CN111126645A - Wind power prediction algorithm based on data mining technology and improved support vector machine - Google Patents

Wind power prediction algorithm based on data mining technology and improved support vector machine Download PDF

Info

Publication number
CN111126645A
CN111126645A CN201811283481.6A CN201811283481A CN111126645A CN 111126645 A CN111126645 A CN 111126645A CN 201811283481 A CN201811283481 A CN 201811283481A CN 111126645 A CN111126645 A CN 111126645A
Authority
CN
China
Prior art keywords
support vector
vector machine
data
wind power
power 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.)
Pending
Application number
CN201811283481.6A
Other languages
Chinese (zh)
Inventor
周文博
高婧
李存斌
孙辰军
周景
蔺帅帅
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 Hebei Electric Power Co Ltd
North China Electric Power University
Original Assignee
State Grid Corp of China SGCC
State Grid Hebei Electric Power Co Ltd
North China Electric Power University
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 Hebei Electric Power Co Ltd, North China Electric Power University filed Critical State Grid Corp of China SGCC
Priority to CN201811283481.6A priority Critical patent/CN111126645A/en
Publication of CN111126645A publication Critical patent/CN111126645A/en
Pending legal-status Critical Current

Links

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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Evolutionary Computation (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Operations Research (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a wind power prediction algorithm based on a data mining technology and an improved support vector machine, which comprises the following steps: step 1: processing data; processing the error and missing original data, and performing denoising analysis on the processed original data by adopting wavelet transformation to obtain a group of new data with higher credibility; step 2: initializing parameters; determining a kernel parameter sigma of a support vector machine algorithm2And a penalty factor C; and step 3: optimizing parameters; kernel parameter sigma of support vector machine algorithm through cuckoo search algorithm2Carrying out iterative optimization on the penalty factor C; and 4, step 4: constructing a support vector machine, namely obtaining optimal parameters through the cuckoo search algorithm in the step 3, and constructing the support vector machine; and 5: support vector machine algorithm based on the stepsCarrying out wind power prediction by measuring machine training; step 6: and outputting a wind power generation prediction result.

Description

Wind power prediction algorithm based on data mining technology and improved support vector machine
Technical Field
The invention relates to the technical field of power system application management, in particular to a wind power prediction algorithm based on a data mining technology and an improved support vector machine.
Background
Currently, most scholars focus on two research methods, namely statistical model research and physical model research, aiming at the research of wind power generation prediction. Although both methods can predict the wind power generation situation, the two methods are substantially different. The former method is a method of performing an intensive study on historical data of wind power generation, predicting the operation condition of a fan at the next stage according to the historical data, and obtaining a wind power generation value, such as AR, MA, ARMA and the like. The physical model mainly takes meteorological information data and the like as independent variables to construct a wind power generation curve, so that the value of the next stage is simulated and predicted. However, with the development of information technology, a supervisory control and data acquisition System (SCADA) has been widely used for monitoring wind turbines and acquiring various data sources for weather, power generation, and the like. Because of the enormous and complex data, traditional prediction methods have not been suitable and need to be modified to handle large data. Therefore, new intelligent algorithms such as a neural network, a particle swarm optimization algorithm, an SVM and the like are rapidly developed, and the SVM algorithm can effectively solve the problems of classification and complex nonlinear programming and is widely applied to the aspects of wind speed prediction, power generation amount prediction and the like. How to optimize the nuclear parameters and the penalty factors is a key problem in research of SVM algorithm, and parameters obtained by different methods can have different influences on a prediction result.
A large number of data errors and omissions will occur due to instability of the information system and sensor defects. Therefore, how to extract correct information from a large amount of data and how to process unreasonable data by using valuable information still remains a problem to be solved.
Therefore, a data mining technology and a wind power generation prediction algorithm of the DM-WT-CS-SVM are expected to solve the problem that the SVM algorithm in the prior art carries out reasonable parameter optimization on nuclear parameters and penalty factors.
Disclosure of Invention
The invention aims to provide a wind power generation prediction algorithm based on a data mining technology and a DM-WT-CS-SVM, which corrects original data based on a DM technology, adopts a WT method to reduce noise as an effective signal processing method, finally utilizes an SVM to predict wind power, and utilizes a CS algorithm to optimize parameters.
The invention discloses a wind power generation prediction algorithm based on a data mining technology and a DM-WT-CS-SVM, which comprises the following steps:
step 1: processing data; collecting original data, processing the error and missing original data, and then performing denoising analysis on the processed original data by adopting wavelet transformation to obtain a group of new data with higher credibility, wherein the formula of the wavelet transformation is as follows:
Figure RE-GDA0001943839980000021
sets of frequency signals, Detailsj＝Approxj-approxj-1The method comprises the steps of representing periodic variation information of an original signal, wherein DWT is discrete wavelet transform, m is a scale factor, N is 1,2 and …, N is sampling time, N is total amount of samples, and psi (x) is a wavelet function;
step 2: initializing parameters; determining support vector machineNuclear parameter of method sigma2And a penalty factor C;
and step 3: optimizing parameters; kernel parameter sigma of support vector machine algorithm through cuckoo search algorithm2And carrying out iterative optimization on the penalty factor C, wherein the fitness function is as follows:
Figure RE-GDA0001943839980000022
where n sample volumes of training set, yiIs the output of the training set and is,
Figure RE-GDA0001943839980000023
is the corresponding optimization result;
and 4, step 4: constructing a support vector machine, namely obtaining optimal parameters through the cuckoo search algorithm in the step 3, and constructing the support vector machine;
and 5: carrying out wind power prediction on the training of the support vector machine according to the steps of the support vector machine algorithm;
step 6: and outputting a wind power generation prediction result.
Preferably, the step 3 parameter optimization specifically includes the following steps:
step 3.1: selecting a kernel parameter sigma in a random solution according to the size of the fitness function2And taking the better solution of the penalty factor C as an initial solution;
step 3.2: updating the positions of cuckoos by iteration;
step 3.3: according to the discovery probability paDiscard partial solutions, pa∈[0,1]；
Step 3.4: generating the same number of new solutions as the number of discarded solutions of step 3.3;
step 3.5: calculating fitness functions of the initial solution and the new solution, and selecting the solution with the smaller fitness function as the initial solution of the next iteration;
step 3.6: if the iteration times reach the maximum iteration times, returning to the parameter values and executing the step 4, and if the iteration times do not reach the maximum iteration times, returning to the step 3.1 to start the next iteration.
Preferably, the iterative formula in step 3.2 is as follows:
Figure RE-GDA0001943839980000031
in the formula
Figure RE-GDA0001943839980000032
Denotes the ith solution of the t generation, α0Is a constant, usually takes on the value 0.01, xbestRepresents the optimal solution under the current condition, mu and nu obey the standard normal distribution,
Figure RE-GDA0001943839980000033
lambda is 1.5
Preferably, the step 3.4 generates a new solution by a cuckoo search algorithm in a cross operation and mixed variation manner, and the formula is as follows:
Figure RE-GDA0001943839980000034
where r is a scaling factor, in the interval 0,1]The random numbers are uniformly distributed on the random number,
Figure RE-GDA0001943839980000035
and
Figure RE-GDA0001943839980000036
two random solutions during the t-th iteration are shown.
Preferably, the maximum number of iterations is 400, and when the number of iterations is 400, the optimization process is terminated.
Preferably, the raw data comprises: wind speed, wind direction, and historical power generation data.
Preferably, the step 1 is to process the error original data by adopting a curve fitting method in a data mining algorithm; and filling the missing original data by adopting an average value method. The wind power generation prediction algorithm based on the data mining technology and the DM-WT-CS-SVM, disclosed by the invention, adopts the data mining technology to mine the relation between the wind speed and the wind power output, and unreasonable original data are modified. The wind power generation prediction algorithm is based on a Wavelet Transform (WT) method, and can eliminate the high frequency part of the original signal. Secondly, in order to improve the accuracy of the prediction result, parameters of a kernel function and penalty factors of a Support Vector Machine (SVM) are simulated, and the simulation capability of the SVM is improved by adopting a Cuckoo Search (CS) algorithm. The invention not only can be used for the dispatching department to optimize the output of the conventional unit and arrange for standby according to the output curve of the wind power plant, but also can provide necessary scientific basis for the power generation plan of the power grid for making days.
Drawings
FIG. 1 is a schematic flow diagram of a wind power generation prediction algorithm of the present invention.
Detailed Description
In order to make the implementation objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are only some, but not all embodiments of the invention. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1: a wind power prediction method based on data mining and SVM algorithm improvement mainly comprises the following steps:
1. receipt collection
Firstly, a fan is selected as an example, meteorological data (including wind speed and wind direction) and electric energy output data are collected, and the time interval is 1 hour. And processing the data errors and data missing in the data.
2. Data processing
Data processing plays an important role in wind power prediction. The DM can obtain the most valuable information from the diversity of capacity, speed and data. Also, the WT method is a useful signal transformation tool. In the report, the DM technology is used for researching the relation between the wind speed and the energy output, error data is processed, and a WT method is used for reducing the noise of a sample.
(1) Data errors. The data error of the original data set can be divided into two types, namely that the wind speed, the direction or the energy output value is negative, and that the energy output is 0 when the wind speed value is not 0. Therefore, it is necessary to correct these data, and DM is a suitable method. Since wind speed has the greatest effect on energy output, curve fitting is used herein to model the relationship between these two variables. Common methods including gaussian fitting, fourier and polynomial fitting (including first, second and third order polynomials) are compared based on the adjusted r-squares and SSE. The result shows that the Fourier fitting method is the optimal method with the maximum square of the adjusting coefficient r and the minimum SSE. Therefore, the present invention will employ Fourier fitting to fit curves of wind speed and power output. From the fit results, the rules for processing the error data are as follows: a) when the wind speed is less than 4m/s, the wind power is 0; b) when the wind speed is more than 4m/s, the output value is calculated by using a Fourier fitting function, and the formula is as follows.
f(x)＝7.187-5.542cos(0.217x)-4.413sin(0.217x)
In the formula: x is the wind speed value, f (x) is the wind energy output
(2) Data is lost. There are many samples in the original dataset that are filled with "-", which means that this value exceeds the statistic due to equipment failure. For these samples, two solutions need to be considered, a) when the wind speed value is not "-", the Fourier fitting equation is also applicable; b) if both wind speed and output power are "-", the gap is filled with the average of the last and second day.
(3) And (5) performing wavelet transformation. To improve the efficiency of prediction, WT analysis is applied herein to noise reduction of the data set. By this method, the raw data of the DM processing is decomposed into an approximate component a1 and a detail component D1. The original signal has a high similarity to the approximation component a 1. A1 is a low frequency part, which changes slowly. It is also the frame and outline of the entire signal, reflecting most of the information. D1 is a high frequency part and a noise part, and changes rapidly. It reflects the details of the signal, a small fraction of all information. In order to eliminate irregular signals in the original signal, noise parts are excluded herein. Therefore, simulation was performed by taking a1 as the input signal.
3. And (4) selecting input variables. Wind output is affected by various factors such as wind speed, temperature, and wind direction, but is mainly affected by wind conditions. Thus, the present invention only takes wind speed and wind direction as input variables.
SVM prediction. A vector sample with n dimensions can be represented as: (x)1,y1),(x2,y2),…,(xn,yn) Can pass through an optimal decision function
Figure RE-GDA0001943839980000051
Mapping non-linearities
Figure RE-GDA0001943839980000052
Into a linear function in a higher dimensional space. ω and b can be found by the structural risk minimization principle (SRM), and the formula is as follows.
Figure RE-GDA0001943839980000053
Solving the above equation by Lagrange function can be converted to solving a linear problem, as follows.
Figure RE-GDA0001943839980000061
α and b can be found by the above equation and a regression function is obtained:
Figure RE-GDA0001943839980000062
K(x,xi) As kernel functions, in the invention
Figure RE-GDA0001943839980000063
As a kernel function.
At this time, in order to optimize the penalty factor in the decision functionC and kernel parameter σ in kernel function2In the present invention, a cuckoo search algorithm is used.
5. And calculating a result. After punishment factors and kernel parameters can be obtained through cuckoo search, a regression function of the SVM can be obtained, a plurality of data are selected as training samples, a trained SVM model can be obtained, at the moment, values of the testing samples are brought into the trained SVM model, prediction data can be obtained, a prediction value is compared with an actual value and is compared and analyzed with other algorithms, and the fact that a DM (demodulation) technology, a WT (WT) method and a CS (circuit switched) algorithm are introduced is obtained, so that the accuracy of a prediction result can be remarkably improved, and the probability of extreme errors can be reduced.
Finally, it should be pointed out that: the above examples are only for illustrating the technical solutions of the present invention, and are not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (7)

1. A wind power prediction algorithm based on a data mining technology and an improved support vector machine is characterized by comprising the following steps:
step 1: processing data; collecting original data, processing the error and missing original data, and then performing denoising analysis on the processed original data by adopting wavelet transformation to obtain a group of new data with higher credibility, wherein the formula of the wavelet transformation is as follows:
Figure FDA0001848531920000011
Figure FDA0001848531920000012
wherein ApproxJ,tFor the low scale of the signal, trends of the original signal, Details, can be capturedjBeing a collection of high-frequency signals, Detailsj＝Approxj-approxj-1The method comprises the steps of representing periodic variation information of an original signal, wherein DWT is discrete wavelet transform, m is a scale factor, N is 1,2 and …, N is sampling time, N is total amount of samples, and psi (x) is a wavelet function;
step 2: initializing parameters; determining a kernel parameter sigma of a support vector machine algorithm2And a penalty factor C;
and step 3: optimizing parameters; kernel parameter sigma of support vector machine algorithm through cuckoo search algorithm2And carrying out iterative optimization on the penalty factor C, wherein the fitness function is as follows:
Figure FDA0001848531920000013
where n sample volumes of training set, yiIs the output of the training set and is,
Figure FDA0001848531920000014
is the corresponding optimization result;
and 4, step 4: constructing a support vector machine, namely obtaining optimal parameters through the cuckoo search algorithm in the step 3, and constructing the support vector machine;
and 5: carrying out wind power prediction on the training of the support vector machine according to the steps of the support vector machine algorithm;
step 6: and outputting a wind power generation prediction result.
2. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 1, characterized in that: the step 3 of parameter optimization specifically comprises the following steps:
step 3.1: selecting a kernel parameter sigma in a random solution according to the size of the fitness function2And taking the better solution of the penalty factor C as an initial solution;
step 3.2: updating the positions of cuckoos by iteration;
step 3.3:according to the discovery probability paDiscard partial solutions, pa∈[0,1]；
Step 3.4: generating the same number of new solutions as the number of discarded solutions of step 3.3;
step 3.5: calculating fitness functions of the initial solution and the new solution, and selecting the solution with the smaller fitness function as the initial solution of the next iteration;
step 3.6: if the iteration times reach the maximum iteration times, returning to the parameter values and executing the step 4, and if the iteration times do not reach the maximum iteration times, returning to the step 3.1 to start the next iteration.
3. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 2, characterized in that: the iterative formula in step 3.2 is as follows:
Figure FDA0001848531920000021
in the formula
Figure FDA0001848531920000022
Denotes the ith solution of the t generation, α0Is a constant, usually takes on the value 0.01, xbestRepresents the optimal solution under the current condition, mu and nu obey the standard normal distribution,
Figure FDA0001848531920000023
lambda is 1.5.
4. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 2, characterized in that: step 3.4, generating a new solution by adopting a cross operation and mixed variation mode through a cuckoo search algorithm, wherein the formula is as follows:
Figure FDA0001848531920000024
where r is a scaling factor, in the interval 0,1]Upper uniformityThe random number of the distribution is determined,
Figure FDA0001848531920000025
and
Figure FDA0001848531920000026
two random solutions during the t-th iteration are shown.
5. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 2, characterized in that: the maximum iteration number is 400, and when the iteration number is 400
The optimization process terminates.
6. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 1, characterized in that: the raw data includes: wind speed, wind direction, and historical power generation data.
7. The wind power prediction algorithm based on data mining technology and improved support vector machine of claim 1, characterized in that: the step 1 is to process the error original data by adopting a curve fitting method in a data mining algorithm; and filling the missing original data by adopting an average value method.
CN201811283481.6A 2018-10-31 2018-10-31 Wind power prediction algorithm based on data mining technology and improved support vector machine Pending CN111126645A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811283481.6A CN111126645A (en) 2018-10-31 2018-10-31 Wind power prediction algorithm based on data mining technology and improved support vector machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811283481.6A CN111126645A (en) 2018-10-31 2018-10-31 Wind power prediction algorithm based on data mining technology and improved support vector machine

Publications (1)

Publication Number Publication Date
CN111126645A true CN111126645A (en) 2020-05-08

Family

ID=70485186

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811283481.6A Pending CN111126645A (en) 2018-10-31 2018-10-31 Wind power prediction algorithm based on data mining technology and improved support vector machine

Country Status (1)

Country Link
CN (1) CN111126645A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112015784A (en) * 2020-09-07 2020-12-01 华北电力大学（保定） Wind condition data mining method and device, wind measuring device and data mining equipment
CN112765544A (en) * 2020-12-30 2021-05-07 国网湖南省电力有限公司 Strong wind power correction and forecast method under influence of typhoon of power transmission line
CN114154684A (en) * 2021-11-15 2022-03-08 国家电网有限公司 Short-term photovoltaic power prediction method based on data mining and multi-core support vector machine

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112015784A (en) * 2020-09-07 2020-12-01 华北电力大学（保定） Wind condition data mining method and device, wind measuring device and data mining equipment
CN112015784B (en) * 2020-09-07 2024-02-13 华北电力大学（保定） Wind condition data mining method and device, wind measuring device and data mining equipment
CN112765544A (en) * 2020-12-30 2021-05-07 国网湖南省电力有限公司 Strong wind power correction and forecast method under influence of typhoon of power transmission line
CN112765544B (en) * 2020-12-30 2023-09-15 国网湖南省电力有限公司 Method for correcting and forecasting strong wind power under influence of typhoon of power transmission line
CN114154684A (en) * 2021-11-15 2022-03-08 国家电网有限公司 Short-term photovoltaic power prediction method based on data mining and multi-core support vector machine

Similar Documents

Publication Publication Date Title
CN114004139B (en) Photovoltaic power generation power prediction method
CN109711609B (en) Photovoltaic power station output power prediction method based on wavelet transformation and extreme learning machine
CN111126645A (en) Wind power prediction algorithm based on data mining technology and improved support vector machine
CN111369057A (en) Air quality prediction optimization method and system based on deep learning
CN112668611B (en) Kmeans and CEEMD-PE-LSTM-based short-term photovoltaic power generation power prediction method
CN114792156A (en) Photovoltaic output power prediction method and system based on curve characteristic index clustering
CN113627685B (en) Wind driven generator power prediction method considering wind power internet load limit
CN111008726A (en) Class image conversion method in power load prediction
CN110555548A (en) ICS-ELM ultra-short-term wind power prediction method based on data mining original error correction
CN117218425A (en) Power generation loss analysis method and system for photovoltaic power station
CN115034485A (en) Wind power interval prediction method and device based on data space
CN116151464A (en) Photovoltaic power generation power prediction method, system and storable medium
CN115081597A (en) Method, system, equipment and medium for performing carbon check by using electricity based on machine learning
CN117054803A (en) Method and system for identifying grounding faults of distribution network containing distributed photovoltaic
CN114118401A (en) Neural network-based power distribution network flow prediction method, system, device and storage medium
CN117833216A (en) Photovoltaic power station generated power prediction method and device based on hybrid neural network
CN118095891A (en) Active power distribution network payload prediction method and system considering source load meteorological characteristic decoupling
CN111950752A (en) Photovoltaic power station generating capacity prediction method, device and system and storage medium thereof
CN115983507B (en) Method and system for predicting broadband oscillation risk of section of power grid of transmitting end source
CN114281846B (en) New energy power generation prediction method based on machine learning
CN113779861B (en) Photovoltaic Power Prediction Method and Terminal Equipment
CN115514315A (en) Distributed photovoltaic power station PR prediction method based on optimized GRU model
CN114034966A (en) Power transmission line fault identification method and device based on support vector machine
Fang et al. Research on Short-Term and Medium-Term Power Load Forecasting Based on STL-LightGBM
CN118260541B (en) Power grid measurement data checking method and system based on artificial intelligence algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20200508