CN113570120A - Wind power prediction method and device based on improved particle swarm optimization - Google Patents

Wind power prediction method and device based on improved particle swarm optimization Download PDF

Info

Publication number
CN113570120A
CN113570120A CN202110767036.2A CN202110767036A CN113570120A CN 113570120 A CN113570120 A CN 113570120A CN 202110767036 A CN202110767036 A CN 202110767036A CN 113570120 A CN113570120 A CN 113570120A
Authority
CN
China
Prior art keywords
wind power
particle
particle swarm
power prediction
normalization
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
CN202110767036.2A
Other languages
Chinese (zh)
Inventor
邹定江
刘天羽
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.)
Shanghai Dianji University
Original Assignee
Shanghai Dianji 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 Shanghai Dianji University filed Critical Shanghai Dianji University
Priority to CN202110767036.2A priority Critical patent/CN113570120A/en
Publication of CN113570120A publication Critical patent/CN113570120A/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
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Abstract

The invention relates to a wind power prediction method and a wind power prediction device based on an improved particle swarm algorithm, wherein the method comprises the following steps of: s1: acquiring historical wind power data, and preprocessing the historical wind power data; s2: dividing historical wind power data into a training set and a test set, and constructing an initial wind power prediction model based on a BP neural network based on the training set; s3: optimizing an initial wind power prediction model by adopting a particle swarm algorithm with an inertia weight value decreasing in a nonlinear mode based on a test set to obtain an optimal prediction model; s4: and predicting the wind power by using the optimal model to obtain a prediction result and performing inverse normalization on the data. Compared with the prior art, the method has the advantages of high model convergence speed and quick and accurate prediction result.

Description

Wind power prediction method and device based on improved particle swarm optimization
Technical Field
The invention relates to the field of wind power prediction, in particular to a wind power prediction method and device based on an improved particle swarm algorithm.
Background
At present, the prediction methods for wind power can be divided into physical methods, statistical methods and composite methods. The physical method comprises the steps of calculating the wind direction and the wind speed at the height position of a hub of the wind generating set according to parameters such as meteorological information, the landform and the geographic position of the wind power plant, and the like, then calculating the output power according to a standard wind speed power curve of the wind generating set, and further solving the wind power prediction result. The statistical method is the statistics of the power change inference of the wind power plant in the future according to the historical power data and meteorological data of the wind power plant. The composite method combines a physical method and a statistical method, and most of the existing power prediction systems adopt the composite method for prediction. When wind power is predicted, in the prior art, simple simulation is only carried out according to historical generated power data of a wind power plant, a plurality of factors influencing the output of the wind power plant are ignored, the accuracy of a prediction result is low easily, and when the historical wind power data of the wind power plant are analyzed, a large amount of data are lost and abnormal, so that the accuracy of the prediction result is further reduced.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a wind power prediction method and device based on an improved particle swarm algorithm.
The purpose of the invention can be realized by the following technical scheme:
a wind power prediction method based on an improved particle swarm algorithm comprises the following steps:
s1: acquiring historical wind power data, and preprocessing the historical wind power data;
s2: dividing historical wind power data into a training set and a test set, and constructing an initial wind power prediction model based on a BP neural network based on the training set;
s3: optimizing an initial wind power prediction model by adopting a particle swarm algorithm with an inertia weight value decreasing in a nonlinear mode based on a test set to obtain an optimal prediction model;
s4: and predicting the wind power by using the optimal model to obtain a prediction result and performing inverse normalization on the data. Preferably, the particle velocity formula of the particle swarm algorithm with the inertia weight value decreasing nonlinearly is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles.
Preferably, the formula of the nonlinear decrement weight is as follows:
Figure BDA0003152148720000021
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.
Preferably, the step of optimizing the initial wind power prediction model in step S3 includes:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
Preferably, the preprocessing mode in step S1 includes normalization, and the formula of the normalization is:
Figure BDA0003152148720000022
wherein, A is the original data set before normalization, B is the data set after normalization, Amax、AminRespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
A wind power prediction device based on an improved particle swarm algorithm comprises a preprocessing module, a pre-training module, a model optimization module and a prediction module,
the preprocessing module acquires historical wind power data and preprocesses the historical wind power data;
the pre-training module divides historical wind power data into a training set and a test set, and an initial wind power prediction model based on a BP neural network is constructed based on the training set;
the model optimization module adopts a particle swarm algorithm with the inertia weight value decreasing nonlinearly and optimizes the initial wind power prediction model based on a test set to obtain an optimal prediction model;
and the prediction module predicts the wind power by using the optimal model to obtain a prediction result and performs inverse normalization on the data.
Preferably, the particle velocity formula of the particle swarm algorithm with the inertia weight value decreasing nonlinearly is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles.
Preferably, the formula of the nonlinear decrement weight is as follows:
Figure BDA0003152148720000031
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.
Preferably, the step of optimizing the initial wind power prediction model by the model optimization module includes:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
Preferably, the preprocessing module performs preprocessing in a manner including normalization, where the normalization is represented by the following formula:
Figure BDA0003152148720000032
wherein, A is the original data set before normalization, B is the data set after normalization, Amax、AminRespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
Compared with the prior art, the method of the invention enables the inertia weight to be reduced in a nonlinear way on the basis of the inertia weight of the particle swarm particles, so that the overall searching capability of the particles at the initial stage of iteration is stronger, and the searching area is larger, thereby being not easy to miss the searching of the overall optimal solution; in the later iteration stage, the local searching capability of the particles is stronger, the searching range is smaller, the optimal solution can be searched more finely, the iteration times are reduced, and the convergence speed is higher. According to the invention, the BP neural network is optimized by improving the particle swarm algorithm, so that an optimal wind power prediction model can be obtained, the convergence speed is accelerated, and the prediction accuracy is improved. The method can quickly find out the global optimal solution, establish the optimal prediction model, effectively improve the prediction efficiency and the prediction accuracy of the wind power and have strong practicability.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. Note that the following description of the embodiments is merely a substantial example, and the present invention is not intended to be limited to the application or the use thereof, and is not limited to the following embodiments.
Examples
A wind power prediction method based on an improved particle swarm algorithm is shown in figure 1 and comprises the following steps:
s1: and acquiring historical wind power data, and preprocessing the historical wind power data.
And acquiring historical wind power data, wherein factors influencing the wind power comprise wind direction, wind speed, air density, surface roughness, air temperature and air pressure.
Preprocessing historical wind power data to obtain preprocessed historical wind power data; the preprocessing is mainly the normalization processing of data, and because the dimensions of the data are different, the data difference is large, and the data need to be dimensionless, the normalization is needed; the formula of the normalization process is:
Figure BDA0003152148720000041
wherein, A is the original data set before normalization, B is the data set after normalization, Amax、AminRespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
S2: dividing historical wind power data into a training set and a testing set, and constructing an initial wind power prediction model based on a BP neural network based on the training set.
S3: and optimizing the initial wind power prediction model by adopting a particle swarm algorithm with the inertia weight value decreasing in a nonlinear manner based on the test set to obtain an optimal prediction model.
The particle velocity formula of the particle swarm algorithm with the inertia weight value being decreased in a nonlinear way is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles. In the formula, w is usually a constant, and determines the convergence and exploration capacity of a particle swarm optimization algorithm, so that the current flight speed of the particle is determined, and the inertia weight is improved to be reduced in a nonlinear way:
Figure BDA0003152148720000051
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.Get wmax＝0.9，wmin＝0.4。
In this step, based on the particle velocity formula, the step of optimizing the initial wind power prediction model in step S3 includes:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
S4: and predicting the wind power by using the optimal model to obtain a prediction result and performing inverse normalization on the data.
Based on the same inventive concept, the embodiment of the invention also provides a wind power prediction device based on the improved particle swarm optimization, and as the principle of solving the problems of the devices is similar to the wind power prediction method based on the improved particle swarm optimization, the implementation of the method can be referred, and repeated expenditure is not repeated.
The wind power prediction device comprises a preprocessing module, a pre-training module, a model optimization module and a prediction module,
the preprocessing module acquires historical wind power data and preprocesses the historical wind power data.
The preprocessing mode in the preprocessing module comprises normalization processing, and the formula of the normalization processing is as follows:
Figure BDA0003152148720000052
wherein A is an original data set before normalization, B is a data set after normalization,Amax、Aminrespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
The pre-training module divides historical wind power data into a training set and a test set, and an initial wind power prediction model based on a BP neural network is constructed based on the training set;
and the model optimization module adopts a particle swarm algorithm with the inertia weight value decreasing in a nonlinear mode and optimizes the initial wind power prediction model based on the test set to obtain an optimal prediction model.
The particle velocity formula of the particle swarm algorithm with the inertia weight value being decreased in a nonlinear way is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles.
The formula of the nonlinear decrement weight is as follows:
Figure BDA0003152148720000061
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.
The step of optimizing the initial wind power prediction model by the model optimization module comprises the following steps:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
And the prediction module predicts the wind power by using the optimal model to obtain a prediction result and performs inverse normalization on the data.
The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments may be implemented in other various manners, and various omissions, substitutions, and changes may be made without departing from the technical spirit of the present invention.

Claims (10)

1. A wind power prediction method based on an improved particle swarm algorithm is characterized by comprising the following steps:
s1: acquiring historical wind power data, and preprocessing the historical wind power data;
s2: dividing historical wind power data into a training set and a test set, and constructing an initial wind power prediction model based on a BP neural network based on the training set;
s3: optimizing an initial wind power prediction model by adopting a particle swarm algorithm with an inertia weight value decreasing in a nonlinear mode based on a test set to obtain an optimal prediction model;
s4: and predicting the wind power by using the optimal model to obtain a prediction result and performing inverse normalization on the data.
2. The wind power prediction method based on the improved particle swarm algorithm according to claim 1, wherein the particle velocity formula of the particle swarm algorithm with the inertia weight value decreasing nonlinearly is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles.
3. The wind power prediction method based on the improved particle swarm optimization algorithm according to claim 2, wherein the formula of the nonlinear decrement weight is as follows:
Figure FDA0003152148710000011
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.
4. The improved particle swarm optimization-based wind power prediction method according to claim 1, wherein the step of optimizing the initial wind power prediction model in step S3 comprises:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
5. The wind power prediction method based on the improved particle swarm algorithm according to claim 1, wherein the preprocessing in step S1 includes normalization, and the formula of the normalization is:
Figure FDA0003152148710000021
wherein, A is the original data set before normalization, B is the data set after normalization, Amax、AminRespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
6. A wind power prediction device based on an improved particle swarm algorithm is characterized by comprising a preprocessing module, a pre-training module, a model optimization module and a prediction module,
the preprocessing module acquires historical wind power data and preprocesses the historical wind power data;
the pre-training module divides historical wind power data into a training set and a test set, and an initial wind power prediction model based on a BP neural network is constructed based on the training set;
the model optimization module adopts a particle swarm algorithm with the inertia weight value decreasing nonlinearly and optimizes the initial wind power prediction model based on a test set to obtain an optimal prediction model;
and the prediction module predicts the wind power by using the optimal model to obtain a prediction result and performs inverse normalization on the data.
7. The improved particle swarm optimization-based wind power prediction device according to claim 6, wherein the particle velocity formula of the particle swarm optimization with the non-linear decreasing inertia weight is as follows:
Vid(t+1)＝wVid(t)+c1r1[Pid(t)-Xid(t)]+c2r2[Gid(t)-Xid(t)]
wherein w is a non-linear decreasing weight, Vid(t +1) is the updated velocity of the ith particle, Vid(t) is the current velocity of the ith particle, c1、c2As an acceleration factor, r1、r2Is [0, 1 ]]Random number of inner, Pid(t) is the optimum position of the ith particle, Xid(t) is the current position of the ith particle, Gid(t) is the optimum position for all particles.
8. The improved particle swarm optimization-based wind power prediction device according to claim 7, wherein the formula of the nonlinear decreasing weight is as follows:
Figure FDA0003152148710000022
wherein, wminIs the lower limit of inertia weight, wmaxIs the upper limit of inertia weight, a is constant coefficient, and determines the decreasing path, t is the current iteration number, tmaxIs the maximum iteration number.
9. The improved particle swarm optimization-based wind power prediction device according to claim 6, wherein the step of optimizing the initial wind power prediction model by the model optimization module comprises:
the method comprises the following steps: initializing parameters including population particle number, initial position and speed of particles, maximum iteration number and minimum mean square error of BP neural network;
step two: calculating the fitness of each particle in the particle swarm, comparing the fitness of each particle with the fitness of the best position flown by the particle, and selecting the smaller one as the best position at the moment;
step three: and judging whether the current iteration times reach an iteration threshold or whether the minimum mean square error is equal to the minimum mean square error of the BP neural network, if so, finishing the optimization of the particles to obtain an optimal prediction model, otherwise, updating the speed and the position of the particles, and returning to the third step.
10. The improved particle swarm algorithm-based wind power prediction device according to claim 6, wherein the preprocessing module performs preprocessing in a normalization manner, and the normalization processing has a formula:
Figure FDA0003152148710000031
wherein, A is the original data set before normalization, B is the data set after normalization, Amax、AminRespectively, the maximum and minimum values in the data set before normalization, Bmax、BminThe maximum and minimum values of the normalized data set, respectively.
CN202110767036.2A 2021-07-07 2021-07-07 Wind power prediction method and device based on improved particle swarm optimization Pending CN113570120A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110767036.2A CN113570120A (en) 2021-07-07 2021-07-07 Wind power prediction method and device based on improved particle swarm optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110767036.2A CN113570120A (en) 2021-07-07 2021-07-07 Wind power prediction method and device based on improved particle swarm optimization

Publications (1)

Publication Number Publication Date
CN113570120A true CN113570120A (en) 2021-10-29

Family

ID=78163939

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110767036.2A Pending CN113570120A (en) 2021-07-07 2021-07-07 Wind power prediction method and device based on improved particle swarm optimization

Country Status (1)

Country Link
CN (1) CN113570120A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106126961A (en) * 2016-07-28 2016-11-16 上海交通大学 Lycoperdon polymorphum Vitt population satellite clock correction adaptive forecasting method and system
CN106920008A (en) * 2017-02-28 2017-07-04 山东大学 A kind of wind power forecasting method based on Modified particle swarm optimization BP neural network
CN108665322A (en) * 2018-05-18 2018-10-16 河南工业大学 The construction method of grain ration Consumption forecast model, Consumption forecast method and device
CN112686464A (en) * 2021-01-07 2021-04-20 云南电力技术有限责任公司 Short-term wind power prediction method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106126961A (en) * 2016-07-28 2016-11-16 上海交通大学 Lycoperdon polymorphum Vitt population satellite clock correction adaptive forecasting method and system
CN106920008A (en) * 2017-02-28 2017-07-04 山东大学 A kind of wind power forecasting method based on Modified particle swarm optimization BP neural network
CN108665322A (en) * 2018-05-18 2018-10-16 河南工业大学 The construction method of grain ration Consumption forecast model, Consumption forecast method and device
CN112686464A (en) * 2021-01-07 2021-04-20 云南电力技术有限责任公司 Short-term wind power prediction method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
邹定江，等: "改进麻雀算法优化Elman神经网络的短期电力负荷预测", 《上海电机学院学报》, vol. 25, no. 4, pages 221 - 227 *

Similar Documents

Publication Publication Date Title
Jiang et al. Network intrusion detection based on PSO-XGBoost model
US11208985B2 (en) Correction method and apparatus for predicted wind speed of wind farm
CN105488528B (en) Neural network image classification method based on improving expert inquiry method
CN112101430B (en) Anchor frame generation method for image target detection processing and lightweight target detection method
CN112581263A (en) Credit evaluation method for optimizing generalized regression neural network based on wolf algorithm
CN105224738B (en) LSSVM non-gaussian fluctuating wind speed prediction technique
CN110263834B (en) Method for detecting abnormal value of new energy power quality
CN114841257A (en) Small sample target detection method based on self-supervision contrast constraint
CN110784455B (en) Method for optimizing Xgboost model based on linear decreasing weight particle swarm algorithm
CN112800682A (en) Feedback optimization fan blade fault monitoring method
CN116503676B (en) Picture classification method and system based on knowledge distillation small sample increment learning
CN115358305A (en) Incremental learning robustness improving method based on iterative generation of boundary samples
CN114118537A (en) Combined prediction method for carbon emission of airspace flight
CN111598460A (en) Method, device and equipment for monitoring heavy metal content in soil and storage medium
CN114882373A (en) Multi-feature fusion sandstorm prediction method based on deep neural network
CN116485049B (en) Electric energy metering error prediction and optimization system based on artificial intelligence
CN113570120A (en) Wind power prediction method and device based on improved particle swarm optimization
CN112698666B (en) Aircraft route optimization method based on meteorological grid
CN116259057A (en) Method for solving data heterogeneity problem in federal learning based on alliance game
CN110991743B (en) Wind power short-term combination prediction method based on cluster analysis and neural network optimization
CN114781615A (en) Two-stage quantization implementation method and device based on compressed neural network
CN107274357A (en) A kind of optimal gray level image enhancing processing system of parameter
CN113294297A (en) Variable weight adjusting method for wind turbine generator nonlinear model prediction torque control
CN115061501B (en) Method and system for identifying coded data of air formation
CN111914923B (en) Target distributed identification method based on clustering feature extraction

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