CN112134304B - Micro-grid full-automatic navigation method, system and device based on deep learning - Google Patents

Micro-grid full-automatic navigation method, system and device based on deep learning Download PDF

Info

Publication number
CN112134304B
CN112134304B CN202011000664.XA CN202011000664A CN112134304B CN 112134304 B CN112134304 B CN 112134304B CN 202011000664 A CN202011000664 A CN 202011000664A CN 112134304 B CN112134304 B CN 112134304B
Authority
CN
China
Prior art keywords
micro
grid
day
output
energy storage
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202011000664.XA
Other languages
Chinese (zh)
Other versions
CN112134304A (en
Inventor
黄彦璐
俞靖一
张子昊
马溪原
陈元峰
郭晓斌
林冬
向思阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southern Power Grid Digital Grid Research Institute Co Ltd
Original Assignee
Southern Power Grid Digital Grid Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southern Power Grid Digital Grid Research Institute Co Ltd filed Critical Southern Power Grid Digital Grid Research Institute Co Ltd
Priority to CN202011000664.XA priority Critical patent/CN112134304B/en
Publication of CN112134304A publication Critical patent/CN112134304A/en
Application granted granted Critical
Publication of CN112134304B publication Critical patent/CN112134304B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28—Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/22—The renewable source being solar energy
    • H02J2300/24—The renewable source being solar energy of photovoltaic origin
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/28—The renewable source being wind energy
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/50—Photovoltaic [PV] energy
    • Y02E10/56—Power conversion systems, e.g. maximum power point trackers

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention relates to a micro-grid full-automatic navigation method, system and device based on deep learning. The method comprises the following steps: acquiring system net load of a microgrid system at 24 days ago; inputting the system net load into a day-ahead optimization scheduling model of the micro-grid, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the day-ahead optimization scheduling model of the microgrid is a double-layer Bi-LSTM neural network model; adjusting the output of the controllable unit of the day-ahead optimized scheduling strategy of the micro-grid system according to the minimum technical output, the output upper limit value, the climbing constraint and the operation time constraint of the controllable unit; adjusting the energy storage charging and discharging power according to the energy storage charging and discharging power upper limit value, the capacity constraint and the energy storage balance constraint in the scheduling period; and adjusting the exchange power of the micro-grid and the large grid tie line according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint. The method and the device can improve the accuracy and efficiency of the day-ahead optimized scheduling of the microgrid.

Description

Micro-grid full-automatic navigation method, system and device based on deep learning
Technical Field
The invention relates to the field of micro-grid optimized scheduling, in particular to a micro-grid full-automatic navigation method, system and device based on deep learning.
Background
The micro-grid is a small power generation-distribution-utilization system consisting of a controllable distributed power supply unit such as a micro gas turbine and a diesel generator, a fan, a photovoltaic renewable energy power generation unit and energy storage and load, can effectively reduce the operation cost and pollution emission of the system and improve the permeability and power supply reliability of renewable energy sources through the coordinated operation of the internal source, the grid, the load and the energy storage of the system, and is an important component of an intelligent power grid.
Compared with the traditional planning type scheduling method based on engineering personnel experience and cognition and the micro-grid optimization scheduling method based on physical model driving, the micro-grid full-automatic navigation method based on the artificial intelligence technology has the advantages of wide control range and high solving efficiency, and can effectively solve the problems that the planning type scheduling method cannot cover all scenes and the model driving method is low in calculation efficiency and difficult to meet the requirement of operation real-time performance. In addition, with the increase of the complexity of the micro-grid system, factors such as high-proportion access of strong intermittent renewable energy, high-permeability power electronic equipment, multi-energy coupling, multi-participatory gaming, multi-time scale and the like make the traditional micro-grid optimization scheduling method difficult to continue, and the development of the micro-grid intelligent optimization scheduling method based on the artificial intelligence technology, namely the micro-grid full-automatic navigation method, is an effective solution for solving the problem. The day-ahead optimized scheduling of the microgrid is an important means for ensuring the safe, reliable and economic operation of the microgrid, and is a hotspot problem of related research of the microgrid. The traditional microgrid optimization scheduling is usually based on an optimization theory and method, firstly, modeling is carried out on each element in a microgrid, then, models are simplified and processed, and finally, corresponding solving algorithms are researched to solve the models. The objective function of the model is generally the minimum running cost, and a multi-objective optimization model is established by comprehensively considering economic, environmental and social benefits in relevant research; common model modeling methods include mixed integer programming, dynamic programming, model predictive control, distributed optimization, Lyapunov optimization, and the like; common model solving algorithms include genetic algorithms, particle swarm algorithms, active evolution algorithms, lagrangian relaxation methods, and the like. In recent years, with the high-proportion access of renewable energy sources such as fans, photovoltaic and the like, the operation uncertainty of a micro-grid is remarkably increased, and how to deal with uncertainty factors becomes a difficult problem of optimal scheduling of the micro-grid. A common solution for the problems is to convert the uncertainty problem into a certainty problem for modeling solution, which mainly comprises scene-based random optimization, opportunity constraint optimization, robust optimization and the like, but all the methods have certain limitations, such as large random optimization calculation amount, over conservative robust optimization and the like, and when the uncertainty factor of the operation of the microgrid system is complex, the probability modeling is difficult, so that the optimized scheduling efficiency of the microgrid is low.
Artificial intelligence is an important driving force of a new technological revolution and industrial change, machine learning is the core of artificial intelligence, a computer is specially researched to simulate or realize the learning behavior of human beings so as to obtain new knowledge or skills, and the research on the application of the computer in system scheduling and running is helpful for breaking through the limitation of the traditional solution. Machine learning belongs to a model-free idea, a data driving method is adopted to replace physical modeling, internal mechanisms of physical models such as elements and the like do not need to be considered, the method is insensitive to the physical model of a research object, and system environment information is fully mined through continuous learning of known samples such as historical data, decisions and the like, so that a stable control strategy is directly obtained. At present, the application of machine learning algorithms such as deep learning in system optimization scheduling mainly focuses on load prediction, including improvements in load prediction accuracy, prediction speed, model generalization and the like, and solves the problem of random disturbance caused by access of distributed energy; the existing research mostly aims at the characteristics of the micro-grid power data such as scale, randomness, diversity and the like, and adopts deep learning to accurately extract the characteristics of input data, fully explores the relevance among data and outputs a relatively stable result.
Disclosure of Invention
The invention aims to provide a micro-grid full-automatic navigation method, system and device based on deep learning, so that the solving efficiency is remarkably improved while the solving precision of micro-grid optimal scheduling is ensured.
In order to achieve the purpose, the invention provides the following scheme:
a micro-grid full-automatic navigation method based on deep learning comprises the following steps:
acquiring system net load of a microgrid system at 24 days ago;
inputting the system net load into a micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the running time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to obtain the adjusted output of the controllable unit;
adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
and adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line.
Optionally, the adjusting the controllable unit output in the day-ahead optimization scheduling policy of the microgrid system according to the minimum technical output of the controllable unit, the upper limit of the controllable unit output, the controllable unit climbing constraint and the running time constraint to obtain the adjusted controllable unit output specifically includes:
comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technical output which is 0.5 time of the controllable unit output and the minimum technical output;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output;
when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output, judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit, adjusting the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system to be the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit, judging whether the output of the controllable unit in two adjacent time periods in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit or not;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the climbing constraint of the unit, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the output of the controllable unit in the previous time interval and the maximum upward output;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output;
when the output of two adjacent time intervals of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the units, judging whether the running state of each time interval of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum running time constraint and the minimum shutdown time constraint;
adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be the minimum technical output;
and adjusting the output of the controllable unit corresponding to the time interval which does not meet the minimum shutdown time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be 0.
Optionally, the adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling policy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power specifically includes:
judging whether the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is greater than the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the energy storage charging and discharging power, adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system to the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the energy storage charging and discharging power, judging whether the operation capacity of an energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint;
according to the maximum charge state of the energy storage battery, adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
according to the minimum state of charge of the energy storage battery, adjusting energy storage discharge power corresponding to a time period which does not meet the minimum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
when the operation capacity of the energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint, judging whether the energy storage discharge power of the energy storage battery in the last time interval of the scheduling cycle in the day-ahead optimization scheduling strategy of the micro-grid system meets the energy storage balance constraint in the scheduling cycle;
and if the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle, adjusting the energy storage discharge power of the last period of the energy storage battery scheduling cycle to be the opposite number of the sum of the energy storage discharge powers of all periods except the last period in the energy storage battery scheduling cycle.
Optionally, the adjusting, according to the upper limit of the exchange power of the microgrid and the large grid tie line and the system power balance constraint, the exchange power of the microgrid and the large grid tie line in the day-ahead optimization scheduling policy of the microgrid system to obtain the adjusted exchange power of the microgrid and the large grid tie line specifically includes:
judging whether the exchange power of the microgrid and a large power grid tie line in the day-ahead optimization scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid tie line;
when the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the exchange power of the micro-grid and the large grid connecting line, the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to be the upper limit value of the exchange power of the micro-grid and the large grid connecting line;
when the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is not more than the upper limit value of the exchange power of the micro-grid and the large grid tie line, adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
Optionally, the inputting the system payload into the microgrid day-ahead optimization scheduling model and outputting the day-ahead optimization scheduling policy of the microgrid system, and the method further includes:
carrying out normalization processing on training sample data to obtain normalized training sample data; training the microgrid day-ahead optimization scheduling model by adopting a mean square error as a loss function and a weight updating method based on an Adam optimization algorithm based on the normalized training sample data to obtain a trained microgrid day-ahead optimization scheduling model;
the method comprises the following steps of adjusting the exchange power of the microgrid and the large power grid tie line in a day-ahead optimization scheduling strategy of the microgrid system according to the upper limit value of the exchange power of the microgrid and the large power grid tie line and the system power balance constraint to obtain the adjusted exchange power of the microgrid and the large power grid tie line, and then further comprising the following steps:
and performing off-line training on the day-ahead optimization scheduling model of the microgrid according to the adjusted output of the controllable unit, the adjusted energy storage charge-discharge power, the adjusted exchange power of the microgrid and a large power grid connecting line and the system net load of the microgrid system in 24 days ahead, and correcting the weight parameters in the day-ahead optimization scheduling model of the microgrid.
The invention also provides a micro-grid full-automatic navigation system based on deep learning, which comprises:
the system net load obtaining module is used for obtaining the system net load of the micro-grid system at 24 days ago;
the micro-grid day-ahead optimization scheduling strategy generating module is used for inputting the system net load into a micro-grid day-ahead optimization scheduling model and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
the controllable unit output adjusting module is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system according to the minimum technical output of the controllable unit, the upper limit value of the controllable unit output, the climbing constraint of the controllable unit and the running time constraint to obtain the adjusted controllable unit output;
the energy storage charging and discharging power adjusting module is used for adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
and the micro-grid and large-grid tie line exchange power adjusting module is used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system according to the micro-grid and large-grid tie line exchange power upper limit value and the system power balance constraint to obtain the adjusted micro-grid and large-grid tie line exchange power.
Optionally, the controllable unit output adjustment module specifically includes:
the minimum technology processing and comparing unit is used for comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technology output which is 0.5 time of the controllable unit output and the minimum technology output;
the controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0 when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output; when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output;
the controllable unit processing upper limit judging unit is used for judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the controllable unit output upper limit value or not when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output;
the controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the controllable unit output upper limit value when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is larger than the controllable unit output upper limit value;
the climbing constraint judging unit is used for judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit in two adjacent time intervals or not when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit;
the controllable unit output adjusting unit is used for adjusting the controllable unit output of the next time period in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the controllable unit output of the previous time period and the maximum output when the outputs of the two adjacent time periods of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system do not meet the climbing constraint of the unit; when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output;
the operation time constraint judging unit is used for judging whether the operation state of each time interval of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum operation time constraint and the minimum shutdown time constraint or not when the output of two adjacent time intervals of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the unit climbing constraint and climbing constraint;
the controllable unit output adjusting unit is used for adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output; and adjusting the output of the controllable unit corresponding to the time interval which does not meet the minimum shutdown time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be 0.
Optionally, the energy storage charging and discharging power adjustment module specifically includes:
the energy storage charging and discharging power upper limit value judging unit is used for judging whether the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is greater than the energy storage charging and discharging power upper limit value or not;
the energy storage charging and discharging power adjusting unit is used for adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system to be the upper limit value of the energy storage charging and discharging power when the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is larger than the upper limit value of the energy storage charging and discharging power;
the capacity constraint judging unit of the energy storage battery is used for judging whether the running capacity of the energy storage battery in each time period in the day-ahead optimized scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint or not when the energy storage charge-discharge power in the day-ahead optimized scheduling strategy of the micro-grid system is not larger than the upper limit value of the energy storage charge-discharge power;
the energy storage charging and discharging power adjusting unit is used for adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the microgrid system according to the maximum charge state of the energy storage battery; according to the minimum state of charge of the energy storage battery, adjusting energy storage discharge power corresponding to a time period which does not meet the minimum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
the energy storage balance constraint judging unit in the scheduling cycle is used for judging whether the energy storage discharge power of the last time period of the energy storage battery scheduling cycle in the day-ahead optimized scheduling strategy of the microgrid system meets the energy storage balance constraint in the scheduling cycle or not when the running capacity of the energy storage battery in each time period in the day-ahead optimized scheduling strategy of the microgrid system meets the maximum capacity constraint and the minimum capacity constraint;
the energy storage charging and discharging power adjusting unit is used for adjusting the energy storage discharging power of the last period of the energy storage battery scheduling cycle to be the opposite number of the sum of the energy storage discharging power of all periods except the last period in the energy storage battery scheduling cycle when the energy storage discharging power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle.
Optionally, the microgrid and grid tie line exchange power adjustment module specifically includes:
the upper limit value judging unit is used for judging whether the exchange power of the microgrid and the large power grid tie line in the day-ahead optimized scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid tie line;
the microgrid and large power grid connecting line exchange power adjusting unit is used for adjusting the microgrid and large power grid connecting line exchange power in the day-ahead optimization scheduling strategy of the microgrid system to be the upper limit value of the microgrid and large power grid connecting line exchange power when the microgrid and large power grid connecting line exchange power in the day-ahead optimization scheduling strategy of the microgrid system is larger than the upper limit value of the microgrid and large power grid connecting line exchange power;
the micro-grid and large-grid tie line exchange power adjusting unit is further used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint when the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the micro-grid and large-grid tie line exchange power; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
The invention also provides a micro-grid full-automatic navigation device based on deep learning, which comprises: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring system net load of a microgrid system at 24 days ago;
inputting the system net load into a micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the running time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to obtain the adjusted output of the controllable unit;
adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
and adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the method, the double-layer Bi-LSTM neural network model trained based on the Adam optimization algorithm is adopted to generate the day-ahead optimization scheduling strategy of the microgrid, and the refined mathematical modeling of each element of the microgrid in model driving is avoided. And aiming at the generated day-ahead optimization scheduling strategy of the micro-grid, the output of a controllable unit, the energy storage charging and discharging power and the exchange power of the micro-grid and a large power grid connecting line in the day-ahead optimization scheduling strategy are adjusted according to each operation constraint condition of the micro-grid system, so that the accuracy of optimization scheduling of the micro-grid is improved, the rapid balance of the power of the micro-grid can be realized, and the intelligent degree of the scheduling operation of the micro-grid is further improved. The microgrid full-automatic navigation method can quickly and accurately acquire the day-ahead control strategy of each element of the microgrid aiming at any scene of actual operation of the microgrid, and compared with the traditional microgrid optimization scheduling method, the microgrid full-automatic navigation method can remarkably improve the solving efficiency while ensuring the solving precision.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a deep learning-based microgrid full-automatic navigation method of the invention;
FIG. 2 is a schematic structural diagram of a double-layer Bi-LSTM neural network model in the present invention;
FIG. 3 is a schematic structural diagram of LSTM in the Bi-layer Bi-LSTM neural network model;
fig. 4 is a schematic structural diagram of the deep learning-based micro-grid full-automatic navigation system of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flow chart of a deep learning-based microgrid full-automatic navigation method. As shown in fig. 1, the deep learning-based microgrid full-automatic navigation method of the present invention includes the following steps:
step 100: and acquiring the system net load of the microgrid system 24 days ago.
Step 200: and inputting the system net load into the micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system. The day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power. The day-ahead optimization scheduling model of the microgrid is a double-layer Bi-LSTM neural network model trained on the basis of an Adam optimization algorithm. The Bi-LSTM neural network model is a two-way long-short term memory network model, as shown in fig. 2, in the Bi-layer Bi-LSTM neural network, the output of the first layer Bi-LSTM is used as the input of the second layer Bi-LSTM neural network.
The LSTM (long-short term memory network) is a special recurrent neural network, controls forgetting and refreshing of information by adding a memory unit and introducing a gate control unit such as an input gate, a forgetting gate and an output gate, can effectively solve the problems of gradient dispersion and gradient explosion of the recurrent neural network in the long sequence training process, is suitable for processing data information with a longer time sequence, and has a structure shown in figure 3.
The forward propagation process of the basic structural unit of LSTM is shown in formulas (1) - (6).
gf＝σ(Wf[ht-1,xt]+bf) (1)
gi＝σ(Wi[ht-1,xt]+bi) (2)
go＝σ(Wo[ht-1,xt]+bo) (3)
Figure BDA0002694183950000111
Figure BDA0002694183950000112
Figure BDA0002694183950000113
Wherein x istAn input vector of LSTM; c. CtAnd htInternal state vectors and output state vectors of the LSTM, respectively; gf,gi,goRespectively are control variables of a forgetting gate, an input gate and an output gate; sigma is an activation function, and a Sigmoid function is generally used; wfAnd bfA parameter vector for a forgetting gate; wi,bi,Wc,bcIs a parameter vector of the output gate; woAnd boIs a parameter vector of the output gate;
Figure BDA0002694183950000114
representing the hadamard product.
As can be seen from the basic structural unit of the LSTM, at each time stamp t, the LSTM can only extract the feature information of the current input and past time series, and ignore the feature information of the future time series. The structure provides complete past and future context characteristic information of each time stamp of an input sequence of the output layer, enriches the expression capability of the model and does not increase the requirement on data volume. The load prediction data and the scheduling decision information of the microgrid system are typical sequential data with time sequence, the day-ahead optimization scheduling period of the microgrid is longer, and the LSTM which is skilled in processing the longer sequential data is adopted to learn the input and output mapping relation more appropriately.
In the step, before a day-ahead optimization scheduling strategy of the micro-grid system is generated according to the system net load, the constructed day-ahead optimization scheduling model of the micro-grid needs to be trained. The input of the microgrid day-ahead optimization scheduling model is system net load of 24 days ahead, and the system net load can be obtained from wind turbines, photovoltaic and load prediction data of 24 days ahead; the output of the micro-grid day-ahead optimization scheduling model is a scheduling decision result of 24 day-ahead periods, and specifically comprises the output of each controllable unit in each period, the charge-discharge power of the energy storage battery in each period and the exchange power of the micro-grid and the large grid connecting line. Therefore, input data and output data of the microgrid day-ahead optimization scheduling model which are actually collected form a training sample. Before network training, training sample data needs to be subjected to normalization processing. Specifically, the system net load data and the controllable unit output data are normalized to be between [0 and 1], and the charge and discharge power of the energy storage battery and the exchange power of the micro-grid and the large grid tie line are normalized to be between [ -1 and 1], as shown in formulas (7) to (12).
Pnet,t＝PLoad,t-PWT,t-PPV,t (7)
Pes,t＝Pdis,t-Pcha,t (8)
Figure BDA0002694183950000121
Figure BDA0002694183950000122
Figure BDA0002694183950000123
Figure BDA0002694183950000124
In the formula, PLoad,tRepresenting the load size of the system during the time period t; pWT,tAnd PPV,tRespectively representing the predicted output of the fan and the photovoltaic in the time period t; pGi,tRepresenting the output of the controllable unit i in the time t; pcha,tAnd Pdis,tRespectively representing the charging power and the discharging power of the energy storage battery in a time period t; pGrid,tRepresenting the exchange power of the microgrid with the large grid during a period t, where PGrid,tIf the power is more than 0, the micro-grid buys electricity from the large grid, otherwise, the micro-grid sells electricity to the large grid; pes,tFor charging and discharging power of the energy storage battery in time period t, Pnet,tIs the net load of the system over time period t; pes,tFor charging and discharging power of the energy storage battery in time period t, Pes,tDischarge is indicated by > 0, and charge is indicated by contraries; pnet,t0,PGi,t0,Pes,t0,PGrid,t0And the normalized values of the system net load, the output of the controllable unit, the energy storage charge-discharge power and the exchange power of the micro-grid and the large-grid connecting line are respectively represented.
And then, training the microgrid day-ahead optimization scheduling model by adopting a mean square error as a loss function and a weight updating method based on an Adam optimization algorithm based on the normalized training sample data to obtain the trained microgrid day-ahead optimization scheduling model.
The loss function is:
Figure BDA0002694183950000131
the weight update formula of the Adam algorithm is shown in equations (14) to (16):
Figure BDA0002694183950000132
Figure BDA0002694183950000133
Figure BDA0002694183950000134
in the formula (P)Gi,t0,Pes,t0,PGrid,t0) Is the actual value of the output vector;
Figure BDA0002694183950000135
is a predicted value of the output vector; thetatFor neural nets to be updatedA network weight parameter; delta is the learning rate; epsilon is a smoothing parameter; m ist,vtRespectively a first moment mean value and a second moment mean value of the gradient; beta is a1,β2Is the attenuation factor.
The day-ahead optimization scheduling strategy of the microgrid system output by the day-ahead optimization scheduling model probably does not meet the power balance constraint of the microgrid system, the operation constraint of each element and the like, so that the result output by the microgrid system needs to be correspondingly adjusted, and the specific adjustment process is from step 300 to step 500.
Step 300: and adjusting the controllable unit output in the day-ahead optimization scheduling strategy of the microgrid system according to the minimum technical output of the controllable unit, the upper limit value of the controllable unit output, the climbing constraint of the controllable unit and the running time constraint to obtain the adjusted controllable unit output. The method comprises the following specific steps:
(1) and comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technical output of 0.5 time and the minimum technical output.
And when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output, adjusting the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system to 0 to finish the adjustment of the output of the controllable unit.
And when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to the minimum technical output, and finishing the adjustment of the controllable unit output.
And (3) when the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output, continuing to execute the step (2).
(2) And judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the output upper limit value of the controllable unit.
And if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit, adjusting the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system to the upper limit value of the output of the controllable unit, and finishing the adjustment of the output of the controllable unit.
And (4) if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the output upper limit value of the controllable unit, continuing to execute the step (3).
(3) And judging whether the output of the controllable unit in two adjacent time periods in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit.
When the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit climbing constraint, the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system is adjusted to be the sum of the output of the controllable unit in the previous time interval and the maximum upward output, namely the output of the controllable unit in the next time interval is the sum of the output of the controllable unit in the previous time interval and the maximum upward output, and the adjustment of the output of the controllable unit is completed.
When the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system is adjusted to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulating output, namely the output of the controllable unit in the next time interval is equal to the output of the controllable unit in the previous time interval-the minimum up-regulating output, and the adjustment of the output of the controllable unit is completed.
And (4) when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit, continuing to execute the step (4).
(4) And judging whether the operation state of each time interval of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system meets the minimum operation time constraint and the minimum shutdown time constraint.
And adjusting the controllable unit output corresponding to the time interval which does not meet the minimum operation time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be the minimum technical output, and finishing the adjustment of the controllable unit output.
And adjusting the output of the controllable unit to be 0 corresponding to the time interval which does not meet the minimum shutdown time constraint in the day-ahead optimized scheduling strategy of the micro-grid system, and finishing the adjustment of the output of the controllable unit.
And if the operation state of the controllable unit in each time period in the day-ahead optimization scheduling strategy of the micro-grid system meets the minimum operation time constraint and the minimum shutdown time constraint, the output of the controllable unit is not required to be adjusted.
Step 400: and adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power. The specific process is as follows:
(1) and judging whether the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is greater than the upper limit value of the energy storage charging and discharging power.
And when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system is greater than the upper limit value of the energy storage charging and discharging power, adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system to the upper limit value of the energy storage charging and discharging power, and completing the adjustment of the energy storage charging and discharging power.
And (3) when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system is not greater than the upper limit value of the energy storage charging and discharging power, continuing to execute the step (2).
(2) And judging whether the running capacity of the energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint.
And adjusting the energy storage charging power corresponding to the time period which does not meet the maximum capacity constraint in the day-ahead optimization scheduling strategy of the microgrid system according to the maximum charge state of the energy storage battery, so as to finish the adjustment of the energy storage charging and discharging power.
And adjusting the energy storage discharge power corresponding to the time period which does not meet the minimum capacity constraint in the day-ahead optimization scheduling strategy of the microgrid system according to the minimum charge state of the energy storage battery, so as to finish the adjustment of the energy storage charge and discharge power.
And (4) when the running capacity of the energy storage battery in each period in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint, continuing to execute the step (3).
(3) And judging whether the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the micro-grid system meets the energy storage balance constraint in the scheduling cycle.
If the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle, the energy storage discharge power of the last period of the energy storage battery scheduling cycle is adjusted to be the opposite number of the sum of the energy storage discharge powers of all periods except the last period in the energy storage battery scheduling cycle so as to meet the energy storage balance constraint in the scheduling cycle, namely the sum of the energy storage discharge power of the last period and the energy storage discharge power of all other periods in the scheduling cycle is 0. And finishing the adjustment of the energy storage charging and discharging power.
And if the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the micro-grid system meets the energy storage balance constraint in the scheduling cycle, the energy storage discharge power does not need to be adjusted.
Step 500: and adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line. The specific process is as follows:
(1) and judging whether the exchange power of the microgrid and a large power grid connecting line in the day-ahead optimization scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid connecting line.
When the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is greater than the upper limit value of the exchange power of the micro-grid and the large grid tie line, the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to the upper limit value of the exchange power of the micro-grid and the large grid tie line, and the adjustment of the exchange power of the micro-grid and the large grid tie line is completed;
and (3) when the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the exchange power of the micro-grid and the large grid tie line, continuing to execute the step (2).
(2) Adjusting the exchange power adjustment of the microgrid and a large power grid connecting line in the day-ahead optimization scheduling strategy of the microgrid system according to the adjusted output of the controllable unit, the adjusted energy storage charge-discharge power and the system power balance constraint; and (4) adjusting the exchange power of the micro-grid and the large grid tie line, namely system net load, adjusted output of the controllable unit, and adjusted energy storage charging and discharging power.
When the day-ahead optimization scheduling strategy of the microgrid system output by the day-ahead optimization scheduling model is adjusted, the element constraints are processed first, and the system constraints are processed later, namely (1) - (4) in the step 300, (1) - (2) in the step 400 and (1) in the step 500 are processed first, then (3) in the step 400 is processed, and finally (2) in the step 500 is processed again.
Based on the steps, corresponding scheduling decision results can be directly output according to day-ahead prediction data of any scene of wind power, photovoltaic and load, and rapid balance and auxiliary decision of day-ahead power of the microgrid are achieved. In order to further improve the accuracy of the microgrid optimization scheduling, input data of a microgrid day-ahead optimization scheduling model (namely, system net load of the microgrid system in 24 day-ahead periods) and a corresponding adjusted day-ahead optimization scheduling strategy (namely, adjusted controllable unit output, adjusted energy storage charging and discharging power, adjusted microgrid and large power grid connection line exchange power) are stored, continuous offline training is performed on the microgrid day-ahead optimization scheduling model along with the passage of time and continuous accumulation of historical sample data, and weight parameters are corrected, so that the precision and the efficiency of the microgrid day-ahead optimization scheduling can be continuously improved.
According to the method, the day-ahead optimization scheduling decision of the microgrid needs to meet time coupling constraints such as controllable unit climbing constraint, minimum operation and outage time constraint, energy storage battery capacity and charging and discharging power relation constraint and the like, so that the scheduling decision result of the microgrid at the current time period is influenced by the fan output, photovoltaic output and the operation working conditions of loads at the past and future time periods, therefore, the Bi-LSTM neural network is selected to construct the deep learning model of the day-ahead optimization scheduling of the microgrid, the characteristic information of the operation working conditions of the microgrid can be more effectively extracted, the mapping relation between the operation scene (system net load) of the microgrid and the scheduling decision result can be more accurately described, the optimization scheduling result is more accurate, and the efficiency is higher.
Corresponding to the micro-grid full-automatic navigation method based on deep learning, the invention also provides a micro-grid full-automatic navigation system based on deep learning. Fig. 4 is a schematic structural diagram of the deep learning-based micro-grid full-automatic navigation system of the present invention. As shown in fig. 4, the deep learning-based micro-grid full-automatic navigation system of the present invention includes the following structures:
the system payload obtaining module 401 is configured to obtain a system payload of the microgrid system at a 24-day previous time period.
A microgrid day-ahead optimization scheduling strategy generation module 402, configured to input the system payload into a microgrid day-ahead optimization scheduling model, and output a day-ahead optimization scheduling strategy of the microgrid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charging and discharging power and micro-grid and large grid tie line exchange power.
And a controllable unit output adjusting module 403, configured to adjust the controllable unit output in the day-ahead optimization scheduling policy of the microgrid system according to the minimum technical output of the controllable unit, the upper limit value of the controllable unit output, the controllable unit climbing constraint and the running time constraint, so as to obtain the adjusted controllable unit output.
And the energy storage charging and discharging power adjusting module 404 is configured to adjust the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery, and the energy storage balance constraint in the scheduling period, so as to obtain the adjusted energy storage charging and discharging power.
The microgrid-to-large grid tie line exchange power adjustment module 405 is configured to adjust the microgrid-to-large grid tie line exchange power in the day-ahead optimization scheduling strategy of the microgrid system according to a microgrid-to-large grid tie line exchange power upper limit value and system power balance constraint, so as to obtain the adjusted microgrid-to-large grid tie line exchange power.
As another embodiment, in the deep learning-based fully automatic navigation system for a microgrid of the present invention, the controllable unit output adjustment module 403 specifically includes:
and the minimum technology processing and comparing unit is used for comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technology output which is 0.5 time of the controllable unit output and the minimum technology output.
The controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0 when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output; and when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output.
And the controllable unit processing upper limit judging unit is used for judging whether the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the controllable unit output upper limit value or not when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output.
The controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the controllable unit output upper limit value when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is larger than the controllable unit output upper limit value.
And the climbing constraint judging unit is used for judging whether the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system meets the unit climbing constraint and descending climbing constraint when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit.
The controllable unit output adjusting unit is further used for adjusting the controllable unit output of a later time period in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the controllable unit output of the previous time period and the maximum upper regulation output when the outputs of two adjacent time periods of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system do not meet the climbing constraint of the unit; and when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output.
And the operation time constraint judging unit is used for judging whether the operation state of each time interval of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum operation time constraint and the minimum shutdown time constraint or not when the output of two adjacent time intervals of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the unit climbing constraint and climbing constraint.
The controllable unit output adjusting unit is used for adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output; and adjusting the output of the controllable unit corresponding to the time interval which does not meet the minimum shutdown time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be 0.
As another embodiment, in the fully automatic navigation system of a microgrid based on deep learning of the present invention, the energy storage charging/discharging power adjustment module 404 specifically includes:
and the energy storage charging and discharging power upper limit value judging unit is used for judging whether the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is greater than the energy storage charging and discharging power upper limit value.
And the energy storage charging and discharging power adjusting unit is used for adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system to be the energy storage charging and discharging power upper limit value when the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is larger than the energy storage charging and discharging power upper limit value.
And the capacity constraint judging unit of the energy storage battery is used for judging whether the running capacity of the energy storage battery in each time period in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint or not when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the energy storage charging and discharging power.
The energy storage charging and discharging power adjusting unit is further used for adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the microgrid system according to the maximum charge state of the energy storage battery; and adjusting the energy storage discharge power corresponding to the time period which does not meet the minimum capacity constraint in the day-ahead optimization scheduling strategy of the micro-grid system according to the minimum charge state of the energy storage battery.
And the energy storage balance constraint judging unit in the scheduling cycle is used for judging whether the energy storage discharge power of the last time period of the energy storage battery scheduling cycle in the day-ahead optimized scheduling strategy of the microgrid system meets the energy storage balance constraint in the scheduling cycle or not when the running capacity of the energy storage battery in each time period in the day-ahead optimized scheduling strategy of the microgrid system meets the maximum capacity constraint and the minimum capacity constraint.
The energy storage charging and discharging power adjusting unit is further configured to adjust the energy storage discharging power at the last period of the energy storage battery scheduling cycle to be an inverse number of the sum of the energy storage discharging power at all periods except the last period in the energy storage battery scheduling cycle when the energy storage discharging power at the last period of the energy storage battery scheduling cycle in the day-ahead optimized scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle.
As another embodiment, in the fully automatic navigation system of a microgrid based on deep learning of the present invention, the microgrid and grid tie line exchange power adjustment module 405 specifically includes:
and the upper limit value judgment unit of the exchange power of the microgrid and the large power grid connecting line is used for judging whether the exchange power of the microgrid and the large power grid connecting line in the day-ahead optimized scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid connecting line.
And the micro-grid and large-grid tie line exchange power adjusting unit is used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system to be the micro-grid and large-grid tie line exchange power upper limit value when the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the micro-grid and large-grid tie line exchange power upper limit value.
The micro-grid and large-grid tie line exchange power adjusting unit is further used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint when the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the micro-grid and large-grid tie line exchange power; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
The invention also provides a micro-grid full-automatic navigation device based on deep learning, which comprises: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring system net load of a microgrid system at 24 days ago;
inputting the system net load into a micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the running time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to obtain the adjusted output of the controllable unit;
adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
and adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (4)

1. A micro-grid full-automatic navigation method based on deep learning is characterized by comprising the following steps:
acquiring system net load of a microgrid system at 24 days ago;
inputting the system net load into a micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the running time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to obtain the adjusted output of the controllable unit;
adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the operation time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the microgrid system is adjusted to obtain the adjusted output of the controllable unit, and the method specifically comprises the following steps:
comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technical output which is 0.5 time of the controllable unit output and the minimum technical output;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output;
when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output, judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit, adjusting the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system to be the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit, judging whether the output of the controllable unit in two adjacent time periods in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit or not;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the climbing constraint of the unit, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the output of the controllable unit in the previous time interval and the maximum upward output;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output;
when the output of two adjacent time intervals of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the units, judging whether the running state of each time interval of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum running time constraint and the minimum shutdown time constraint;
adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be the minimum technical output;
adjusting the output of a controllable unit corresponding to a time period which does not meet the minimum shutdown time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be 0;
the method for adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power specifically comprises the following steps:
judging whether the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is greater than the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the energy storage charging and discharging power, adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system to the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the energy storage charging and discharging power, judging whether the operation capacity of an energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint;
according to the maximum charge state of the energy storage battery, adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
according to the minimum state of charge of the energy storage battery, adjusting energy storage discharge power corresponding to a time period which does not meet the minimum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
when the operation capacity of the energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint, judging whether the energy storage discharge power of the energy storage battery in the last time interval of the scheduling cycle in the day-ahead optimization scheduling strategy of the micro-grid system meets the energy storage balance constraint in the scheduling cycle;
if the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle, adjusting the energy storage discharge power of the last period of the energy storage battery scheduling cycle to be the opposite number of the sum of the energy storage discharge powers of all periods except the last period in the energy storage battery scheduling cycle;
the method comprises the following steps of adjusting the exchange power of the microgrid and the large power grid tie line in a day-ahead optimization scheduling strategy of the microgrid system according to the upper limit value of the exchange power of the microgrid and the large power grid tie line and the system power balance constraint to obtain the adjusted exchange power of the microgrid and the large power grid tie line, and specifically comprises the following steps:
judging whether the exchange power of the microgrid and a large power grid tie line in the day-ahead optimization scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid tie line;
when the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the exchange power of the micro-grid and the large grid connecting line, the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to be the upper limit value of the exchange power of the micro-grid and the large grid connecting line;
when the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is not more than the upper limit value of the exchange power of the micro-grid and the large grid tie line, adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
2. The deep learning-based microgrid full-automatic navigation method according to claim 1, characterized in that the system payload is input into a microgrid day-ahead optimization scheduling model and a day-ahead optimization scheduling strategy of the microgrid system is output, and the method further comprises the following steps:
carrying out normalization processing on training sample data to obtain normalized training sample data; training the microgrid day-ahead optimization scheduling model by adopting a mean square error as a loss function and a weight updating method based on an Adam optimization algorithm based on the normalized training sample data to obtain a trained microgrid day-ahead optimization scheduling model;
the method comprises the following steps of adjusting the exchange power of the microgrid and the large power grid tie line in a day-ahead optimization scheduling strategy of the microgrid system according to the upper limit value of the exchange power of the microgrid and the large power grid tie line and the system power balance constraint to obtain the adjusted exchange power of the microgrid and the large power grid tie line, and then further comprising the following steps:
and performing off-line training on the day-ahead optimization scheduling model of the microgrid according to the adjusted output of the controllable unit, the adjusted energy storage charge-discharge power, the adjusted exchange power of the microgrid and a large power grid connecting line and the system net load of the microgrid system in 24 days ahead, and correcting the weight parameters in the day-ahead optimization scheduling model of the microgrid.
3. The utility model provides a full-automatic navigation system of little electric wire netting based on deep learning which characterized in that includes:
the system net load obtaining module is used for obtaining the system net load of the micro-grid system at 24 days ago;
the micro-grid day-ahead optimization scheduling strategy generating module is used for inputting the system net load into a micro-grid day-ahead optimization scheduling model and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
the controllable unit output adjusting module is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system according to the minimum technical output of the controllable unit, the upper limit value of the controllable unit output, the climbing constraint of the controllable unit and the running time constraint to obtain the adjusted controllable unit output;
the energy storage charging and discharging power adjusting module is used for adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
the micro-grid and large-grid tie line exchange power adjusting module is used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system according to the micro-grid and large-grid tie line exchange power upper limit value and the system power balance constraint to obtain the adjusted micro-grid and large-grid tie line exchange power;
the controllable unit output adjusting module specifically comprises:
the minimum technology processing and comparing unit is used for comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technology output which is 0.5 time of the controllable unit output and the minimum technology output;
the controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0 when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output; when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output;
the controllable unit processing upper limit judging unit is used for judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the controllable unit output upper limit value or not when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output;
the controllable unit output adjusting unit is used for adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the controllable unit output upper limit value when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is larger than the controllable unit output upper limit value;
the climbing constraint judging unit is used for judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit in two adjacent time intervals or not when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit;
the controllable unit output adjusting unit is further used for adjusting the controllable unit output of a later time period in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the controllable unit output of the previous time period and the maximum upper regulation output when the outputs of two adjacent time periods of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system do not meet the climbing constraint of the unit; when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output;
the operation time constraint judging unit is used for judging whether the operation state of each time interval of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum operation time constraint and the minimum shutdown time constraint or not when the output of two adjacent time intervals of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system meets the unit climbing constraint and climbing constraint;
the controllable unit output adjusting unit is used for adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output; adjusting the output of a controllable unit corresponding to a time period which does not meet the minimum shutdown time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be 0;
the energy storage charging and discharging power adjusting module specifically comprises:
the energy storage charging and discharging power upper limit value judging unit is used for judging whether the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is greater than the energy storage charging and discharging power upper limit value or not;
the energy storage charging and discharging power adjusting unit is used for adjusting the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system to be the upper limit value of the energy storage charging and discharging power when the energy storage charging and discharging power in the day-ahead optimized scheduling strategy of the microgrid system is larger than the upper limit value of the energy storage charging and discharging power;
the capacity constraint judging unit of the energy storage battery is used for judging whether the running capacity of the energy storage battery in each time period in the day-ahead optimized scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint or not when the energy storage charge-discharge power in the day-ahead optimized scheduling strategy of the micro-grid system is not larger than the upper limit value of the energy storage charge-discharge power;
the energy storage charging and discharging power adjusting unit is further used for adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the microgrid system according to the maximum charge state of the energy storage battery; according to the minimum state of charge of the energy storage battery, adjusting energy storage discharge power corresponding to a time period which does not meet the minimum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
the energy storage balance constraint judging unit in the scheduling cycle is used for judging whether the energy storage discharge power of the last time period of the energy storage battery scheduling cycle in the day-ahead optimized scheduling strategy of the microgrid system meets the energy storage balance constraint in the scheduling cycle or not when the running capacity of the energy storage battery in each time period in the day-ahead optimized scheduling strategy of the microgrid system meets the maximum capacity constraint and the minimum capacity constraint;
the energy storage charging and discharging power adjusting unit is used for adjusting the energy storage discharging power of the last period of the energy storage battery scheduling cycle to be the opposite number of the sum of the energy storage discharging power of all periods except the last period in the energy storage battery scheduling cycle when the energy storage discharging power of the last period of the energy storage battery scheduling cycle in the day-ahead optimized scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle;
the micro-grid and large-grid tie line exchange power adjustment module specifically comprises:
the upper limit value judging unit is used for judging whether the exchange power of the microgrid and the large power grid tie line in the day-ahead optimized scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid tie line;
the microgrid and large power grid connecting line exchange power adjusting unit is used for adjusting the microgrid and large power grid connecting line exchange power in the day-ahead optimization scheduling strategy of the microgrid system to be the upper limit value of the microgrid and large power grid connecting line exchange power when the microgrid and large power grid connecting line exchange power in the day-ahead optimization scheduling strategy of the microgrid system is larger than the upper limit value of the microgrid and large power grid connecting line exchange power;
the micro-grid and large-grid tie line exchange power adjusting unit is further used for adjusting the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint when the micro-grid and large-grid tie line exchange power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the micro-grid and large-grid tie line exchange power; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
4. The utility model provides a full-automatic navigation head of little electric wire netting based on deep learning which characterized in that includes: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring system net load of a microgrid system at 24 days ago;
inputting the system net load into a micro-grid day-ahead optimization scheduling model, and outputting a day-ahead optimization scheduling strategy of the micro-grid system; the microgrid day-ahead optimization scheduling model is a double-layer Bi-LSTM neural network model trained based on an Adam optimization algorithm; the day-ahead optimization scheduling strategy of the micro-grid system comprises controllable unit output, energy storage charge-discharge power and micro-grid and large grid tie line exchange power;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the running time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to obtain the adjusted output of the controllable unit;
adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power;
adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the upper limit value of the exchange power of the micro-grid and the large grid tie line and the system power balance constraint to obtain the adjusted exchange power of the micro-grid and the large grid tie line;
according to the minimum technical output of the controllable unit, the upper limit value of the output of the controllable unit, the climbing constraint of the controllable unit and the operation time constraint, the output of the controllable unit in the day-ahead optimization scheduling strategy of the microgrid system is adjusted to obtain the adjusted output of the controllable unit, and the method specifically comprises the following steps:
comparing the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system with the minimum technical output which is 0.5 time of the controllable unit output and the minimum technical output;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is less than or equal to 0.5 times of the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to 0;
when the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system is greater than 0.5 times of the minimum technical output and less than the minimum technical output, adjusting the controllable unit output in the day-ahead optimized scheduling strategy of the micro-grid system to be the minimum technical output;
when the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than or equal to the minimum technical output, judging whether the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is greater than the upper limit value of the output of the controllable unit, adjusting the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system to be the upper limit value of the output of the controllable unit;
if the output of the controllable unit in the day-ahead optimized scheduling strategy of the micro-grid system is not greater than the upper limit value of the output of the controllable unit, judging whether the output of the controllable unit in two adjacent time periods in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the unit or not;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the climbing constraint of the unit, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the sum of the output of the controllable unit in the previous time interval and the maximum upward output;
when the output of the controllable unit in two adjacent time intervals in the day-ahead optimized scheduling strategy of the micro-grid system does not meet the unit down-climbing constraint, adjusting the output of the controllable unit in the next time interval in the day-ahead optimized scheduling strategy of the micro-grid system to be the difference between the output of the controllable unit in the previous time interval and the maximum down-regulation output;
when the output of two adjacent time intervals of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the climbing constraint and the climbing constraint of the units, judging whether the running state of each time interval of the controllable units in the day-ahead optimized scheduling strategy of the micro-grid system meets the minimum running time constraint and the minimum shutdown time constraint;
adjusting the controllable unit output corresponding to the time period which does not meet the minimum operation time constraint in the day-ahead optimization scheduling strategy of the micro-grid system to be the minimum technical output;
adjusting the output of a controllable unit corresponding to a time period which does not meet the minimum shutdown time constraint in the day-ahead optimized scheduling strategy of the micro-grid system to be 0;
the method for adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the microgrid system according to the energy storage charging and discharging power upper limit value, the capacity constraint of the energy storage battery and the energy storage balance constraint in the scheduling period to obtain the adjusted energy storage charging and discharging power specifically comprises the following steps:
judging whether the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is greater than the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the energy storage charging and discharging power, adjusting the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system to the upper limit value of the energy storage charging and discharging power;
when the energy storage charging and discharging power in the day-ahead optimization scheduling strategy of the micro-grid system is not greater than the upper limit value of the energy storage charging and discharging power, judging whether the operation capacity of an energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint;
according to the maximum charge state of the energy storage battery, adjusting energy storage charging power corresponding to a time period which does not meet the maximum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
according to the minimum state of charge of the energy storage battery, adjusting energy storage discharge power corresponding to a time period which does not meet the minimum capacity constraint in a day-ahead optimization scheduling strategy of the micro-grid system;
when the operation capacity of the energy storage battery in each time interval in the day-ahead optimization scheduling strategy of the micro-grid system meets the maximum capacity constraint and the minimum capacity constraint, judging whether the energy storage discharge power of the energy storage battery in the last time interval of the scheduling cycle in the day-ahead optimization scheduling strategy of the micro-grid system meets the energy storage balance constraint in the scheduling cycle;
if the energy storage discharge power of the last period of the energy storage battery scheduling cycle in the day-ahead optimization scheduling strategy of the microgrid system does not meet the energy storage balance constraint in the scheduling cycle, adjusting the energy storage discharge power of the last period of the energy storage battery scheduling cycle to be the opposite number of the sum of the energy storage discharge powers of all periods except the last period in the energy storage battery scheduling cycle;
the method comprises the following steps of adjusting the exchange power of the microgrid and the large power grid tie line in a day-ahead optimization scheduling strategy of the microgrid system according to the upper limit value of the exchange power of the microgrid and the large power grid tie line and the system power balance constraint to obtain the adjusted exchange power of the microgrid and the large power grid tie line, and specifically comprises the following steps:
judging whether the exchange power of the microgrid and a large power grid tie line in the day-ahead optimization scheduling strategy of the microgrid system is greater than the upper limit value of the exchange power of the microgrid and the large power grid tie line;
when the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is larger than the upper limit value of the exchange power of the micro-grid and the large grid connecting line, the exchange power of the micro-grid and the large grid connecting line in the day-ahead optimization scheduling strategy of the micro-grid system is adjusted to be the upper limit value of the exchange power of the micro-grid and the large grid connecting line;
when the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system is not more than the upper limit value of the exchange power of the micro-grid and the large grid tie line, adjusting the exchange power of the micro-grid and the large grid tie line in the day-ahead optimization scheduling strategy of the micro-grid system according to the adjusted controllable unit output, the adjusted energy storage charge-discharge power and the system power balance constraint; and (3) the exchange power of the adjusted micro-grid and the large grid tie line is equal to the system net load, the adjusted output of the controllable unit and the adjusted energy storage charging and discharging power.
CN202011000664.XA 2020-09-22 2020-09-22 Micro-grid full-automatic navigation method, system and device based on deep learning Active CN112134304B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011000664.XA CN112134304B (en) 2020-09-22 2020-09-22 Micro-grid full-automatic navigation method, system and device based on deep learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011000664.XA CN112134304B (en) 2020-09-22 2020-09-22 Micro-grid full-automatic navigation method, system and device based on deep learning

Publications (2)

Publication Number Publication Date
CN112134304A CN112134304A (en) 2020-12-25
CN112134304B true CN112134304B (en) 2022-04-01

Family

ID=73842174

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011000664.XA Active CN112134304B (en) 2020-09-22 2020-09-22 Micro-grid full-automatic navigation method, system and device based on deep learning

Country Status (1)

Country Link
CN (1) CN112134304B (en)

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110738344B (en) * 2018-07-20 2022-07-05 中国农业大学 Distributed reactive power optimization method and device for load prediction of power system
CN109462231B (en) * 2018-11-15 2020-09-01 合肥工业大学 Load optimization scheduling method, system and storage medium for residential micro-grid
CN109948845A (en) * 2019-03-15 2019-06-28 国网江苏省电力有限公司经济技术研究院 A kind of distribution network load shot and long term Memory Neural Networks prediction technique
CN110135637A (en) * 2019-05-13 2019-08-16 武汉科技大学 Micro-capacitance sensor short-term load forecasting method based on shot and long term memory and adaptive boosting
CN110071530B (en) * 2019-05-20 2020-10-02 中国矿业大学 Wind power climbing coordination scheduling method of energy storage-containing power system based on LSTM
CN110334879A (en) * 2019-07-11 2019-10-15 华北电力大学 Power grid bus reactive load forecasting method and device
CN111666763A (en) * 2020-05-28 2020-09-15 平安科技（深圳）有限公司 Network structure construction method and device for multitask scene
CN111626527B (en) * 2020-06-10 2023-02-03 太原理工大学 Intelligent power grid deep learning scheduling method considering fast/slow charging/discharging form of schedulable electric vehicle

Also Published As

Publication number Publication date
CN112134304A (en) 2020-12-25

Similar Documents

Publication Publication Date Title
Yousri et al. Recent methodology based Harris Hawks optimizer for designing load frequency control incorporated in multi-interconnected renewable energy plants
Bose Artificial intelligence techniques in smart grid and renewable energy systems—some example applications
Yang et al. Disturbance observer based fractional-order integral sliding mode frequency control strategy for interconnected power system
CN105846461B (en) Control method and system for large-scale energy storage power station self-adaptive dynamic planning
Sivalingam et al. A modified whale optimization algorithm-based adaptive fuzzy logic PID controller for load frequency control of autonomous power generation systems
CN110070292B (en) Micro-grid economic dispatching method based on cross variation whale optimization algorithm
Capizzi et al. Recurrent neural network-based control strategy for battery energy storage in generation systems with intermittent renewable energy sources
Bakeer et al. A sophisticated modeling approach for photovoltaic systems in load frequency control
Yin et al. Emotional deep learning programming controller for automatic voltage control of power systems
Rosas et al. Itineraries for charging and discharging a BESS using energy predictions based on a CNN-LSTM neural network model in BCS, Mexico
Zhang et al. Deep reinforcement learning for load shedding against short-term voltage instability in large power systems
Chatziagorakis et al. Enhancement of hybrid renewable energy systems control with neural networks applied to weather forecasting: the case of Olvio
Hu et al. Deep reinforcement learning based coordinated voltage control in smart distribution network
CN113872213B (en) Autonomous optimization control method and device for power distribution network voltage
CN117039981A (en) Large-scale power grid optimal scheduling method, device and storage medium for new energy
Yin et al. Expandable deep width learning for voltage control of three-state energy model based smart grids containing flexible energy sources
CN114430165A (en) Micro-grid group intelligent coordination control method and device based on depth model prediction
Almasoudi et al. Nonlinear coordination strategy between renewable energy sources and fuel cells for frequency regulation of hybrid power systems
Elymany et al. Hybrid optimized-ANFIS based MPPT for hybrid microgrid using zebra optimization algorithm and artificial gorilla troops optimizer
El Bourakadi et al. Multi-agent system based on the fuzzy control and extreme learning machine for intelligent management in hybrid energy system
CN112134304B (en) Micro-grid full-automatic navigation method, system and device based on deep learning
CN116054152A (en) Method for participating in collaborative optimization control of source network and load storage of power distribution network by considering economic benefit of wind-solar energy storage
Shi et al. Distribution network distributed energy storage configuration optimization method considering variance of network loss sensitivity
Parabhane et al. Artificial neural networks based power management scheme with enhanced stability for a solar panel/wind turbine generator/fuel cell/battery/power supply designed for industrial loads
Srisailam et al. Load frequency control in deregulated power system with renewable energy sources: Hybrid GOA‐SNN technique

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant