CN108710966B - Photovoltaic power generation power prediction method based on multi-cluster ESN neural network - Google Patents
Photovoltaic power generation power prediction method based on multi-cluster ESN neural network Download PDFInfo
- Publication number
- CN108710966B CN108710966B CN201810352268.XA CN201810352268A CN108710966B CN 108710966 B CN108710966 B CN 108710966B CN 201810352268 A CN201810352268 A CN 201810352268A CN 108710966 B CN108710966 B CN 108710966B
- Authority
- CN
- China
- Prior art keywords
- cluster
- output
- neural network
- esn
- prediction
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 37
- 238000010248 power generation Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000004451 qualitative analysis Methods 0.000 claims abstract description 19
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 17
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 19
- 238000005259 measurement Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims description 3
- 230000006855 networking Effects 0.000 claims description 3
- 230000000737 periodic effect Effects 0.000 claims description 3
- 239000002243 precursor Substances 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012886 linear function Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008121 plant development Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000002910 structure generation Methods 0.000 description 1
Images
Classifications
-
- 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
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
-
- 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
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention discloses a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network, and belongs to the field of photovoltaic power generation. Firstly, constructing a multi-cluster ESN neural network, and adopting an improved ESN network, namely the multi-cluster ESN network; secondly, establishing a multi-cluster ESN neural network prediction model, and analyzing the influence of 24-step hysteresis information on the accuracy of the prediction model by adding corresponding historical 24-step hysteresis information in an input layer when the prediction model is established; and finally, evaluating the prediction performance of the multi-cluster ESN neural network. And quantitative analysis and qualitative analysis are respectively carried out on the prediction results, so that reliable and efficient operation of the photovoltaic power generation system and safe dispatching of a power grid are ensured.
Description
Technical Field
The invention relates to the technical field of power prediction of photovoltaic power generation systems, in particular to a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network.
Background
In recent years, with the increase of global energy demand, renewable energy sources (such as wind energy and solar energy) have attracted attention, wherein solar energy has the advantages of being renewable, pollution-free, safe and reliable. Due to its ready availability, government support and continued development of technology, large-scale Photovoltaic (PV) systems have been widely used around the world. However, the power generated by a photovoltaic system is affected by many uncontrollable factors, such as temperature, humidity, wind speed, wind direction, solar radiation intensity, seasonality, etc., so that the power generation of the photovoltaic system exhibits characteristics of high nonlinearity, randomness, complexity, etc. In order to ensure reliable and efficient operation of a photovoltaic power generation system and safe scheduling of a power grid, the method has important significance for accurately predicting the photovoltaic power generation amount.
For the prediction of the photovoltaic power generation power, a plurality of documents and patents are provided for the prediction, such as the application number of 201310040995X, the name of the invention is 'a photovoltaic power generation prediction system based on a T-S type fuzzy neural network'; the invention has the application number of 201510762155.3, and is named as a short-term power prediction method of a photovoltaic power generation system; the application number is 201410299370.X, and the name is 'photovoltaic power generation prediction method based on GRNN neural network', and the like. Still other papers have also described Photovoltaic Power Generation Power predictions, such as the design of Photovoltaic Power Generation and Model Optimization by Das, Utpal Kumar et al, An analog sensitive for short-term Photovoltaic Power plant development by Alessandrini S et al, and the like.
These patents and papers employ different techniques and methods for modeling and predicting photovoltaic power generation systems. However, they have the disadvantages of complicated training process and the like, and do not perform qualitative analysis on the prediction result. Therefore, the invention provides a photovoltaic power generation power prediction method based on a Multi-cluster Echo State network (MCESN), and the method is used for carrying out quantitative and qualitative analysis on test results respectively.
Disclosure of Invention
The invention aims to provide a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network, and the method is used for respectively carrying out quantitative analysis and qualitative analysis on prediction results and ensuring the reliable and efficient operation of a photovoltaic power generation system and the safe scheduling of a power grid.
The invention provides a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network, which comprises the following steps:
s1: constructing a multi-cluster ESN neural network;
s2, initializing each parameter of the constructed multi-cluster ESN neural network;
s3: considering influence factors, establishing a multi-cluster ESN prediction model;
s4: calling sample information, and training the initialized multi-cluster ESN neural network;
s5: calling a test sample to test the network;
s6: analyzing and evaluating the output result obtained in the test stage;
further, the multi-cluster ESN neural network model is constructed by the following steps:
s11: initializing n precursor nodes in a two-dimensional plane, and fully connecting the nodes;
s12: adding local area nodes to the two-dimensional plane;
s13: connecting the newly added local node with the existing node according to the local priority connection rule;
s14: repeating the steps S12 and S13 until the total number of the nodes reaches the set size N of the reserve pool;
s15: generating a reserve pool connection weight matrix, and calculating to obtain a state updating equation of the reserve pool:
x(t+1)＝f(Winu(t+1)+Wresx(t)+Wbacky(t)+v(t))
the scales of the multiple ESN network input units u (t), the pool unit x (t) and the output unit y (t) are K, N, L respectively. Win、Wres、Wout、WbackRespectively representing an NxK input connection weight matrix, an NxN internal connection weight matrix, an LxN output connection weight matrix and an NxL feedback connection weight matrix. f denotes the excitation function (usually Sigmoid function) of the internal unit, and v (t) is the noise signal.
Further, the output calculation of the network in S15 adopts the following function:
y(t+1)＝fout(Woutx(t+1))
wherein f isoutWhich is the activity function (usually a linear function or Sigmoid function) of the output unit, a linear output unit is used here.
Further, the photovoltaic power generation prediction model based on the multi-cluster ESN neural network comprises a single-step MCESN prediction model and a multi-step MCESN prediction model;
the single-step MCESN prediction model is a 1-hour-ahead short-term prediction, and comprises four different input-output short-term prediction models based on temperature and humidity measurement values and two different input-output short-term prediction models based on periodic characteristics, as shown in FIG. 4, wherein P ispv(t-1)、Ppv(T-24) historical one-step and historical 24-step photovoltaic power generation power values, Tpv(t-1)、Hpv(t-1) corresponding historical one-step temperature and humidity measurements, PpvAnd (t) is the photovoltaic power generation power value needing to be predicted.
The multi-step MCESN prediction model is a multi-input multi-output model with advanced 24-hour medium-term prediction.
Further, the expression of the multi-step MCESN prediction model is as follows:
wherein, the input signals are respectively the power average value of 24 hours of the current day (k)
Corresponding average value of temperature
And days k per month, the output is the power value for the next 24 hours. The left side of the equal sign represents the photovoltaic power generation power value to be predicted which is ahead of one day (24 hours), namely the output of the network; the right side of the equation represents the input signal of the network;
an input-output nonlinear function fitted to the neural network.
Further, all power, temperature history data is expressed in hours as a two-dimensional (2-D) matrix:
wherein, the row number m and the column number s respectively represent the total days and the hours of each day.
Further, the evaluation based on predictive performance includes quantitative and qualitative analysis;
quantitative analysis is to quantify the error between the predicted value and the measured value;
the qualitative analysis is to analyze the internal characteristic difference between the measured value and the predicted value;
further, the quantitative analysis mainly evaluates the error size of the prediction result from four indexes, namely standard square root error (NRMSE), Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and correlation coefficient (r); qualitative analysis included statistical features, seasonality, non-stationarity, and complexity. The quantitative analysis was calculated according to the following formula:
wherein ltestTo test the sample length, σ2The variance of the expected signal is shown as y (t), d (t), which are the actual output and expected output values of the network during the test phase.
In the qualitative analysis, the statistical characteristic difference refers to the mean value, standard deviation and histogram distribution of the comparison measured value and the predicted value; the seasonal characteristic difference is reflected by a surface grid map and a gray scale map of a two-dimensional matrix of the predicted value and the measured value; the non-stationarity difference is reflected by comparing the autocorrelation coefficients of the predicted value and the measured value; the complexity is analyzed by a visual graph networking method.
The invention has the advantages that: firstly, a multi-cluster ESN neural network is established, a photovoltaic power generation power prediction method based on the multi-cluster ESN neural network is provided, and a multi-cluster ESN prediction model with high training speed and high prediction precision is designed; secondly, the proposed multi-cluster ESN prediction model can obtain better performance through quantitative analysis and qualitative analysis evaluation.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings, in which:
FIG. 1 is an overall flow chart of a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network;
FIG. 2 is a flow chart of the generation of a multi-cluster structure;
FIG. 3 is a multi-cluster ESN neural network structure;
FIG. 4 is a diagram of different input-output models of the short-term prediction model 1 hour ahead;
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings; it should be understood that the preferred embodiments are illustrative of the invention only and are not limiting upon the scope of the invention.
FIG. 1 is an overall flow chart of a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network, which comprises three parts, namely, firstly establishing a multi-cluster ESN neural network, then establishing an ESN prediction model to analyze data, and finally evaluating the difference between a predicted result and a measured result through quantitative and qualitative analysis; FIG. 2 is a flow chart of a multi-cluster structure generation process; FIG. 3 is a multi-cluster ESN neural network structure, wherein the network comprises an input layer, a reservoir layer, and an output layer; FIG. 4 is a diagram of different input-output models of a short-term prediction model 1 hour ahead, a multi-cluster ESN neural network prediction model is established considering temperature and humidity measurement values and periodic characteristics of photovoltaic power generation power, Ppv(t-1)、Ppv(T-24) historical one-step and historical 24-step photovoltaic power generation power values, Tpv(t-1)、Hpv(t-1) corresponding historical one-step temperature and humidity measurements, Ppv(t) is the photovoltaic power generation power value to be predicted, as shown in the figure: the invention provides a photovoltaic power generation power prediction method based on a multi-cluster ESN neural network, which comprises the following steps:
s1: constructing a multi-cluster ESN neural network;
s2, initializing each parameter of the constructed multi-cluster ESN neural network, including nodes of an input layer, a hidden layer and an output layer
The number, input and output connection weights, node threshold functions and the like;
s3: considering influence factors, establishing an accurate input-output prediction model;
s4: calling sample information, and training the initialized multi-cluster ESN neural network;
s5: calling a test sample to test the network;
s6: analyzing and evaluating the output result obtained in the test stage;
the multi-cluster ESN neural network model is constructed by the following steps:
s11: initializing n precursor nodes in a two-dimensional plane, and fully connecting the nodes;
s12: adding local area nodes to the two-dimensional plane;
s13: connecting the newly added local node with the existing node according to the local priority connection rule;
s14: repeating the steps S12 and S13 until the total number of the nodes reaches the set size N of the reserve pool;
s15: generating a reserve pool connection weight matrix, and calculating to obtain a state update equation of the reserve pool and the output of the network:
x(t+1)＝f(Winu(t+1)+Wresx(t)+Wbacky(t)+v(t))
the scales of the multiple ESN network input units u (t), the pool unit x (t) and the output unit y (t) are K, N, L respectively. Win、Wres、Wout、WbackRespectively representing an NxK input connection weight matrix, an NxN internal connection weight matrix, an LxN output connection weight matrix and an NxL feedback connection weight matrix. f denotes the excitation function (usually Sigmoid function) of the internal unit, and v (t) is the noise signal.
The output calculation of the network in S15 adopts the following function:
y(t+1)＝fout(Woutx(t+1))
wherein f isoutWhich is the activity function (usually a linear function or Sigmoid function) of the output unit, a linear output unit is used here.
The photovoltaic power generation prediction model based on the multi-cluster ESN neural network comprises 1-hour advanced short-term prediction and 24-hour advanced middle-term prediction;
the single-step MCESN prediction model includes four different input-output short-term prediction models based on temperature and humidity measurements and two different input-output short-term prediction models based on cycle characteristics, as shown in fig. 4;
the multi-step MCESN prediction model is a multi-input multi-output model.
The expression of the multi-step MCESN multi-input multi-output model is as follows:
wherein, the input signals are respectively the power average value of 24 hours of the current day (k)
Corresponding average value of temperature
And days k per month, the output is the power value for the next 24 hours. The left side of the equal sign represents the photovoltaic power generation power value to be predicted which is ahead of one day (24 hours), namely the output of the network; the right side of the equation represents the input signal of the network;
an input-output nonlinear function fitted to the neural network.
All power, temperature history data are expressed in hours as a two-dimensional (2-D) matrix:
wherein, the row number m and the column number s respectively represent the total days and the hours of each day.
The assessment of the predictive performance includes quantitative and qualitative analysis;
quantitative analysis is to quantify the error between the predicted value and the measured value;
the qualitative analysis is to analyze the internal characteristic difference between the measured value and the predicted value;
the quantitative analysis mainly evaluates the error magnitude of the prediction result from four indexes, namely standard square root error (NRMSE), Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and correlation coefficient (r); qualitative analysis included statistical features, seasonality, non-stationarity, and complexity. The quantitative analysis was calculated according to the following formula:
wherein ltestTo test the sample length, σ2The variance of the expected signal is shown as y (t), d (t), which are the actual output and expected output values of the network during the test phase.
In the qualitative analysis, the statistical characteristic difference refers to the mean value, standard deviation and histogram distribution of the comparison measured value and the predicted value; the seasonal characteristic difference is reflected by a surface grid map and a gray scale map of a two-dimensional matrix of the predicted value and the measured value; the non-stationarity difference is reflected by comparing the autocorrelation coefficients of the predicted value and the measured value; the complexity is analyzed by a visual graph networking method. By performing the above qualitative analysis on the measured value and the predicted value, the difference of the internal characteristics between the measured value and the predicted value can be explored, so that the difference between the predicted value and the measured value can be further reflected. By combining quantitative analysis and qualitative analysis, the error between the actual measurement value and the predicted value can be definitely known, and the difference of the internal characteristics between the actual measurement value and the predicted value can be obtained through various graphs.
Claims (1)
1. A photovoltaic power generation power prediction method based on a multi-cluster ESN neural network is characterized by comprising the following steps: the method comprises the following steps:
s1: constructing a multi-cluster ESN neural network, wherein the multi-cluster ESN neural network is generated by the following steps:
s11: initializing n precursor nodes in a two-dimensional plane, and fully connecting the nodes;
s12: adding local area nodes to the two-dimensional plane;
s13: connecting the newly added local node with the existing node according to the local priority connection rule;
s14: repeating the steps S12 and S13 until the total number of the nodes reaches the set size N of the reserve pool;
s15: generating a reserve pool connection weight matrix, and calculating to obtain a state updating equation of the reserve pool:
x(t+1)＝f(Winu(t+1)+Wresx(t)+Wbacky(t)+v(t))
wherein, the scales of the multi-cluster ESN network input unit u (t), the reserve pool unit x (t) and the output unit y (t) are K, N, L respectively; win、Wres、Wout、WbackRespectively representing an NxK input connection weight matrix, an NxN internal connection weight matrix, an LxN output connection weight matrix and an NxL feedback connection weight matrix, wherein f represents an excitation function of an internal unit and is a Sigmoid function, and v (t) represents a noise signal; the output calculation of the multi-cluster ESN neural network adopts the following functions:
y(t+1)＝fout(Woutx(t+1))
wherein f isoutLinear output units are used here as a function of the activity of the output unit;
s2, initializing each parameter of the constructed multi-cluster ESN neural network;
s3: considering influence factors, establishing a single-step MCESN prediction model and a multi-step MCESN prediction model, wherein the single-step MCESN prediction model is a short-term prediction in advance for 1 hour and comprises four different input-output short-term prediction models based on temperature and humidity measurement values and two different input-output short-term prediction models based on periodic characteristics, and the multi-step MCESN prediction model is a multi-input multi-output model in advance for a middle-term prediction for 24 hours; the multi-step MCESN prediction model is described by an expression:
wherein, the input signals are respectively the power average value of 24 hours of the current day
Corresponding average value of temperature
And the number of days per month k, the output being the power value P from the first hour to the last hour of the following day1(k+1),P2(k+1),…,P24(k +1), the left side of the equal sign represents the photovoltaic power generation power value to be predicted which is one day ahead, namely the network output; the right side of the equation represents the input signal of the network;
an input-output nonlinear function fitted to the neural network; in order to solve the average power and the corresponding temperature average value of the multi-input multi-output model, all power and temperature historical data are expressed as a two-dimensional matrix according to hours:
wherein, the number of rows m and the number of columns s respectively represent the total days and the hours of each day;
s4: calling sample information, and training the initialized multi-cluster ESN neural network;
s5: calling a test sample to test the network;
s6: for the output result obtained in the testing stage, evaluating the prediction performance, wherein the evaluation comprises quantitative and qualitative analysis; the quantitative analysis is used for quantifying the error between a predicted value and a measured value, and the qualitative analysis is used for analyzing the internal characteristic difference between the measured value and the predicted value; the quantitative analysis mainly evaluates the error size of a prediction result from four indexes, namely standard square root error NRMSE, average absolute error MAE, root mean square error RMSE and correlation coefficient r; the qualitative analysis includes statistical features, seasonality, non-stationarity, and complexity; calculated according to the following formula:
wherein ltestTo test the sample length, σ2The variance of the expected signal is shown as y (t), d (t) and the actual output and the expected output value of the network in the testing stage respectively;
the statistical characteristic difference refers to the mean value, standard deviation and histogram distribution of the comparison measured value and the predicted value;
the seasonal characteristic difference is reflected by a surface grid map and a gray scale map of a two-dimensional matrix of predicted values and measured values;
the non-stationarity difference is reflected by comparing the autocorrelation coefficients of the predicted value and the measured value;
the complexity is analyzed by a visual graph networking method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810352268.XA CN108710966B (en) | 2018-04-19 | 2018-04-19 | Photovoltaic power generation power prediction method based on multi-cluster ESN neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810352268.XA CN108710966B (en) | 2018-04-19 | 2018-04-19 | Photovoltaic power generation power prediction method based on multi-cluster ESN neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108710966A CN108710966A (en) | 2018-10-26 |
| CN108710966B true CN108710966B (en) | 2021-10-22 |
Family
ID=63867276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810352268.XA Active CN108710966B (en) | 2018-04-19 | 2018-04-19 | Photovoltaic power generation power prediction method based on multi-cluster ESN neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108710966B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978238B (en) * | 2019-03-05 | 2022-05-03 | 电子科技大学 | Destination prediction method based on deep echo state network |
| CN112364527B (en) * | 2020-12-02 | 2021-09-07 | 岳文琦 | Debutanizer soft measurement modeling method based on ALIESN online learning algorithm |
| CN117674143B (en) * | 2024-02-01 | 2024-04-12 | 西华大学 | Short-term photovoltaic power prediction method and device based on echo pulse nerve P system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766986A (en) * | 2017-10-31 | 2018-03-06 | 天津大学 | Leak integral form echo state network on-line study photovoltaic power Forecasting Methodology |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11238337B2 (en) * | 2016-08-22 | 2022-02-01 | Applied Brain Research Inc. | Methods and systems for implementing dynamic neural networks |
-
2018
- 2018-04-19 CN CN201810352268.XA patent/CN108710966B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766986A (en) * | 2017-10-31 | 2018-03-06 | 天津大学 | Leak integral form echo state network on-line study photovoltaic power Forecasting Methodology |
Non-Patent Citations (2)
| Title |
|---|
| 基于改进回声状态网的时间序列预测;林健等;《渤海大学学报(自然科学版)》;20150915(第03期);全文 * |
| 用于混沌时间序列预测的多簇回响状态网络;宋青松等;《物理学报》;20090715(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108710966A (en) | 2018-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tabone et al. | Modeling variability and uncertainty of photovoltaic generation: A hidden state spatial statistical approach | |
| CN107194495B (en) | Photovoltaic power longitudinal prediction method based on historical data mining | |
| CN112100911B (en) | Solar radiation prediction method based on depth BILSTM | |
| CN103996084B (en) | Wind power probability Forecasting Methodology based on longitudinal moment Markov chain model | |
| CN108710966B (en) | Photovoltaic power generation power prediction method based on multi-cluster ESN neural network | |
| CN113554466A (en) | Short-term power consumption prediction model construction method, prediction method and device | |
| Yin et al. | Forecasting method of monthly wind power generation based on climate model and long short-term memory neural network | |
| CN106446440A (en) | Short-term photovoltaic generation power prediction method based on online sequential extreme learning machine | |
| CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
| CN113516271A (en) | Wind power cluster power day-ahead prediction method based on space-time neural network | |
| CN116029419A (en) | Deep learning-based long-term new energy daily average generation power prediction method and system | |
| CN116451821A (en) | Deep learning-based large-area new energy generation power prediction method and system | |
| CN116014722A (en) | Sub-solar photovoltaic power generation prediction method and system based on seasonal decomposition and convolution network | |
| CN117114438A (en) | Building area energy system cold and hot load data driving prediction method with flexibility and interpretability | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| Ibrahim et al. | LSTM neural network model for ultra-short-term distribution zone substation peak demand prediction | |
| CN110659775A (en) | LSTM-based improved electric power short-time load prediction algorithm | |
| CN112581311B (en) | Method and system for predicting long-term output fluctuation characteristics of aggregated multiple wind power plants | |
| CN113722951B (en) | Scatterer three-dimensional finite element grid optimization method based on neural network | |
| Hua | Power quality prediction of active distribution network based on CNN-LSTM deep learning model | |
| Obeidi et al. | Sunspot number-based neural network model for global solar radiation estimation in Ghardaïa | |
| CN114510872A (en) | Cloud server aging prediction method based on self-attention mechanism DLSTM | |
| Cho et al. | Application of Parallel ANN-PSO to Hourly Solar PV Estimation | |
| CN113962432A (en) | Wind power prediction method and system integrating three-dimensional convolution and light-weight convolution threshold unit | |
| Tian et al. | Fusion modeling for wind power forecasting based on redundant method elimination |
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 |