CN115000961B - Line loss prediction calculation method and system - Google Patents
Line loss prediction calculation method and system Download PDFInfo
- Publication number
- CN115000961B CN115000961B CN202210929518.8A CN202210929518A CN115000961B CN 115000961 B CN115000961 B CN 115000961B CN 202210929518 A CN202210929518 A CN 202210929518A CN 115000961 B CN115000961 B CN 115000961B
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- output voltage
- photovoltaic field
- data
- grading
- 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
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 23
- 238000005286 illumination Methods 0.000 claims abstract description 43
- 238000003062 neural network model Methods 0.000 claims abstract description 22
- 230000005540 biological transmission Effects 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 11
- 230000003203 everyday effect Effects 0.000 claims description 4
- 230000005684 electric field Effects 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- H02J3/38—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
- 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
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Power Engineering (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention provides a line loss prediction calculation method and a system, which comprises the following steps: collecting time-series output voltage, wind direction, wind speed and illumination intensity data of each photovoltaic field; grading the collected data; grouping data into sample data, carrying out neural network model training, collecting output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field in real time, processing the data into grading data, and predicting the change rate of the voltage of the photovoltaic field to be measured in the next period; and calculating the output voltage of the next period, and predicting and calculating the line loss of the photovoltaic field to be measured in the next period according to the output voltage of the next period. The technical problem that line loss prediction calculation is inaccurate is solved through the scheme.
Description
Technical Field
The invention relates to the field of line loss calculation of power transmission lines, in particular to a line loss prediction calculation method and system based on machine learning.
Background
Although the power transmission line uses good conductors such as copper and aluminum as leads to transmit electric energy, the power transmission line still has a certain resistance value, especially when the line is relatively long, the resistance value is considerable when accumulated, and the consumed electric energy accounts for seven percent and eight percent of the transmitted electric energy. According to the principle of power consumption, when the transmission line is fixed, i.e. the resistance is constant, the loss is higher if the voltage level is lower. Distributed photovoltaic power plants are generally large in distribution area, and there may be a large distance from the photovoltaic field to the inverter, so that the line resistance is large, and meanwhile, the output voltage of the photovoltaic field is generally low, so that the line loss is relatively large. Because the photovoltaic grid fluctuates greatly, power supplementation is usually required to output stable power, and it is important to accurately predict and calculate the change of line loss in order to accurately control the grid power.
However, the photovoltaic power generation is affected by meteorological factors such as irradiance, sunshine duration and cloud amount, and when the illumination fluctuates, the output voltage of the photovoltaic field fluctuates, so that the line loss is not fixed, and the ultra-short-term fluctuation problem exists. According to the loss principle, predicting the output voltage is the key to accurately calculating the line loss. At present, ultra-short-term prediction of a photovoltaic field is mainly performed through historical weather, real-time weather, historical sunshine, real-time sunshine and the like, for example, CN102281016A discloses a clear sky photovoltaic ultra-short-term power prediction method based on a real-time radiation acquisition technology, and CN101969207A discloses a photovoltaic ultra-short-term power prediction method combining satellite remote sensing and meteorological telemetry technologies.
Disclosure of Invention
The invention provides a line loss prediction calculation method and system, and aims to solve the problem of large calculation error of ultra-short-term line loss in a photovoltaic system.
In one aspect of the present invention, a line loss prediction calculation method is provided, where the line loss prediction calculation method is applied to a distributed photovoltaic field, and the distributed photovoltaic field includes a plurality of independent photovoltaic fields; the method is characterized in that: the line loss prediction calculation method comprises the following steps: synchronously acquiring output voltage, wind direction, wind speed and illumination intensity of each independent photovoltaic field in a first period to obtain time-series output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field; calculating a change rate sequence of the time-series output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence; classifying the output voltage change of the Tth +1 th period of the photovoltaic field to be tested into a label, inputting the output voltage change of the Tth period of all other photovoltaic fields, the wind direction, the wind speed, the illumination intensity classification sequence and the coordinate of the corresponding photovoltaic field to form sample data, and performing neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested; wherein the serial numbers of the T grading sequences are 1, 2, 3 and 4, 8230, N-1 and N are total data amount; acquiring output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field in real time and processing the data into grading data, inputting real-time grading data except the photovoltaic field to be detected and corresponding photovoltaic field coordinate data into a trained neural network model of the photovoltaic field to be detected, predicting output voltage change grading of the photovoltaic field to be detected in the next period, and further determining the change rate of the voltage of the next period; and calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the voltage of the next period, and predicting and calculating the line loss of the next period of the photovoltaic field to be measured according to the output voltage of the next period.
Furthermore, the distributed photovoltaic fields refer to more than 10 independent photovoltaic fields in the same photovoltaic management system, and the same photovoltaic management system is characterized in that each photovoltaic field is connected to the same transmission scheduling system and is managed by the same system.
Further, sample data in which the output voltage change rates are all 0 are removed.
Further, each independent photovoltaic field in the distributed photovoltaic fields is sequentially set as a photovoltaic field to be tested, and a model is trained for each independent photovoltaic field.
Further, data was collected every day every 2 minutes from sunrise, and data was collected for the entire year.
The invention provides a line loss prediction calculation system, which is applied to a distributed photovoltaic field, wherein the distributed photovoltaic field comprises a plurality of independent photovoltaic fields; the method is characterized in that:
the line loss prediction computing system comprises the following steps: the acquisition module is used for synchronously acquiring the output voltage, the wind direction, the wind speed and the illumination intensity of each independent photovoltaic field in a first period to obtain time-series output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field; the grading module is used for calculating a change rate sequence of the time-series output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence; the training module is used for carrying out neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested by using the output voltage variation classification of the T +1 th period of the photovoltaic field to be tested as a label and the output voltage variation, wind direction, wind speed, illumination intensity classification sequence of the T th period of all other photovoltaic fields and the coordinate of the corresponding photovoltaic field as input to form the sample data; wherein the serial numbers of the T hierarchical sequences are 1, 2, 3 and 4 \8230, the value of N-1, N is the total data volume; the prediction module is used for acquiring the output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field in real time and processing the data into grading data, inputting the real-time grading data of the photovoltaic field to be detected and the coordinate data of the corresponding photovoltaic field into the trained neural network model of the photovoltaic field to be detected, predicting the output voltage change grading of the next period of the photovoltaic field to be detected, and further determining the change rate of the voltage of the next period; and the calculating module is used for calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the voltage of the next period, and predicting and calculating the line loss of the next period of the photovoltaic field to be measured according to the output voltage of the next period.
Furthermore, the distributed photovoltaic fields refer to more than 10 independent photovoltaic fields in the same photovoltaic management system, and the same photovoltaic management system is characterized in that each photovoltaic field is connected to the same transmission scheduling system and is managed by the same system.
Further, sample data in which the output voltage change rates are all 0 are removed.
Further, each independent photovoltaic field in the distributed photovoltaic fields is sequentially set as a photovoltaic field to be tested, and a model is trained for each independent photovoltaic field.
Further, data was collected every day every 2 minutes from sunrise, and data was collected for the entire year.
According to the technical scheme, the photovoltaic electric field to be measured is subjected to ultra-short-term output voltage change prediction according to the real-time output voltage change rate of other photovoltaic fields in the distributed photovoltaic field and the conditions of real-time sunlight, wind direction and the like, so that the problem of inaccuracy caused by direct use of meteorological data is avoided, and the accuracy rate of predicting and calculating line loss is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed 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 creative efforts.
FIG. 1 is a schematic view of two photovoltaic fields of the present invention;
fig. 2 is a flowchart of a line loss prediction calculation method according to the present invention.
Detailed Description
The invention is described in detail with reference to the drawings and the detailed description.
As shown in fig. 1, the output voltage of the photovoltaic field a is reduced due to the covering of the black clouds, the photovoltaic field B is not affected at this time, but the black clouds move to the photovoltaic field B as time goes on, and similar influence is caused on the photovoltaic field B in a certain period of time later, so that the change rate of the output voltage of the photovoltaic field B can be predicted according to the position, the wind direction, the illumination and the change rate of the output voltage of the photovoltaic field a; based on the principle, the output voltage change condition of the subsequent photovoltaic field B is predicted according to the current output voltage change condition of the photovoltaic field A; for a system with a plurality of photovoltaic fields, as long as the output of one field is changed, the change of other scenes in the subsequent time can be predicted in time so as to facilitate the system to make adjustment in time, and the following implementation mode is specifically adopted:
in one embodiment, a line loss prediction calculation method is provided, and is applied to a distributed photovoltaic field, where the distributed photovoltaic field includes a plurality of independent photovoltaic fields, and a flow of the method is shown in fig. 2.
Compared with a centralized photovoltaic field, the distributed photovoltaic field is built in a desert, gobi and other areas by centralized large-area photovoltaic, and the distributed photovoltaic field fully utilizes idle land resources and can be distributed on a roof, a water surface and the like; the distributed photovoltaic field comprises a plurality of disconnected and independent photovoltaic electric fields, and the scale, the output power, the position coordinates and the like of each photovoltaic electric field are possibly different, such as roof electric fields of a plurality of common users, electric fields of a plurality of fishery aquaculture water surfaces and the like; the output voltage of the distributed photovoltaic field is low, so that the line loss fluctuation is large; furthermore, the distributed photovoltaic farms mean that more than 10 independent photovoltaic farms are arranged in the same photovoltaic management system, and the same photovoltaic management system is characterized in that each photovoltaic farm is connected to the same transmission scheduling system and managed by the same system; distributed photovoltaic fields generally have the characteristics of wide range and uneven distribution; distributed photovoltaic fields managed by the same photovoltaic management center may be distributed in a range of hundreds of square kilometers, so that weather conditions between the distributed photovoltaic fields may be greatly different, and some photovoltaic fields may be covered by dark clouds and some may be clear in weather at the same time.
Synchronously acquiring output voltage, wind direction, wind speed and illumination intensity of each independent photovoltaic field in a first period to obtain time-sequenced output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field;
for the convenience of subsequent data processing, the output voltage of each independent photovoltaic field is synchronously collected in the same period.
For example, the voltage data acquisition is performed in a time period of 2 minutes, that is, the output voltage of the photovoltaic scene is acquired every 2 minutes.
Exemplarily, in order to maintain synchronism, all photovoltaic fields adopt the following time sequence (with a period of 2 minutes) 12; in the example, from 12 o' clock, the voltage was collected every 2 minutes.
The first period can be freely set, obviously, the shorter the period is, the denser the data is, the higher the prediction precision is by adopting the method, but the model training is slower; the longer the period, the less the data, and although a certain precision is sacrificed, the processing speed is greatly improved, and those skilled in the art can freely select the data according to the system hardware condition and the requirement on the precision, which is not limited by the present application.
Based on the foregoing time sequence, illustratively, the photovoltaic field a acquires a series of voltage data: 376 370, 365, \ 8230 \ 8230;, which corresponds to time point 12;
similarly, data acquisition is performed on all the independent photovoltaic fields in the management system, and a series of time sequence data of output voltage of each independent photovoltaic field is obtained.
Similar to the way of collecting the output voltage, the wind direction, the wind speed and the illumination intensity of each independent photovoltaic field are collected in the same first period. In order to maintain time synchronization, data acquisition is similarly performed at the same point in time and interval.
And calculating a change rate sequence of the time-series output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence.
On one hand, due to the influence on weather changes, the absolute variation between different photovoltaic fields may be relatively large, but the variation ratios are similar; on the other hand, for the transmission management system, the output voltages of all the current photovoltaic fields are known, and only the voltage after one regulation period needs to be concerned when prediction regulation is carried out, namely the management system is concerned about the fluctuation change condition of the voltage in the short term and the future so as to deal with the fluctuation change condition in advance, so that the management system is more concerned about the change rate of the voltage; meanwhile, the absolute value difference of the output voltages of different fields is large, if the absolute value of the voltage is directly used for prediction training, the dimensionality of the trained data is exponentially increased, and the training process is extremely long; the change rate is usually one to two percentage points, the change range is limited, the grading setting is easy to carry out, and the total data volume after grading is greatly reduced.
Therefore, in order to speed up the training, further, a sequence of rates of change is calculated from the output voltage of each photovoltaic field, the rates of change filtering out the absolute value of the voltage.
Illustratively, the photovoltaic field a acquires a series of data: 376 370, 365, 369, 8230, 0, -0.01596, -0.01351,0.01096, 8230, 8230and the like.
Further, in order to facilitate data training, the change rate is graded into change grades; illustratively, the ranking is done according to the rate of change as described in Table 1.
The above-mentioned rate of change is classified into 0, -4, -3,3 \8230;.
All the photovoltaic fields are subjected to the grading sequence processing to obtain the grading sequence of each independent photovoltaic field, the grading sequences are expressed, the voltage change is unified to the same dimension, the training calculated amount is greatly reduced, and the training speed is improved.
Preferably, similar to the processing of voltage data, in order to reduce the data volume and speed up the training, the wind direction, the wind speed and the illumination intensity are processed in a grading manner to obtain a corresponding grouping sequence.
Illustratively, the wind direction is divided into: north, northeast, east, southeast, south, southwest, west and northwest directions.
Illustratively, wind speeds are classified one per 0.5 level, such as 0, 0.5, 1, 1.5, 2, 2.5 \8230;.
Illustratively, the illumination intensity is graded every 0.5 Wanks, such as 0-0.5 Wanks at grade 1, 0.5 to 1 Wanks at grade 2, 1 to 1.5 Wanks at grade 3 \8230; \ 8230;.
Classifying the output voltage change of the Tth +1 th period of the photovoltaic field to be tested into a label, inputting the output voltage change of the Tth period of all other photovoltaic fields, the wind direction, the wind speed, the illumination intensity classification sequence and the coordinate of the corresponding photovoltaic field to form sample data, and performing neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested; the T grading sequence numbers are 1, 2, 3 and 4, 8230, N-1 and N are total data amount.
Illustratively, the output voltage of the T +1 th cycle of the photovoltaic field a with respect to the T cycle is changed to 3 steps, more specifically, the T +1 th cycle in the photovoltaic field a is 13 points 02 minutes, and the output voltage at the T cycle 13 is changed to 3 steps; in the T period, namely 13 points, the output voltage changes, the wind direction, the wind speed, the illumination intensity and the coordinates of other photovoltaic fields are divided into (2, east, 1,9, (73.2, 94.3)), (2, east, 1, 10, (73.2, 93.3)), (1, east, 1, 10, (73.7, 95.3)), (1, northeast, 1,9, (76.2, 92.4)) \\ 8230 \
Then, as shown in the following formula, the output voltage variation, wind direction, wind speed, illumination intensity and coordinate of the T periods of other photovoltaic fields and the voltage variation of the T period of the current photovoltaic field to be measured form a sample
All training data can be acquired by the same method, and model training can be performed after enough training data are acquired.
Preferably, in order to cover all seasonal variations, data of at least one year is acquired as training data; in detail, data may be collected for a whole year, one data per day every 2 minutes from sunrise, the total number of times collected being the total data volume N.
Preferably, in order to reduce the data amount, sample data with output voltage change rates all being 0 are removed, that is, when the output voltage in one sample changes all being 0, that is, the output voltage at the sampling time does not change, and this data has no change meaning and cannot contribute to prediction, so that the data is removed.
And training by adopting sample data to obtain the trained neural network model. For a specific neural network model, the present application is not limited specifically, and since data is processed, most neural networks in the prior art can be applied to the present application, such as CNN, GAN networks, and the like.
Further, each independent photovoltaic field in the distributed photovoltaic fields is sequentially set as a photovoltaic field to be tested, and a model is trained for each independent photovoltaic field, so that each independent photovoltaic field can be predicted.
The method comprises the steps of collecting output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field in real time, preprocessing the data, inputting the preprocessed output voltage change, wind direction, wind speed, illumination intensity and coordinate data outside the photovoltaic field to be detected into a trained neural network model, predicting the output voltage change grade of the photovoltaic field to be detected in the next period, and determining the voltage change rate.
Illustratively, the output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field are collected in real time, and are processed in the same way as training data, namely the voltage change, the wind direction, the wind speed and the illumination intensity are classified; more specific examples, take 15: the data for the other photovoltaic fields in the system at 00 hours are (1, west, 1,9, (73.2, 94.3)), (0, northwest, 1, 10, (73.2, 93.3)), (0, west, 1, 10, (73.7, 95.3)), (1, northwest, 1,9, (76.2, 92.4)) \\ 8230; \ 8230;, inputting a trained neural network model of the photovoltaic field to be predicted, outputting a prediction result of the next period of the photovoltaic field to be predicted with a change grade of 1 grade, and predicting that the change rate of the photovoltaic field to be predicted in the next period is +0.5%.
And calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the output voltage of the next period, and predicting and calculating the loss of the photovoltaic field to be measured according to the output voltage of the next period.
Exemplarily, the photovoltaic field to be tested 15: the voltage output by 00 is 376V, the predicted change rate is +0.5%, the predicted voltage in the next period is 376 x (1 + 0.5%), and the predicted line loss can be calculated through the predicted voltage.
In a second embodiment, another aspect provides a line loss prediction computing system, which is applied to a distributed photovoltaic field, where the distributed photovoltaic field includes a plurality of independent photovoltaic fields.
The system comprises:
the acquisition module is used for synchronously acquiring the output voltage, the wind direction, the wind speed and the illumination intensity of each independent photovoltaic field in a first period to obtain time-sequenced output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field;
and the grading module is used for calculating a change rate sequence of the time-series output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence.
The training module is used for carrying out neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested by using the output voltage variation classification of the T +1 th period of the photovoltaic field to be tested as a label and the output voltage variation, wind direction, wind speed, illumination intensity classification sequence of the T th period of all other photovoltaic fields and the coordinate of the corresponding photovoltaic field as input to form the sample data; the numbers of the T hierarchical sequences are 1, 2, 3 and 4 \8230, the values of 8230are \8230, and the N-1, N is the total data volume.
And the prediction module is used for acquiring the output voltage change, wind direction, wind speed and illumination intensity data of each independent photovoltaic field in real time, preprocessing the output voltage change, wind direction, wind speed, illumination intensity and coordinate data of the photovoltaic field to be detected after preprocessing are input into the trained neural network model, predicting the output voltage change grade of the photovoltaic field to be detected in the next period, and further determining the voltage change rate.
And the calculation module is used for calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the output voltage of the next period, and predicting and calculating the loss of the photovoltaic field to be measured according to the output voltage of the next period.
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.
The present invention is not limited to the specific module configuration described in the related art. The prior art mentioned in the background section and the detailed description section can be used as part of the invention to understand the meaning of some technical features or parameters. The scope of the present invention is defined by the claims.
Claims (10)
1. A line loss prediction calculation method is applied to a distributed photovoltaic field, and the distributed photovoltaic field comprises a plurality of independent photovoltaic fields; the method is characterized in that:
the line loss prediction calculation method comprises the following steps:
s1, synchronously acquiring output voltage, wind direction, wind speed and illumination intensity of each independent photovoltaic field in a first period to obtain time-sequenced output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field;
s2, calculating a change rate sequence of the time-sequenced output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence;
s3, classifying the output voltage change rate of the T +1 th period of the photovoltaic field to be tested as a label, inputting the output voltage change rate of the T-th period of all other photovoltaic fields, the wind direction, the wind speed, the illumination intensity classification sequence and the coordinate of the corresponding photovoltaic field to form sample data, and performing neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested; wherein the number of the grading sequence is 1, 2, 3 and 4 \8230, the value of N-1, N is the total data volume;
s4, collecting the output voltage change rate, the wind direction, the wind speed and the illumination intensity data of each independent photovoltaic field in real time and processing the data into grading data, inputting the real-time grading data of the photovoltaic field to be detected and the corresponding photovoltaic field coordinate data into the trained neural network model of the photovoltaic field to be detected, predicting the output voltage change rate grading of the next period of the photovoltaic field to be detected, and further determining the change rate of the voltage of the next period;
and S5, calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the voltage of the next period, and predicting and calculating the line loss of the next period of the photovoltaic field to be measured according to the output voltage of the next period.
2. A line loss prediction calculation method as defined in claim 1, wherein: the distributed photovoltaic fields refer to more than 10 independent photovoltaic fields in the same photovoltaic management system, and the same photovoltaic management system is characterized in that each photovoltaic field is connected to the same transmission scheduling system and is managed by the same transmission scheduling system.
3. A line loss prediction calculation method as defined in claim 1, wherein: removing sample data with output voltage change rate of 0.
4. A line loss prediction calculation method as defined in claim 1, wherein: and sequentially setting each independent photovoltaic field in the distributed photovoltaic fields as a photovoltaic field to be tested, and training a model for each independent photovoltaic field.
5. A line loss prediction calculation method as defined in claim 1, wherein: data was collected every 2 minutes from sunrise every day for the entire year.
6. A line loss prediction computing system for application to a distributed photovoltaic farm, the distributed photovoltaic farm comprising a plurality of independent photovoltaic farms; the method is characterized in that:
the line loss prediction calculation system comprises:
the acquisition module is used for synchronously acquiring the output voltage, the wind direction, the wind speed and the illumination intensity of each independent photovoltaic field in a first period to obtain time-sequenced output voltage, wind direction, wind speed and illumination intensity data of each independent photovoltaic field;
the grading module is used for calculating a change rate sequence of the time-series output voltage data of each independent photovoltaic field, grading the change rate, and grading the wind direction, the wind speed and the illumination intensity to obtain a grading sequence;
the training module is used for forming sample data by taking the output voltage change rate of the T +1 th period of the photovoltaic field to be tested as a label and the output voltage change rate, the wind direction, the wind speed, the illumination intensity grading sequence of the T th period of all other photovoltaic fields and the coordinate of the corresponding photovoltaic field as input, and performing neural network model training by using the sample data to obtain a trained neural network model of the photovoltaic field to be tested; wherein the number of the grading sequence is 1, 2, 3, 4, 8230, N-1, N is total data amount;
the prediction module is used for acquiring the output voltage change rate, the wind direction, the wind speed and the illumination intensity data of each independent photovoltaic field in real time and processing the data into grading data, inputting the real-time grading data of the photovoltaic field to be detected and the corresponding photovoltaic field coordinate data into the trained neural network model of the photovoltaic field to be detected, predicting the output voltage change rate grading of the next period of the photovoltaic field to be detected, and further determining the change rate of the voltage of the next period;
and the calculation module is used for calculating the output voltage of the next period according to the real-time output voltage of the photovoltaic field to be measured and the change rate of the voltage of the next period, and predicting and calculating the line loss of the next period of the photovoltaic field to be measured according to the output voltage of the next period.
7. A line loss prediction computing system as defined in claim 6, wherein: the distributed photovoltaic fields refer to more than 10 independent photovoltaic fields in the same photovoltaic management system, and the same photovoltaic management system refers to the fact that all the photovoltaic fields are connected to the same transmission scheduling system and are managed by the same transmission scheduling system.
8. A line loss prediction computing system as defined in claim 6, wherein: removing sample data with output voltage change rate of 0.
9. A line loss prediction computing system as defined in claim 6, wherein: and sequentially setting each independent photovoltaic field in the distributed photovoltaic fields as a photovoltaic field to be tested, and training a model for each independent photovoltaic field.
10. A line loss prediction computing system as defined in claim 6, wherein: data was collected every 2 minutes from sunrise every day for the entire year.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210929518.8A CN115000961B (en) | 2022-08-04 | 2022-08-04 | Line loss prediction calculation method and system |
| CN202211603224.2A CN115940145A (en) | 2022-08-04 | 2022-08-04 | Line loss prediction method for distributed power station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210929518.8A CN115000961B (en) | 2022-08-04 | 2022-08-04 | Line loss prediction calculation method and system |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211603224.2A Division CN115940145A (en) | 2022-08-04 | 2022-08-04 | Line loss prediction method for distributed power station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115000961A CN115000961A (en) | 2022-09-02 |
| CN115000961B true CN115000961B (en) | 2022-11-08 |
Family
ID=83021888
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211603224.2A Pending CN115940145A (en) | 2022-08-04 | 2022-08-04 | Line loss prediction method for distributed power station |
| CN202210929518.8A Active CN115000961B (en) | 2022-08-04 | 2022-08-04 | Line loss prediction calculation method and system |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211603224.2A Pending CN115940145A (en) | 2022-08-04 | 2022-08-04 | Line loss prediction method for distributed power station |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN115940145A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103439572A (en) * | 2013-08-15 | 2013-12-11 | 国家电网公司 | Electricity larceny prevention monitoring method based on photovoltaic power generation power prediction |
| CN105243447A (en) * | 2015-10-16 | 2016-01-13 | 河南工业大学 | Grain yield prediction method and device |
| CN106374831A (en) * | 2016-09-26 | 2017-02-01 | 国电南瑞科技股份有限公司 | Distributed photovoltaic monitoring system with power generation forecasting and monitoring method |
| CN109978005A (en) * | 2019-02-25 | 2019-07-05 | 深圳市中电数通智慧安全科技股份有限公司 | A kind of fire alarm method, device, storage medium and terminal device |
| CN110363334A (en) * | 2019-06-21 | 2019-10-22 | 西安工程大学 | Grid-connected grid line loss prediction technique based on Grey Neural Network Model |
| CN110766212A (en) * | 2019-10-15 | 2020-02-07 | 哈尔滨工程大学 | Ultra-short-term photovoltaic power prediction method for historical data missing electric field |
| CN112686472A (en) * | 2021-01-22 | 2021-04-20 | 国网河南省电力公司许昌供电公司 | Power prediction method for distributed photovoltaic equivalent power station |
| CN112994010A (en) * | 2021-01-27 | 2021-06-18 | 河海大学 | AC/DC hybrid power distribution network security domain calculation method considering photovoltaic output fluctuation |
| WO2021238505A1 (en) * | 2020-05-27 | 2021-12-02 | 华北电力大学 | Federated learning-based regional photovoltaic power probability prediction method, and cooperative regulation and control system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322364B (en) * | 2019-06-19 | 2020-08-25 | 山东大学 | Short-term photovoltaic power generation prediction method and system |
-
2022
- 2022-08-04 CN CN202211603224.2A patent/CN115940145A/en active Pending
- 2022-08-04 CN CN202210929518.8A patent/CN115000961B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103439572A (en) * | 2013-08-15 | 2013-12-11 | 国家电网公司 | Electricity larceny prevention monitoring method based on photovoltaic power generation power prediction |
| CN105243447A (en) * | 2015-10-16 | 2016-01-13 | 河南工业大学 | Grain yield prediction method and device |
| CN106374831A (en) * | 2016-09-26 | 2017-02-01 | 国电南瑞科技股份有限公司 | Distributed photovoltaic monitoring system with power generation forecasting and monitoring method |
| CN109978005A (en) * | 2019-02-25 | 2019-07-05 | 深圳市中电数通智慧安全科技股份有限公司 | A kind of fire alarm method, device, storage medium and terminal device |
| CN110363334A (en) * | 2019-06-21 | 2019-10-22 | 西安工程大学 | Grid-connected grid line loss prediction technique based on Grey Neural Network Model |
| CN110766212A (en) * | 2019-10-15 | 2020-02-07 | 哈尔滨工程大学 | Ultra-short-term photovoltaic power prediction method for historical data missing electric field |
| WO2021238505A1 (en) * | 2020-05-27 | 2021-12-02 | 华北电力大学 | Federated learning-based regional photovoltaic power probability prediction method, and cooperative regulation and control system |
| CN112686472A (en) * | 2021-01-22 | 2021-04-20 | 国网河南省电力公司许昌供电公司 | Power prediction method for distributed photovoltaic equivalent power station |
| CN112994010A (en) * | 2021-01-27 | 2021-06-18 | 河海大学 | AC/DC hybrid power distribution network security domain calculation method considering photovoltaic output fluctuation |
Non-Patent Citations (3)
| Title |
|---|
| Ultra-Short-Term Industrial Power Demand Forecasting Using LSTM Based Hybrid Ensemble Learning;Mao Tan等;《IEEE Transactions on Power Systems》;20191231;全文 * |
| 基于CNN-GRU的光伏电站电压轨迹预测;冯裕祺等;《中国电力》;20220731;全文 * |
| 基于模糊神经网络的光伏发电量短期预测;张玉等;《电子设计工程》;20170120(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115000961A (en) | 2022-09-02 |
| CN115940145A (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106485093B (en) | Based on the solar irradiance time series synthetic method for improving Markov chain | |
| CN113496311A (en) | Photovoltaic power station generated power prediction method and system | |
| CN108388956A (en) | Consider the photovoltaic power prediction technique of attenuation | |
| Yang et al. | A novel ARX-based multi-scale spatio-temporal solar power forecast model | |
| CN113722375B (en) | Double-layer hybrid photovoltaic prediction method based on multi-algorithm optimization and computer equipment | |
| CN111626473A (en) | Two-stage photovoltaic power prediction method considering error correction | |
| CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
| Anuradha et al. | Analysis of solar power generation forecasting using machine learning techniques | |
| CN115441447B (en) | New energy power generation power prediction method | |
| CN113837426A (en) | Weather-typing-based photovoltaic power prediction method | |
| Savchenko et al. | Improving the efficiency of solar power plants based on forecasting the intensity of solar radiation using artificial neural networks | |
| CN103700028A (en) | Power consumption control method and device | |
| CN115000961B (en) | Line loss prediction calculation method and system | |
| CN116800172B (en) | Photoelectric conversion method and energy storage system | |
| Wang et al. | Neural network based irradiance mapping model of solar PV power forecasting using sky image | |
| CN103268416A (en) | Wind speed prediction method based on STCP-BP | |
| CN116484998A (en) | Distributed photovoltaic power station power prediction method and system based on meteorological similar day | |
| CN115276002A (en) | Intelligent power distribution property studying and judging method based on power load curve | |
| CN112116127B (en) | Photovoltaic power prediction method based on association of meteorological process and power fluctuation | |
| Cheng et al. | Temperature & humidity-weighted-modified GRNN based prediction model of photovoltaic power generation | |
| Zhao et al. | An Integrated Fishery Meteorological Information Service Terminal Based on End-Side Deep Learning Technology | |
| CN118052338B (en) | Distributed photovoltaic power prediction method based on wide-area cooperative training | |
| CN110874611B (en) | Wind power cluster power ultra-short term prediction error grade grading compensation method | |
| Alanqar et al. | Solar Power Monitoring and Forecasting System using Artificial Intelligence | |
| Ibrahim | Modelling of intraday Photovoltaic power production |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230321 Address after: 257200 No. 3, Haining Road, Hekou District, Shandong, Dongying Patentee after: State Grid Shandong electric power company Dongying Hekou District Power Supply Co. Patentee after: Dongying Power Industry Bureau of State Grid Shandong Electric Power Company Address before: 257200 No. 3, Haining Road, Hekou District, Shandong, Dongying Patentee before: State Grid Shandong electric power company Dongying Hekou District Power Supply Co. |
|
| TR01 | Transfer of patent right |