CN111799847A - Predictive control method of risk-considering two-stage random model of active power distribution network - Google Patents

Predictive control method of risk-considering two-stage random model of active power distribution network Download PDF

Info

Publication number
CN111799847A
CN111799847A CN202010688381.2A CN202010688381A CN111799847A CN 111799847 A CN111799847 A CN 111799847A CN 202010688381 A CN202010688381 A CN 202010688381A CN 111799847 A CN111799847 A CN 111799847A
Authority
CN
China
Prior art keywords
distribution network
power distribution
oltc
active
active power
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.)
Granted
Application number
CN202010688381.2A
Other languages
Chinese (zh)
Other versions
CN111799847B (en
Inventor
周文涛
张宁
曹振博
齐祥和
陈懿
孟凡晨
王泽黎
杜孟珂
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.)
State Grid Corp of China SGCC
State Grid Beijing Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
State Grid Beijing Electric Power 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 State Grid Corp of China SGCC, State Grid Beijing Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202010688381.2A priority Critical patent/CN111799847B/en
Publication of CN111799847A publication Critical patent/CN111799847A/en
Application granted granted Critical
Publication of CN111799847B publication Critical patent/CN111799847B/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
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • 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
    • H02J3/38—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
    • H02J3/48—Controlling the sharing of the in-phase component
    • 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
    • H02J3/50—Controlling the sharing of the out-of-phase component
    • 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
    • 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)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Power Engineering (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Public Health (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses a risk-related prediction control method for a two-stage stochastic model of an active power distribution network, which comprises the following steps of: 1) photovoltaic output and load fluctuation are respectively designed to obey Beta distribution and normal distribution, a large-scale uncertain scene set is generated, and then the large-scale uncertain scene set is reduced; 2) establishing an active and reactive power coordinated scheduling model of the active power distribution network by adopting a two-stage random optimization method; 3) an active and reactive power coordinated scheduling model of the active power distribution network is perfected; 4)forming an optimal scheduling strategy; 5) at each sampling initial time tkRolling solving an optimal scheduling strategy within a prediction period, wherein only [ t ] is performedk,tk+Tc]Control strategy of each regulated resource in time period and at tk,tk+4Tc]The method can flexibly adjust and quickly respond to an operation control object, and can effectively deal with the influence of the error between a predicted value and an actual value on an adjustment strategy so as to improve the voltage operation level of the system.

Description

Predictive control method of risk-considering two-stage random model of active power distribution network
Technical Field
The invention belongs to the technical field of power system automation, and relates to a risk-related predictive control method for a two-stage stochastic model of an active power distribution network.
Background
Active elements such as Distributed Generation (DG), electric vehicles and Energy Storage Systems (ESS) are widely connected, so that power flow of a power distribution network is changed from traditional one-way transmission to two-way flow, uncertainty is enhanced, power and voltage fluctuation of the active power distribution network are greatly improved, and a flexible and effective operation control means is more important. Traditional mechanical equipment such as an on-load tap changer (OLTC) and a Capacitor Bank (CB) has low regulation speed and limited action times. Devices such as DGs and ESS are generally connected with a power distribution network through novel power electronic devices, can quickly track power fluctuation and make timely adjustment, have the characteristics of flexible adjustment and quick response, and are a good operation control means. In the future, the boundary between Active Distribution Network (ADN) active scheduling and reactive power optimization becomes more fuzzy, and it is a necessary trend to comprehensively control various distributed energy sources and various active management means and realize comprehensive and efficient utilization of the energy sources in the whole network.
With the continuous increase of random operation conditions such as new energy power generation and power load change, an operation scheduling theory considering uncertainty factors becomes a current hot problem, and the research of an operation theory considering risks is not limited. An opportunity constraint method is generally adopted in the safe and economic dispatching operation research considering the risk, but the solving process of the opportunity constraint method is relatively complex. In order to visually describe the operation risk of the power distribution network, randomness and uncertainty factors are quantitatively processed, a more accurate and appropriate method is needed, and meanwhile, the high efficiency of a calculation method of the method needs to be considered. In recent years, a number of scholars have extended the conditional value at risk (CVaR) method to the field of power system scheduling and operation research, and have made some technical breakthroughs. The risk measure is an analysis and estimation of risk level, and is one of the most important links in the risk management process, including measuring the probability of loss caused by various risks and the extent and degree of occurrence of the loss. Due to the fact that uncertainty of boundary conditions is enhanced, high-risk operation intervals under certain small-probability events can be formed in ADN operation control, and accurate evaluation of boundary risk of power distribution network scheduling operation under the condition of large-scale access of renewable energy sources is beneficial to reducing operation loss to the maximum extent and obtaining profit of system operation. In addition, in the actual operation process of the ADN, with the increase of prediction uncertainty, the adaptability of the traditional day-ahead planning and scheduling strategy is weak, while Model Predictive Control (MPC) can realize rolling optimization and consider the dynamic performance of the system, and if the MPC can be applied to the operation cost management of the power distribution network, the control of the tail risk is expected to be further enhanced.
Aiming at the active and reactive coordinated scheduling problem of ADN, more researches are carried out to comprehensively control various distributed energy sources and active management means so as to realize safe and economic operation of an active power distribution network, but the ADN has defects in the following aspects: firstly, response characteristics of control decision objects in the active power distribution network are different, and less research relates to designing part of operation control objects (such as DGs, ESS and the like) which are flexibly adjusted and quickly respond as decision variables of the second stage, so that a two-stage optimized operation method is formulated. Secondly, a large amount of distributed energy is accessed, and the uncertainty brought to the safe, reliable and economic operation of the system also needs corresponding deep analysis and overall optimization. In the traditional research, the expected value of the operation cost is simply used as an objective function, the boundary risk is ignored, and the scientificity and comprehensiveness are lost. How to measure the tail risk of high loss during the operation of the active power distribution network under the uncertain operation condition still needs to be explored. Meanwhile, as the prediction time scale becomes longer, the prediction accuracy is reduced, as the prediction uncertainty increases, the fluctuation of the operation parameters of the active power distribution network and the external disturbance are more frequent, the operation strategy of the active power distribution network needs finer time granularity and more flexible time scale, and the deterministic modeling or simple rule control aiming at a typical mode possibly faces failure, so the operation level of the system voltage is seriously influenced.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a risk-related predictive control method of a two-stage random model of an active power distribution network.
In order to achieve the above purpose, the predictive control method for the two-stage stochastic model of the active power distribution network considering the risk of the invention comprises the following steps:
1) considering the uncertainty and the randomness of photovoltaic output and load fluctuation, respectively setting the photovoltaic output and the load fluctuation to obey Beta distribution and normal distribution, performing repeated random sampling and statistical analysis on a simulation result by using Monte Carlo simulation in a set time scale to generate a large-scale uncertain scene set, and then reducing the large-scale uncertain scene set;
2) establishing an active and reactive power coordinated scheduling model of the active power distribution network by adopting a two-stage random optimization method, wherein the whole optimization control process is modeled in stages, and in the first stage, the action state and the action quantity of slow-motion equipment in the active power distribution network are determined without considering uncertainty; in the second stage, adding uncertainty, introducing condition risk value to measure the boundary risk of the operation of the active power distribution network, and measuring the loss degree by taking economic loss as a measurement index;
3) carrying out linear modeling processing on an OLTC (active load controller), a CB (circuit board), a PV (photovoltaic) inverter and an ESS (ESS) in the active power distribution network, and perfecting an active and reactive power coordination scheduling model of the active power distribution network by considering the operation constraint condition of the ADN (adaptive data network);
4) solving an active and reactive power coordinated scheduling model of the active power distribution network to obtain an optimized decision variable of two stages, and determining an output plan of each regulated resource in a scheduling period to form an optimal scheduling strategy;
5) at intervals of time TcStarting rolling optimization, and at each sampling initial time tkTaking the sampling value at the current moment as the initial state, aiming at the current prediction period TNRolling to solve the optimal scheduling strategy within the prediction period, wherein only t is executedk,tk+Tc]Control strategy of each regulated resource in time period and at tk,tk+4Tc]Keeping OLTC tap position and CB gear fixed in time period and waiting for next time tk+1＝tk+TcComes and then shifts the time window backward by a time interval TcAnd completing the two-stage stochastic model prediction control for promoting the coordinated optimization operation of the active power distribution network.
The specific operation of the step 4) is as follows: converting the active and reactive power coordinated scheduling model of the active power distribution network obtained in the step 3) into a mixed integer second-order cone planning problem, calculating optimization decision variables of two stages by solving the mixed integer second-order cone optimization problem, and determining output plans of each regulated and controlled resource in a scheduling period to form an optimal scheduling strategy.
And in the step 1), a large-scale uncertain scene set is reduced by adopting a synchronous back substitution reduction method.
The specific operation of the step 2) is as follows:
an active and reactive power coordinated scheduling model of the active power distribution network is established by adopting a two-stage random optimization method, wherein the whole optimization control process is modeled in stages, and the objective function of the model considers the electricity purchasing cost of a main network, the DG electricity generation cost, the network loss cost and the operation and maintenance costs of OLTC, CB, SVC, PV inverter and ESS, wherein the objective function EC of the first stageSIUncertainty is not considered, uncertainty is considered in the second phase, and the objective function EC of the second phaseSDScene-dependent changes;
wherein, the objective function of the first stage is:
Figure BDA0002588443330000051
wherein omegaGIs a contact node set of a regional power distribution network and an active power distribution network, omegaOLTC、ΩCB、ΩSVCAnd omegaInvFor each regulating device candidate node set, cG、cPVAnd cLossUnit cost for electricity purchase, PV power generation and network loss of active distribution network, respectively, cOLTC、cCB、cSVC、cInvAnd cESSFor the unit regulation cost of OLTC, CB, SVC, photovoltaic inverter and ESS in the active power distribution network,
Figure BDA0002588443330000052
represents the exchange power of the distribution network and the active distribution network connecting line in the t period region,
Figure BDA0002588443330000053
for the generated power of PV at node j during time t,
Figure BDA0002588443330000054
for the loss of line ij during period t,
Figure BDA0002588443330000055
and
Figure BDA0002588443330000056
is a first stage control variable in which,
Figure BDA0002588443330000057
and
Figure BDA0002588443330000058
respectively representing reactive power regulating quantities corresponding to the t period SVC, the photovoltaic inverter, the CB and the ESS,
Figure BDA0002588443330000059
is the gear change identification of the OLTC,
Figure BDA00025884433300000510
and
Figure BDA00025884433300000511
is a variable from 0 to 1, and is,
Figure BDA00025884433300000512
and
Figure BDA00025884433300000513
for OLTC gear change identification when
Figure BDA00025884433300000514
The gear value of the OLTC in the t-th period is smaller than the t-1 period,
Figure BDA00025884433300000515
and
Figure BDA00025884433300000516
meaning the same, T is the duration of the scheduling period.
The objective function for the second stage is:
Figure BDA00025884433300000517
where β is a given confidence level, and β ∈ (0,1), πsRepresenting the occurrence probability of a scene s, wherein rho is risk aversion and is used for representing the operation of a power distribution networkThe degree of aversion to risk by the practitioner.
The specific operation of the step 3) is as follows: and carrying out accurate linear modeling processing on OLTC, CB, SVC, PV inverters and ESS in the active power distribution network, and considering the operation constraint conditions of the active power distribution network so as to perfect an active and reactive power coordination scheduling model of the active power distribution network.
The voltage constraints of OLTC are:
Figure BDA0002588443330000061
wherein the content of the first and second substances,
Figure BDA0002588443330000062
is a voltage reference value, and is,
Figure BDA0002588443330000063
and
Figure BDA0002588443330000064
is the square of the upper and lower limits of the transformation ratio at both sides of the OLTC, lj,tIs the square of the OLTC transformation ratio, lj,tIs a discrete variable;
the number of actions of OLTC is constrained to:
Figure BDA0002588443330000065
wherein the content of the first and second substances,
Figure BDA0002588443330000066
and
Figure BDA0002588443330000067
is a variable from 0 to 1, indicating a change in the OLTC range when
Figure BDA0002588443330000068
The OLTC gear value is larger than the gear at the t-1 time period in the t-th time period, otherwise, the OLTC gear value is smaller than the gear at the t-1 time period in the t-th time period; SRjFor the maximum adjustment range of the OLTC gear,
Figure BDA0002588443330000069
limiting the maximum action times of the OLTC in the T time period;
the constraint conditions of the CB are as follows:
Figure BDA00025884433300000610
wherein the content of the first and second substances,
Figure BDA00025884433300000611
is 0-1 auxiliary variable;
the SVC constraints are:
Figure BDA00025884433300000612
wherein the content of the first and second substances,
Figure BDA0002588443330000071
and
Figure BDA0002588443330000072
respectively representing the upper and lower bounds of SVC reactive compensation output power
The constraint conditions of the photovoltaic inverter are as follows:
Figure BDA0002588443330000073
the modeling process of the ESS is represented as:
Figure BDA0002588443330000074
Figure BDA0002588443330000075
SOCi,T＝SOCi,0
Figure BDA0002588443330000076
Figure BDA0002588443330000077
Figure BDA0002588443330000078
Figure BDA0002588443330000079
Figure BDA00025884433300000710
therein, SOCi,tFor the state of charge, α, of an ESS connected at node i at time tiIs the self-discharge rate of the node;
Figure BDA00025884433300000711
and
Figure BDA00025884433300000712
respectively representing the charging power and the discharging power of the ESS at the node,
Figure BDA00025884433300000713
and
Figure BDA00025884433300000714
the charge-discharge efficiency is shown as follows,
Figure BDA00025884433300000715
and
Figure BDA00025884433300000716
represents the upper limit value of the charging and discharging power of the ESS at the node i,
Figure BDA00025884433300000717
and
Figure BDA00025884433300000718
is a binary variable used to indicate the charging and discharging state of the ESS.
The photovoltaic output is:
Figure BDA00025884433300000719
wherein the content of the first and second substances,
Figure BDA00025884433300000720
and predicting the force of the node j in the t-th period PV under the scene s.
The safety constraints are:
Figure BDA00025884433300000721
Figure BDA00025884433300000722
wherein the content of the first and second substances,
Figure BDA0002588443330000081
and
Figure BDA0002588443330000082
respectively representing the upper and lower limit values of the current of branch ij during period t,
Figure BDA0002588443330000083
and
Figure BDA0002588443330000084
respectively representing the upper and lower limits of the voltage at node j during time t.
The specific operation of the step 5) is as follows:
51) predicting the whole prediction period T according to historical dataNLoad fluctuations and photovoltaic output conditions, wherein uncertainty is not considered in the first phase;
52) introducing uncertainty in the second phase, taking into account [ t ]k+Tc,tk+TN]Generating N initial scenes by the load and photovoltaic output in a time period, and reducing the N initial scenes into N scenes;
53) determining T by solving a single deterministic mixed integer second order cone optimization problem and simultaneously calculating optimization decision variables of two stagesNThe output plan of each regulated resource in the time period is formed to TNAn optimal control strategy within a time period;
54) execute [ t ]k,tk+Tc]Optimal control strategy in time period, and at [ t ]k,tk+4Tc]Keeping OLTC tap position and CB gear fixed in time period, and waiting for next time tk+1＝tk+TcThen the time window is shifted back by one time interval.
The invention has the following beneficial effects:
according to the predictive control method of the two-stage stochastic model of the active power distribution network considering the risks, during specific operation, the whole optimization control process is modeled in stages, and in the first stage, the action state and the action quantity of slow-motion equipment in the active power distribution network are determined without considering uncertainty; in the second stage, uncertainty is added to reduce the action times of slow dynamic equipment, reduce the loss and aging cost of the slow dynamic equipment, fully exert the flexible adjustment characteristics of other equipment, simultaneously introduce conditional risk value to measure the boundary risk of the operation of the active power distribution network, measure the loss degree by taking economic loss as a measurement index, overcome the defects of qualitative analysis and subjective evaluation in the traditional research, and scientifically describe the tail risk and potential loss of the operation of the active power distribution network under the uncertain operation working condition by combining a continuous and scientific conditional risk value calculation method and a quantifiable analysis index. Finally, the model prediction control is applied to the ADN operation optimization considering the operation cost risk, the influence of the error between the predicted value and the actual value on the adjustment strategy can be effectively coped with, the boundary risk of the power distribution network scheduling operation under the large-scale access of renewable energy sources can be accurately evaluated, the operation loss is reduced to the maximum extent, and the system voltage operation level is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a system diagram of the present invention;
FIG. 2 is a schematic diagram of MPC based two-phase stochastic optimization.
Detailed Description
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
The following detailed description is exemplary in nature and is intended to provide further details of the invention. Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention.
Referring to fig. 1, the technical scheme adopted by the invention is as follows: the method comprises the steps of taking the operation cost of the power distribution network as a target, considering the limiting conditions of regulation and control equipment and the operation safety constraint of a power grid, controlling the regulation and control equipment in stages according to different response characteristics of an on-load voltage regulation tap (OLTC), a Capacitor Bank (CB), a Static Var Compensator (SVC), a Photovoltaic (PV) inverter and an ESS, considering uncertainty of load and renewable energy, bringing risks into an optimization problem, introducing a CVaR (dynamic voltage regulation and reactive power compensator) to carry out risk management, establishing a two-stage random optimization model considering the CVaR, and integrating the two-stage random optimization model considering the CVaR into an MPC (MPC) frame to realize rolling optimization solution. The method can realize the stable control of the voltage operation level of the power distribution network, overcomes the defects of qualitative analysis and subjective evaluation in the traditional research, and scientifically describes the tail risk and potential loss of the active power distribution network operation under the uncertain operation condition by combining a continuous and scientific calculation method and a quantifiable analysis index.
The two-stage stochastic model prediction control method for promoting coordinated optimization operation of the active power distribution network comprises the following steps of:
1) considering the uncertainty and the randomness of photovoltaic output and load fluctuation, setting the photovoltaic output and the load fluctuation to respectively obey Beta distribution and normal distribution, utilizing Monte Carlo simulation in a certain time scale, repeatedly and randomly sampling and statistically analyzing a simulation result to generate a large-scale uncertain scene set, and then reducing the large-scale uncertain scene set by adopting a scene reduction technology;
the specific process of the step 1) is as follows:
considering uncertainty of photovoltaic output and load fluctuation, respectively adopting Beta distribution and normal distribution to carry out scene error generation, fusing generated error scene data and predicted scene data within a certain time scale to serve as real-time data which cannot be obtained temporarily in the verification process, wherein the total number of generated scenes is assumed to be N, and the probability corresponding to each scene is pirR is 1,2, …, N. In order to achieve both of the calculation efficiency and the random variation characteristics of the data, it is necessary to extract setting characteristic information from the time-series data by using a scene reduction technique to form a typical scene.
Aiming at the selection problem of the typical scene, the invention adopts the synchronous back-substitution subtraction method to reduce the scale of the typical scene, the number of the reduced scenes is n, and the corresponding probability is pis,s＝1,2,…,n。
2) Establishing an active and reactive power coordinated scheduling model of the active power distribution network by adopting a two-stage random optimization method, wherein the whole control process is modeled in stages, and in the first stage, the action state and the action quantity of slow-motion equipment in the system are determined without considering uncertainty; in the second stage, adding uncertainty, introducing condition risk value to measure boundary risk of Active Distribution Network (ADN) operation, and measuring loss degree by taking economic loss as a measurement index;
the specific process of the step 2) is as follows:
an active and reactive power coordinated scheduling model of the active power distribution network is established by adopting a two-stage random optimization method, wherein the whole control process is modeled in stages, and the economic operation objective function of the active power distribution network takes the electricity purchasing cost of a main network, the DG electricity generation cost and the network loss into considerationAnd the costs and the operating and maintenance costs of the OLTC, CB, SVC, PV inverter, and ESS. Objective function EC of the first stageSIUncertainty is not considered, independent of scene changes (SI), and in the second phase uncertainty, its objective function EC is consideredSDDepending on the scene change (SD).
The objective function of the first stage optimization is:
Figure BDA0002588443330000111
wherein omegaGSet of contact nodes, omega, for regional distribution network and main networkOLTC、ΩCB、ΩSVCAnd omegaInvFor each regulating device candidate node set, cG、cPVAnd cLossUnit cost for electricity purchase, PV generation and network loss of the main network, respectively, cOLTC、cCB、cSVC、cInvAnd cESSFor the unit regulation cost of OLTC, CB, SVC, photovoltaic inverter and ESS,
Figure BDA0002588443330000112
representing the power exchange between the distribution network and the main network connecting line in the t period,
Figure BDA0002588443330000113
for the generated power of PV at node j during time t,
Figure BDA0002588443330000114
for the loss of line ij during period t,
Figure BDA0002588443330000115
and
Figure BDA0002588443330000116
is a first stage control variable in which,
Figure BDA0002588443330000121
and
Figure BDA0002588443330000122
respectively representing reactive power regulating quantities corresponding to the t period SVC, the photovoltaic inverter, the CB and the ESS,
Figure BDA0002588443330000123
is the gear change identification of the OLTC,
Figure BDA0002588443330000124
and
Figure BDA0002588443330000125
a variable of 0-1 for OLTC range change identification when
Figure BDA0002588443330000126
The gear value of the OLTC in the t-th period is smaller than the t-1 period,
Figure BDA0002588443330000127
and
Figure BDA0002588443330000128
meaning the same, T is the duration of the scheduling period.
The construction process of the second stage optimization objective function is as follows:
because uncertain factors are introduced in the optimization process of the second stage, an expected value of an objective function is generally taken as an optimization target, and compared with a deterministic method for accurately predicting the DG output magnitude, the method has certain advantages, but other influence parameters for characterizing cost distribution are often ignored, the expected cost value can present certain distribution characteristics, wherein the possibility of high cost in some cases is high, from the perspective of risk control, a certain risk measurement method needs to be adopted to measure the possibility magnitude of loss caused by various risks and the range and degree of loss occurrence, and the construction of the initial objective function of the second stage is as follows:
Figure BDA0002588443330000129
wherein the content of the first and second substances,
Figure BDA00025884433300001210
and
Figure BDA00025884433300001211
is a controlled variable for the second stage, wherein,
Figure BDA00025884433300001212
Figure BDA00025884433300001213
respectively representing reactive power regulating variables corresponding to the SVC, the photovoltaic inverter and the ESS in a t period under a scene s,
Figure BDA00025884433300001214
representing the power exchange between the power distribution network and the main network connecting line in the t period area under the scene s,
Figure BDA00025884433300001215
for the generated power of PV at node j during time t under scenario s,
Figure BDA00025884433300001216
the network loss of the branch ij in the t period under the scene s is shown.
On the basis of expected cost, the method brings risk into an optimization problem, applies CVaR to risk measurement of the operation cost of the power distribution network, avoids thick tail phenomenon in cost distribution, realizes effective management and control of risk, adjusts deviation of control strategies, enables the adaptability of the strategies to be stronger, and enables given confidence coefficient beta to be E (0,1) and VaRβThe probability that the operation cost is less than xi is larger than the corresponding threshold value when beta, xi is a variable, VaRβIs defined as:
Figure BDA0002588443330000131
CVaRβindicating that the operating cost of the distribution network exceeds VaRβAverage running cost in value, where,
Figure BDA0002588443330000132
wherein omegasFor uncertain scene sets, furthermore, continuous non-negative auxiliary variables eta are introducedSA value of equal to
Figure BDA0002588443330000133
CVaRβCan be expressed as:
Figure BDA0002588443330000134
Figure BDA0002588443330000135
Figure BDA0002588443330000136
Figure BDA0002588443330000137
thus, considering the influence of uncertainty factors, the final objective function of the second stage optimization, which takes into account the running cost risk, is represented as:
Figure BDA0002588443330000138
where β is a given confidence level, and β ∈ (0,1), πsThe probability of occurrence of a scene s is represented, rho is risk aversion degree, rho is used for representing the aversion degree of power distribution network operators to risks, and the larger the rho is, the more the operators tend to avoid the risks, and the decision is relatively conservative.
3) And carrying out linear modeling processing on the OLTC, CB, PV inverter, ESS and other controllable resources in the ADN, and considering the operation constraint conditions of the ADN to perfect an active and reactive power coordination scheduling model of the active power distribution network.
The specific process of the step 3) is as follows:
the method comprises the following steps of carrying out accurate linear modeling processing on OLTC, CB, SVC, PV inverters and ESS in the ADN, considering other operation constraint conditions of the ADN, perfecting an active and reactive power coordination scheduling model of the active power distribution network, and expressing detailed modeling processes and constraint conditions of active management elements as follows:
OLTC
in the operation process of the power distribution network, after the transformer substation is additionally provided with the OLTC, the voltage of the node of the transformer substation can be adjusted, and the voltage constraint of the OLTC is as follows:
Figure BDA0002588443330000141
wherein the content of the first and second substances,
Figure BDA0002588443330000142
is a voltage reference value, and is,
Figure BDA0002588443330000143
and
Figure BDA0002588443330000144
is the square of the upper and lower limits of the transformation ratio at both sides of the OLTC, lj,tIs the square of the OLTC transformation ratio, lj,tFor discrete variables, for clarity, let lj,tFurther processed into a form containing a variable of 0-1, i.e.
Figure BDA0002588443330000145
Wherein r isj,dRepresents the increment of the square of the transformation ratio of the adjacent gears of the OLTC, namely the difference between the square of the transformation ratio of the gear D and the D-1, D is the gear set of the OLTC,
Figure BDA0002588443330000146
being a 0-1 auxiliary variable, since OLTC is limited by equipment wear and aging, the number of OLTC actions needs to be necessarily constrained, namely:
Figure BDA0002588443330000147
wherein the content of the first and second substances,
Figure BDA0002588443330000148
and
Figure BDA0002588443330000149
is a variable from 0 to 1, indicating a change in the OLTC range when
Figure BDA00025884433300001410
The OLTC gear value is larger than the gear at the t-1 time period in the t period, otherwise, the OLTC gear value is smaller than the gear at the t-1 time period in the t period; SRjFor the maximum adjustment range of the OLTC gear,
Figure BDA0002588443330000151
the maximum limit action number of OLTC in the T period.
Capacitor bank
The CB reactive compensation quantity is only related to the number of groups to which the CB reactive compensation quantity is put, and the relationship between the CB reactive compensation quantity and the node voltage is not considered for the moment, and in practical application, the compensation quantity of the CB is a discrete variable, and the reactive compensation quantity is:
Figure BDA0002588443330000152
wherein the content of the first and second substances,
Figure BDA0002588443330000153
the number of groups put into reactive compensation for the CB accessed to the node j at the time t,
Figure BDA0002588443330000154
in the case of a discrete variable, the number of discrete variables,
Figure BDA0002588443330000155
the maximum number of access CB groups for node j,
Figure BDA0002588443330000156
the reactive compensation power that each group of CB can provide is shown, similar to OLTC, and is limited by the loss of CB equipment, and the action times constraint of adding the CB is as follows:
Figure BDA0002588443330000157
wherein the content of the first and second substances,
Figure BDA0002588443330000158
for limiting the number of CB operations, for simplifying the processing
Figure BDA0002588443330000159
Then there are:
Figure BDA00025884433300001510
the constraints of CB are therefore:
Figure BDA00025884433300001511
wherein the content of the first and second substances,
Figure BDA00025884433300001512
is an auxiliary variable of 0-1, similar to
Figure BDA00025884433300001513
And
Figure BDA00025884433300001514
a variable of 0-1, indicating a change in the number of CB input groups, if
Figure BDA00025884433300001515
The number of CB commissioning groups is increased during the t-th period compared to the t-1 period.
Static reactive compensator
SVC can be continuously adjusted, and the SVC has the following constraints:
Figure BDA0002588443330000161
wherein the content of the first and second substances,
Figure BDA0002588443330000162
and
Figure BDA0002588443330000163
respectively representing the upper and lower bounds of the SVC reactive compensation output power.
Photovoltaic inverter
The available reactive power support capacity of the photovoltaic inverter is determined by the apparent power capacity and the active power output of the current photovoltaic, and the constraint conditions of the photovoltaic inverter are as follows:
Figure BDA0002588443330000164
ESS
the modeling process of the ESS can be expressed as:
Figure BDA0002588443330000165
Figure BDA0002588443330000166
SOCi,T＝SOCi,0
Figure BDA0002588443330000167
Figure BDA0002588443330000168
Figure BDA0002588443330000169
Figure BDA00025884433300001610
Figure BDA00025884433300001611
therein, SOCi,tIs the state of charge (SOC) of the ESS connected to node i at time t, αiIs the self-discharge rate of the node;
Figure BDA00025884433300001612
and
Figure BDA00025884433300001613
respectively representing the charging power and the discharging power of the ESS at the node,
Figure BDA00025884433300001614
and
Figure BDA00025884433300001615
the charge-discharge efficiency is shown as follows,
Figure BDA00025884433300001616
and
Figure BDA00025884433300001617
represents the upper limit value of the charging and discharging power of the ESS at the node i,
Figure BDA00025884433300001618
and
Figure BDA00025884433300001619
is a binary variable used to indicate the charging and discharging state of the ESS.
The photovoltaic output is:
Figure BDA0002588443330000171
wherein the content of the first and second substances,
Figure BDA0002588443330000172
and predicting the force of the node j in the t-th period PV under the scene s.
The safety constraints are:
Figure BDA0002588443330000173
Figure BDA0002588443330000174
wherein the content of the first and second substances,
Figure BDA0002588443330000175
and
Figure BDA0002588443330000176
respectively representing the upper and lower limit values of the current of branch ij during period t,
Figure BDA0002588443330000177
and
Figure BDA0002588443330000178
respectively representing the upper and lower limits of the voltage at node j during time t.
4) Carrying out convex relaxation on a power flow equation based on a second-order cone relaxation technology, converting the active and reactive power coordination scheduling model of the active power distribution network obtained in the step 3) into a mixed integer second-order cone planning problem, calculating optimization decision variables of two stages by solving the mixed integer second-order cone optimization problem, and determining an output plan of each regulated and controlled resource in a scheduling period;
the specific process of the step 4) is as follows:
aiming at a radiation type power distribution network, the branch power flow model is constructed by taking photovoltaic power generation as a representative of renewable energy, and then the branch power flow form of the active and reactive power coordinated scheduling model of the active power distribution network is expressed as follows:
Figure BDA0002588443330000179
Figure BDA00025884433300001710
Figure BDA00025884433300001711
Figure BDA00025884433300001712
wherein the content of the first and second substances,(j)representing a set of branch end nodes, pi, with j as head-end node(j)Represents a set of branch head nodes with j as an end node, and Ω is a set of all nodes in the network, ΩLineFor all line sets in the network, Iij,tRepresenting the corresponding branch current at time t, rijRepresenting the resistance, x, of branch ijijRepresenting the reactance value, b, on branch ijjSusceptance, P, representing node jj,tAnd Qj,tRespectively representing the active and reactive injection power of the node j in the period t,
Figure BDA0002588443330000181
the active injection quantity of the upper-level power grid of the node j in the period t is represented,
Figure BDA0002588443330000182
the active output value of the photovoltaic power generation is shown,
Figure BDA0002588443330000183
Figure BDA0002588443330000184
and
Figure BDA0002588443330000185
respectively representing the idle work output by nodes j of a superior power grid, CB, SVC and photovoltaic inverter in the t-th period, Pij,tAnd Qij,tRespectively representing the active and reactive power on branch ij during time t,
Figure BDA0002588443330000186
and
Figure BDA0002588443330000187
respectively representing the active load and the reactive load of a node j in a period t;
order:
Figure BDA0002588443330000188
Figure BDA0002588443330000189
then, the branch load flow form of the active and reactive power coordinated scheduling model of the active power distribution network is expressed as follows:
Figure BDA00025884433300001810
Figure BDA00025884433300001811
Figure BDA00025884433300001812
Figure BDA00025884433300001813
therefore, the active and reactive power coordinated scheduling model of the active power distribution network is converted into a mixed integer second-order cone planning problem, the optimization decision variables of two stages are calculated by solving the mixed integer second-order cone optimization problem, and output plans of various regulated and controlled resources in a scheduling period are determined.
5) At intervals of time TcStarting rolling optimization, and at each sampling initial time tkTaking the sampling value at the current moment as the initial state, aiming at the current prediction period TNRolling to solve the optimal scheduling strategy within the prediction period, wherein only t is executedk,tk+Tc]Control strategy of each regulated resource in time period and at tk,tk+4Tc]Keeping OLTC tap position and CB tap position fixed for a period of time until the next time tk+1＝tk+TcThe time window is shifted backwards by a time interval TcAnd repeating the process.
Referring to fig. 2, the specific process of step 5) is:
the MPC is applied to ADN operation optimization considering the operation cost risk, so that the boundary risk of the dispatching operation of the power distribution network under the large-scale access of renewable energy sources can be accurately evaluated, the operation loss can be reduced to the maximum extent, and the profit of system operation can be obtained. The invention divides the two-stage random optimization into a static optimization stage (first stage) which is advanced by 1h and a prediction rolling optimization stage (second stage) which is advanced by 15min based on an MPC framework, and the coordination relationship between the two stages is shown in FIG. 2.
It should be noted that the tap position of the OLTC and the adjustment amount of the CB are defined as first-stage variables, and the SVC, the photovoltaic inverter, and the ESS can be adjusted in two stages. The two-stage coordination optimization is carried out by integrating the adjusting performance of various control devices, so that the effect of smoothly adjusting reactive power can be realized, the frequent action of related devices can be avoided, and the specific optimization steps in the two stages can be summarized as follows:
51) predicting the whole prediction period T according to historical dataNLoad fluctuations and photovoltaic output conditions within, uncertainty not considered in the first phase;
52) introducing uncertainty in the second phase, taking into account [ t ]k+Tc,tk+TN]Load and photovoltaic output (T) over a period of timec15min), N initial scenes are generated, and the scenes are reduced to N, so as to improve the calculation efficiency.
53) Determining T by solving a single deterministic mixed integer second order cone optimization problem and simultaneously calculating optimization decision variables of two stagesNPlanning the output of each regulated resource in a time period;
54) execute [ t ]k,tk+Tc]Control strategy of each regulated resource in time period and at tk,tk+4Tc]Keeping OLTC tap position and CB gear fixed in time period, and waiting for next time tk+1＝tk+TcThe time window is shifted backward by a time interval and the process is repeated.
According to the invention, the action states and action quantities of slow-motion devices (OLTC and CB) are set as decision variables of a first stage by combining the response characteristics of different control devices, and operation control objects (SVC, inverter and ESS) capable of flexibly adjusting and quickly responding are designed as decision variables of a second stage, so that a two-stage optimized operation method is formulated, the action times of the devices such as OLTC and CB are effectively reduced, the loss and aging cost of the devices are reduced, and the flexible adjustment characteristics of the SVC, the inverter and the ESS are more fully exerted.
In addition, CVaR is introduced to carry out risk control on an optimization target, so that boundary risk of operation of the active power distribution network is measured, economic loss is used as a measurement index to measure loss degree, the defects of qualitative analysis and subjective evaluation in traditional research are overcome, and tail risk and potential loss of operation of the active power distribution network under uncertain operation conditions are scientifically described by combining a continuous and scientific condition risk value calculation method and a quantifiable analysis index.
Finally, the MPC strategy is applied to ADN operation optimization considering the operation cost risk, the influence of the error between the predicted value and the actual value on the adjustment strategy is effectively coped with, the boundary risk of the distribution network scheduling operation under the large-scale access of renewable energy sources is favorably and accurately evaluated, the operation loss is reduced to the maximum extent, and the profit of system operation is obtained.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.

Claims (10)

1. A risk-related prediction control method for a two-stage stochastic model of an active power distribution network is characterized by comprising the following steps:
1) considering the uncertainty and the randomness of photovoltaic output and load fluctuation, respectively setting the photovoltaic output and the load fluctuation to obey Beta distribution and normal distribution, performing repeated random sampling and statistical analysis on a simulation result by using Monte Carlo simulation in a set time scale to generate a large-scale uncertain scene set, and then reducing the large-scale uncertain scene set;
2) establishing an active and reactive power coordinated scheduling model of the active power distribution network by adopting a two-stage random optimization method, wherein the whole optimization control process is modeled in stages, and in the first stage, the action state and the action quantity of slow-motion equipment in the active power distribution network are determined without considering uncertainty; in the second stage, adding uncertainty, introducing condition risk value to measure the boundary risk of the operation of the active power distribution network, and measuring the loss degree by taking economic loss as a measurement index;
3) carrying out linear modeling processing on an OLTC (active load controller), a CB (circuit board), a PV (photovoltaic) inverter and an ESS (ESS) in the active power distribution network, and perfecting an active and reactive power coordination scheduling model of the active power distribution network by considering the operation constraint condition of the ADN (adaptive data network);
4) solving an active and reactive power coordinated scheduling model of the active power distribution network to obtain an optimized decision variable of two stages, and determining an output plan of each regulated resource in a scheduling period to form an optimal scheduling strategy;
5) at intervals of time TcStarting rolling optimization, and at each sampling initial time tkTaking the sampling value at the current moment as the initial state, aiming at the current prediction period TNRolling to solve the optimal scheduling strategy within the prediction period, wherein only t is executedk,tk+Tc]Control strategy of each regulated resource in time period and at tk,tk+4Tc]Keeping OLTC tap position and CB gear fixed in time period and waiting for next time tk+1＝tk+TcComes and then shifts the time window backward by a time interval TcAnd completing the two-stage stochastic model prediction control for promoting the coordinated optimization operation of the active power distribution network.
2. The predictive control method for the risk-considering two-stage stochastic model of the active power distribution network according to claim 1, wherein the specific operations of step 4) are as follows: converting the active and reactive power coordinated scheduling model of the active power distribution network obtained in the step 3) into a mixed integer second-order cone planning problem, calculating optimization decision variables of two stages by solving the mixed integer second-order cone optimization problem, and determining output plans of each regulated and controlled resource in a scheduling period to form an optimal scheduling strategy.
3. The predictive control method for the risk-considering two-stage stochastic model of the active power distribution network according to claim 1, wherein the large-scale uncertain scene set is reduced by a synchronous back-substitution subtraction method in step 1).
4. The predictive control method for the risk-considering two-stage stochastic model of the active power distribution network according to claim 1, wherein the specific operations in step 2) are as follows:
an active and reactive power coordinated scheduling model of the active power distribution network is established by adopting a two-stage random optimization method, wherein the whole optimization control process is modeled in stages, and the objective function of the model considers the electricity purchasing cost of a main network, the DG electricity generation cost, the network loss cost and the operation and maintenance costs of OLTC, CB, SVC, PV inverter and ESS, wherein the objective function EC of the first stageSIUncertainty is not considered, uncertainty is considered in the second phase, and the objective function EC of the second phaseSDScene-dependent changes;
wherein, the objective function of the first stage is:
Figure FDA0002588443320000021
wherein omegaGIs a contact node set of a regional power distribution network and an active power distribution network, omegaOLTC、ΩCB、ΩSVCAnd omegaInvFor each regulating device candidate node set, cG、cPVAnd cLossUnit cost for electricity purchase, PV power generation and network loss of active distribution network, respectively, cOLTC、cCB、cSVC、cInvAnd cESSIs activeThe unit regulation cost of OLTC, CB, SVC, photovoltaic inverter and ESS in the distribution network,
Figure FDA0002588443320000022
represents the exchange power of the distribution network and the active distribution network connecting line in the t period region,
Figure FDA0002588443320000023
for the generated power of PV at node j during time t,
Figure FDA0002588443320000024
for the loss of line ij during period t,
Figure FDA0002588443320000025
and
Figure FDA0002588443320000026
is a first stage control variable in which,
Figure FDA0002588443320000031
and
Figure FDA0002588443320000032
respectively representing reactive power regulating quantities corresponding to the t period SVC, the photovoltaic inverter, the CB and the ESS,
Figure FDA0002588443320000033
is the gear change identification of the OLTC,
Figure FDA0002588443320000034
and
Figure FDA0002588443320000035
is a variable from 0 to 1, and is,
Figure FDA0002588443320000036
and
Figure FDA0002588443320000037
for OLTC gear change identification when
Figure FDA0002588443320000038
The gear value of the OLTC in the t-th period is smaller than the t-1 period,
Figure FDA0002588443320000039
and
Figure FDA00025884433200000310
the meanings are the same, and T is the duration of a scheduling period;
the objective function for the second stage is:
Figure FDA00025884433200000311
where β is a given confidence level, and β ∈ (0,1), πsAnd representing the occurrence probability of the scene s, wherein rho is risk aversion degree, and is used for representing the aversion degree of the power distribution network operating personnel to the risks.
5. The predictive control method for the risk-considering two-stage stochastic model of the active power distribution network according to claim 1, wherein the specific operation of step 3) is: and carrying out accurate linear modeling processing on OLTC, CB, SVC, PV inverters and ESS in the active power distribution network, and considering the operation constraint conditions of the active power distribution network so as to perfect an active and reactive power coordination scheduling model of the active power distribution network.
6. The predictive control method for the risk-taking account of the two-stage stochastic model of the active power distribution network according to claim 5, wherein the voltage constraint of the OLTC is as follows:
Figure FDA00025884433200000312
wherein the content of the first and second substances,
Figure FDA00025884433200000313
is a voltage reference value, and is,
Figure FDA00025884433200000314
and
Figure FDA00025884433200000315
is the square of the upper and lower limits of the transformation ratio at both sides of the OLTC, lj,tIs the square of the OLTC transformation ratio, lj,tIs a discrete variable;
the number of actions of OLTC is constrained to:
Figure FDA0002588443320000041
wherein the content of the first and second substances,
Figure FDA0002588443320000042
and
Figure FDA0002588443320000043
is a variable from 0 to 1, indicating a change in the OLTC range when
Figure FDA0002588443320000044
The OLTC gear value is larger than the gear at the t-1 time period in the t-th time period, otherwise, the OLTC gear value is smaller than the gear at the t-1 time period in the t-th time period; SRjFor the maximum adjustment range of the OLTC gear,
Figure FDA0002588443320000045
limiting the maximum action times of the OLTC in the T time period;
the constraint conditions of the CB are as follows:
Figure FDA0002588443320000046
wherein the content of the first and second substances,
Figure FDA0002588443320000047
is 0-1 auxiliary variable;
the SVC constraints are:
Figure FDA0002588443320000048
wherein the content of the first and second substances,
Figure FDA0002588443320000049
and
Figure FDA00025884433200000410
respectively representing the upper and lower bounds of SVC reactive compensation output power
The constraint conditions of the photovoltaic inverter are as follows:
Figure FDA00025884433200000411
7. the risk-aware predictive control method for two-stage stochastic models of active power distribution networks according to claim 5, wherein the modeling process of the ESS is represented as:
Figure FDA0002588443320000051
Figure FDA0002588443320000052
SOCi,T＝SOCi,0
Figure FDA0002588443320000053
Figure FDA0002588443320000054
Figure FDA0002588443320000055
Figure FDA0002588443320000056
Figure FDA0002588443320000057
therein, SOCi,tFor the state of charge, α, of an ESS connected at node i at time tiIs the self-discharge rate of the node;
Figure FDA0002588443320000058
and
Figure FDA0002588443320000059
respectively representing the charging power and the discharging power of the ESS at the node,
Figure FDA00025884433200000510
and
Figure FDA00025884433200000511
the charge-discharge efficiency is shown as follows,
Figure FDA00025884433200000512
and Pi dch,maxRepresents the upper limit value of the charging and discharging power of the ESS at the node i,
Figure FDA00025884433200000513
and
Figure FDA00025884433200000514
is a binary variable used to indicate the charging and discharging state of the ESS.
8. The predictive control method for the risk-aware two-stage stochastic model of the active power distribution network of claim 5, wherein the photovoltaic output is:
Figure FDA00025884433200000515
wherein the content of the first and second substances,
Figure FDA00025884433200000516
and predicting the force of the node j in the t-th period PV under the scene s.
9. The predictive control method for the risk-aware two-stage stochastic model of the active power distribution network according to claim 5, wherein the safety constraints are:
Figure FDA00025884433200000517
Figure FDA00025884433200000518
wherein the content of the first and second substances,
Figure FDA00025884433200000519
and
Figure FDA00025884433200000520
respectively representing the upper and lower limit values of the current of branch ij during period t,
Figure FDA00025884433200000521
and
Figure FDA00025884433200000522
respectively representing the upper and lower limits of the voltage at node j during time t.
10. The predictive control method for the risk-considering two-stage stochastic model of the active power distribution network according to claim 1, wherein the specific operations of step 5) are as follows:
51) predicting the whole prediction period according to the historical dataTNLoad fluctuations and photovoltaic output conditions, wherein uncertainty is not considered in the first phase;
52) introducing uncertainty in the second phase, taking into account [ t ]k+Tc,tk+TN]Generating N initial scenes by the load and photovoltaic output in a time period, and reducing the N initial scenes into N scenes;
53) determining T by solving a single deterministic mixed integer second order cone optimization problem and simultaneously calculating optimization decision variables of two stagesNThe output plan of each regulated resource in the time period is formed to TNAn optimal control strategy within a time period;
54) execute [ t ]k,tk+Tc]Optimal control strategy in time period, and at [ t ]k,tk+4Tc]Keeping OLTC tap position and CB gear fixed in time period, and waiting for next time tk+1＝tk+TcThen the time window is shifted back by one time interval.
CN202010688381.2A 2020-07-16 2020-07-16 Predictive control method of risk-considering two-stage random model of active power distribution network Active CN111799847B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010688381.2A CN111799847B (en) 2020-07-16 2020-07-16 Predictive control method of risk-considering two-stage random model of active power distribution network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010688381.2A CN111799847B (en) 2020-07-16 2020-07-16 Predictive control method of risk-considering two-stage random model of active power distribution network

Publications (2)

Publication Number Publication Date
CN111799847A true CN111799847A (en) 2020-10-20
CN111799847B CN111799847B (en) 2022-11-22

Family

ID=72807475

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010688381.2A Active CN111799847B (en) 2020-07-16 2020-07-16 Predictive control method of risk-considering two-stage random model of active power distribution network

Country Status (1)

Country Link
CN (1) CN111799847B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113095951A (en) * 2021-05-08 2021-07-09 西安科技大学 Intelligent regulation and control method, equipment and device for gas extraction and storage medium
CN113110561A (en) * 2021-05-24 2021-07-13 四川大学 Random distribution robust optimization-based SMPC algorithm for maintaining formation of satellite
CN113162060A (en) * 2021-03-17 2021-07-23 武汉工程大学 Opportunity constraint optimization-based active power distribution network two-stage reactive power regulation method
CN113364043A (en) * 2021-05-08 2021-09-07 国网浙江省电力有限公司丽水供电公司 Micro-grid group optimization method based on condition risk value
CN113516278A (en) * 2021-04-26 2021-10-19 山东大学 Active power distribution network multi-time scale active and reactive power coordinated optimization scheduling method and system

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103544656A (en) * 2013-10-24 2014-01-29 清华大学 Active power distribution network operational control method based on minimum carbon emission
CN104866924A (en) * 2015-05-26 2015-08-26 清华大学 Active power distribution network planning and operation combined optimization method
CN104993522A (en) * 2015-06-30 2015-10-21 中国电力科学研究院 Active power distribution network multi-time scale coordinated optimization scheduling method based on MPC
CN105550766A (en) * 2015-12-04 2016-05-04 山东大学 Micro-grid robustness multi-target operation optimization method containing renewable energy resources
CN106712031A (en) * 2017-02-10 2017-05-24 东南大学 Sequence-robustness optimizing dispatching system and dispatching method of active power distribution network considering uncertainty
CN106786806A (en) * 2016-12-15 2017-05-31 国网江苏省电力公司南京供电公司 A kind of power distribution network active reactive based on Model Predictive Control coordinates regulation and control method
CN107147152A (en) * 2017-06-15 2017-09-08 广东工业大学 New energy power distribution network polymorphic type active reactive source cooperates with Optimal Configuration Method and system
CN108448585A (en) * 2018-03-29 2018-08-24 清华大学 A kind of electric network swim equation solution method of linearization based on data-driven
CN109687510A (en) * 2018-12-11 2019-04-26 东南大学 A kind of meter and probabilistic power distribution network Multiple Time Scales optimizing operation method
CN109861204A (en) * 2018-12-18 2019-06-07 青岛理工大学 Active distribution network cooperative control system and method based on Model Predictive Control
CN109873447A (en) * 2019-02-19 2019-06-11 国网江苏省电力有限公司南京供电分公司 A kind of multi-source collaboration active-idle regulation method of the more time stages of active distribution network
CN110009262A (en) * 2019-04-28 2019-07-12 国网福建省电力有限公司福州供电公司 A kind of a few days ago-in a few days two stages Optimization Scheduling of active distribution network
CN110148969A (en) * 2019-03-26 2019-08-20 上海电力学院 Active distribution network optimizing operation method based on model predictive control technique
CN110298138A (en) * 2019-07-09 2019-10-01 南方电网科学研究院有限责任公司 A kind of integrated energy system optimization method, device, equipment and readable storage medium storing program for executing
CN110460036A (en) * 2019-05-10 2019-11-15 四川大学 A kind of probabilistic alternating current-direct current power distribution network distributed optimization method of consideration wind-powered electricity generation

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103544656A (en) * 2013-10-24 2014-01-29 清华大学 Active power distribution network operational control method based on minimum carbon emission
CN104866924A (en) * 2015-05-26 2015-08-26 清华大学 Active power distribution network planning and operation combined optimization method
CN104993522A (en) * 2015-06-30 2015-10-21 中国电力科学研究院 Active power distribution network multi-time scale coordinated optimization scheduling method based on MPC
CN105550766A (en) * 2015-12-04 2016-05-04 山东大学 Micro-grid robustness multi-target operation optimization method containing renewable energy resources
CN106786806A (en) * 2016-12-15 2017-05-31 国网江苏省电力公司南京供电公司 A kind of power distribution network active reactive based on Model Predictive Control coordinates regulation and control method
CN106712031A (en) * 2017-02-10 2017-05-24 东南大学 Sequence-robustness optimizing dispatching system and dispatching method of active power distribution network considering uncertainty
CN107147152A (en) * 2017-06-15 2017-09-08 广东工业大学 New energy power distribution network polymorphic type active reactive source cooperates with Optimal Configuration Method and system
CN108448585A (en) * 2018-03-29 2018-08-24 清华大学 A kind of electric network swim equation solution method of linearization based on data-driven
CN109687510A (en) * 2018-12-11 2019-04-26 东南大学 A kind of meter and probabilistic power distribution network Multiple Time Scales optimizing operation method
CN109861204A (en) * 2018-12-18 2019-06-07 青岛理工大学 Active distribution network cooperative control system and method based on Model Predictive Control
CN109873447A (en) * 2019-02-19 2019-06-11 国网江苏省电力有限公司南京供电分公司 A kind of multi-source collaboration active-idle regulation method of the more time stages of active distribution network
CN110148969A (en) * 2019-03-26 2019-08-20 上海电力学院 Active distribution network optimizing operation method based on model predictive control technique
CN110009262A (en) * 2019-04-28 2019-07-12 国网福建省电力有限公司福州供电公司 A kind of a few days ago-in a few days two stages Optimization Scheduling of active distribution network
CN110460036A (en) * 2019-05-10 2019-11-15 四川大学 A kind of probabilistic alternating current-direct current power distribution network distributed optimization method of consideration wind-powered electricity generation
CN110298138A (en) * 2019-07-09 2019-10-01 南方电网科学研究院有限责任公司 A kind of integrated energy system optimization method, device, equipment and readable storage medium storing program for executing

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
NIKOLAOS G. PATERAKIS，等: "A Multi-Objective Optimization Approach to Risk-Constrained Energy and Reserve Procurement Using Demand Response", 《IEEE TRANSACTIONS ON POWER SYSTEMS》 *
任佳依，等: "基于模型预测控制的主动配电网多时间尺度有功无功协调调度", 《中国电机工程学报》 *
王海冰，等: "考虑条件风险价值的两阶段发电调度随机规划模型和方法", 《中国电机工程学报》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113162060A (en) * 2021-03-17 2021-07-23 武汉工程大学 Opportunity constraint optimization-based active power distribution network two-stage reactive power regulation method
CN113516278A (en) * 2021-04-26 2021-10-19 山东大学 Active power distribution network multi-time scale active and reactive power coordinated optimization scheduling method and system
CN113516278B (en) * 2021-04-26 2023-08-22 山东大学 Active power distribution network multi-time scale active and reactive power coordination optimization scheduling method and system
CN113095951A (en) * 2021-05-08 2021-07-09 西安科技大学 Intelligent regulation and control method, equipment and device for gas extraction and storage medium
CN113364043A (en) * 2021-05-08 2021-09-07 国网浙江省电力有限公司丽水供电公司 Micro-grid group optimization method based on condition risk value
CN113095951B (en) * 2021-05-08 2023-09-22 西安科技大学 Intelligent gas extraction regulation and control method, equipment, device and storage medium
CN113110561A (en) * 2021-05-24 2021-07-13 四川大学 Random distribution robust optimization-based SMPC algorithm for maintaining formation of satellite

Also Published As

Publication number Publication date
CN111799847B (en) 2022-11-22

Similar Documents

Publication Publication Date Title
CN111799847B (en) Predictive control method of risk-considering two-stage random model of active power distribution network
Hu et al. Load following of multiple heterogeneous TCL aggregators by centralized control
CN110298138B (en) Comprehensive energy system optimization method, device, equipment and readable storage medium
Mohamed et al. Real-time energy management scheme for hybrid renewable energy systems in smart grid applications
CN106505635B (en) Active scheduling model and scheduling system with minimum wind abandon
CN110690732B (en) Photovoltaic reactive power partition pricing power distribution network reactive power optimization method
CN110782363A (en) AC/DC power distribution network scheduling method considering wind power uncertainty
CN110581571A (en) dynamic optimization scheduling method for active power distribution network
CN114336702B (en) Wind-solar storage station group power distribution collaborative optimization method based on double-layer random programming
CN109378861B (en) Robust optimization scheduling method for active power distribution network considering time-space correlation
CN111092429A (en) Optimized scheduling method of flexible interconnected power distribution network, storage medium and processor
CN114597969B (en) Power distribution network double-layer optimization method considering intelligent soft switch and virtual power plant technology
Wang et al. Two-stage full-data processing for microgrid planning with high penetrations of renewable energy sources
Dong et al. Optimal scheduling framework of electricity-gas-heat integrated energy system based on asynchronous advantage actor-critic algorithm
Wang et al. Multi-objective robust optimization of hybrid AC/DC distribution networks considering flexible interconnection devices
MansourLakouraj et al. Multi-timescale risk-constrained volt/var control of distribution grids with electric vehicles and solar inverters
Laouafi et al. An evaluation of conventional and computational intelligence methods for medium and long-term load forecasting in Algeria
CN116722605A (en) Power distribution network scheduling optimization method based on Internet of things
CN112699562A (en) Method and terminal for constructing power distribution network architecture
CN105207255B (en) A kind of power system peak regulation computational methods suitable for wind power output
Farjah Proposing an efficient wind forecasting agent using adaptive MFDFA
CN113013884B (en) Three-section type reactive voltage control method for photovoltaic power distribution system with high permeability
Han et al. Regression model-based adaptive receding horizon control of soft open points for loss minimization in distribution networks
Sahoo et al. Forecasting Tariff Rates and Enhancing Power Quality in Microgrids: The Synergistic Role of LSTM and UPQC
Wang et al. Intraday net load reserve demand assessment based on catboost and kernel density estimation

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