CN116562650A - Short-term wind power prediction method and device and computer readable storage medium - Google Patents

Short-term wind power prediction method and device and computer readable storage medium Download PDF

Info

Publication number
CN116562650A
CN116562650A CN202310555581.4A CN202310555581A CN116562650A CN 116562650 A CN116562650 A CN 116562650A CN 202310555581 A CN202310555581 A CN 202310555581A CN 116562650 A CN116562650 A CN 116562650A
Authority
CN
China
Prior art keywords
short
wind power
lstm
power prediction
model
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
CN202310555581.4A
Other languages
Chinese (zh)
Inventor
彭丽
阳同光
李婉婷
杨京渝
黄银欢
陈雯静
陈颖倩
蒋锦峰
李泽星
曹京毅
费雅雯
黄锦毅
杨天峰
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.)
Hunan City University
Original Assignee
Hunan City 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 Hunan City University filed Critical Hunan City University
Priority to CN202310555581.4A priority Critical patent/CN116562650A/en
Publication of CN116562650A publication Critical patent/CN116562650A/en
Pending legal-status Critical Current

Links

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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
    • 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
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • 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
    • 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)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Evolutionary Computation (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Operations Research (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The application provides a short-term wind power prediction method, a short-term wind power prediction device and a computer storage medium. Wherein the method comprises the following steps: the terminal equipment collects historical data, wherein the historical data can comprise wind speed; performing modal decomposition on the historical data through a variable modal decomposition VMD to obtain P modal components, wherein P is a positive integer greater than or equal to 1; reconstructing the P modal components into N subsequences based on the approximate entropy, wherein N is less than or equal to P; performing parameter optimization on the LSTM neural network model of the long-term and short-term memory network based on an improved whale optimization algorithm WOA to obtain optimal LSTM parameters; constructing a short-term wind power prediction model based on the N subsequences after reconstruction and the optimal LSTM parameters; and inputting the prediction data into a short-term wind power prediction model to obtain a short-term wind power prediction result. Through the technical scheme provided by the application, the problems of wind power intermittence and fluctuation can be solved, and the accuracy of short-term wind power prediction is improved.

Description

Short-term wind power prediction method and device and computer readable storage medium
Technical Field
The application relates to the technical field of new energy power generation and smart power grids, in particular to a short-term wind power prediction method, a short-term wind power prediction device and a computer readable storage medium.
Background
In the background of huge global carbon dioxide emissions, wind power generation technology is rapidly developing. The strong fluctuation and intermittence of wind energy bring higher operation requirements for the normal operation of the power system, so that the operation risk caused by wind power generation grid connection is reduced, and the accurate prediction of wind power is an important guarantee for reasonably controlling and adjusting the grid connection technology.
In recent years, long-short term memory (LSTM) networks have been well applied in the field of short-term wind power prediction, and LSTM networks can fully mine the inherent correlation between time series data, but have low prediction accuracy when analyzing discontinuous data. Therefore, how to improve the accurate prediction of wind power is a problem to be solved at present.
Disclosure of Invention
The embodiment of the application provides a short-term wind power prediction method, a short-term wind power prediction device and a computer readable storage medium, which can improve the accurate prediction of wind power.
In a first aspect, the present application provides a short-term wind power prediction method, which may be applied to a computer device, a module (e.g., a chip or a processor) in the computer device, and a logic module or software that can implement all or part of the functions of the computer device. The following description will take an example in which the execution subject is a computer device. The method comprises the following steps: the method comprises the steps that a computer device collects historical data, wherein the historical data comprise wind speeds; performing modal decomposition on the historical data through a variational modal decomposition (variational mode decomposition, VMD) to obtain P modal components (intrinsic mode functions, IMF), wherein P is a positive integer greater than or equal to 1; reconstructing the P modal components into N subsequences based on the approximate entropy, wherein N is less than or equal to P; performing parameter optimization on a long-short-term memory network (long-short term memory, LSTM) neural network model based on an improved whale optimization algorithm (the whale optimization algorithm, WOA) to obtain optimal LSTM parameters; constructing a short-term wind power prediction model based on the N subsequences after reconstruction and the optimal LSTM parameters; and inputting the prediction data into a short-term wind power prediction model to obtain a short-term wind power prediction result.
The method has great application value in the field of wind power prediction by combining an optimization algorithm and a neural network, and the scheme provided by the application improves an LSTM model based on improved VMD and approximate entropy combined with improved WOA, and is applied to wind power prediction to obtain a short-term wind power prediction model. Because the whale optimization algorithm has the advantages of few parameters, simple structure, high prediction precision and the like, the improved whale optimization algorithm is utilized to perform optimization training on the parameters in the LSTM neural network, the optimal LSTM parameters can be automatically selected, the precision of the model is improved, and therefore the precision of short-term wind power prediction is improved.
In one possible implementation manner, the performing parameter optimization on the LSTM neural network model based on the improved WOA to obtain an optimal LSTM parameter includes: taking parameters to be optimized of the LSTM neural network model as an initialization solution of the improved WOA; and optimizing the parameters to be optimized of the LSTM neural network model by adopting the improved WOA to obtain the optimal LSTM parameters.
In a possible implementation manner, the optimizing the parameters to be optimized of the LSTM neural network model by using the improved WOA to obtain the optimal LSTM parameters includes: initializing parameters of the improved WOA; determining the current optimal value and the optimal solution of the improved WOA; and obtaining the optimal LSTM parameter according to the current optimal value and the optimal solution of the whale.
In one possible implementation manner, the performing, by the VMD, modal decomposition on the history data to obtain P modal components includes: the historical data is used as an input signal and is decomposed into modal components with different characteristics, and the estimated bandwidth and the minimum constraint condition of each modal component are the sum of all modalities; estimating component bandwidths by using the modal components and the input signals as constraint conditions and using Gaussian smoothing and gradient square norms to obtain a VMD variation model with constraint; converting the constrained VMD variational model into an unconstrained VMD variational model by using a quadratic penalty factor and a Lagrangian multiplier; and determining the P modal components according to the unconstrained VMD variational model and convergence criteria.
In one possible implementation, the reconstructing the P modal components into N subsequences based on approximate entropy includes: and for the P modal components after VMD decomposition, estimating the complexity of the P modal components by adopting approximate entropy to obtain N subsequences after reconstruction, wherein the N subsequences are the subsequences with typical characteristics.
In a possible implementation manner, the constructing a short-term wind power prediction model based on the reconstructed N subsequences and the optimal LSTM parameters includes: respectively taking the N subsequences after reconstruction as the input of an improved LSTM model, and training the improved LSTM model; and obtaining the short-term wind power prediction model based on the trained improved LSTM model and the optimal LSTM parameters.
In one possible implementation, the historical data further includes wind direction, yaw angle, and fan output power.
In a second aspect, embodiments of the present application provide a short-term wind power prediction apparatus. The short-term wind power prediction device can be applied to computer equipment, modules (such as chips or processors) in the computer equipment, and logic modules or software capable of realizing all or part of the functions of the computer equipment. The short-term wind power prediction device has a function of implementing the behavior in the method example of any one of the above-mentioned first aspects. The functions may be implemented by hardware, or may be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the functions described above. The advantages may be seen from the description of the first aspect, which is not repeated here.
In a third aspect, a short-term wind power prediction apparatus is provided, where the short-term wind power prediction apparatus may be a computer device in an embodiment of the method described above, or a chip or a processor provided in the computer device. The short-term wind power prediction device may include a processor, a memory, an input interface for receiving information from other communication devices than the short-term wind power prediction device, and an output interface for outputting information to other communication devices than the short-term wind power prediction device, the processor being coupled to the memory, the memory for storing programs or instructions that, when executed by the processor, cause the short-term wind power prediction device to perform the methods performed by the computer apparatus, or a chip or processor in the computer apparatus, in the above-described method embodiments.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored therein a computer program or computer instructions which, when run on a computer, cause the computer to perform the method of the first aspect or any of the possible implementations of the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product comprising program instructions which, when run on a computer, cause the computer to perform the method of the first aspect or any of the possible implementations of the first aspect.
In a sixth aspect, embodiments of the present application provide a chip system, where the chip system includes a processor for implementing the functions in the methods described above. In one possible implementation, the system on a chip may also include memory for storing program instructions and/or data. The chip system may be formed of a chip or may include a chip and other discrete devices.
In a seventh aspect, the present application provides a short-term wind power prediction system comprising at least one computer device for performing any of the methods of the first aspect described above, when at least one of the aforementioned computer devices is operating in the short-term wind power prediction system.
Drawings
In order to more clearly illustrate the embodiments of the present application, the drawings that are used in the embodiments will be briefly described below. It will be obvious to those skilled in the art that other figures may be obtained from these figures without the inventive effort.
Fig. 1 is a schematic diagram of a network architecture according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of a short-term wind power prediction method according to an embodiment of the present application;
FIG. 3 is an exploded flow chart of a VMD provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of an LSTM model according to an embodiment of the present application;
FIG. 5 is a schematic flow chart of constructing a short-term wind power prediction model according to an embodiment of the present application;
FIG. 6 is a schematic diagram of an original wind speed versus wind power sequence provided by an embodiment of the present application;
FIG. 7 is a schematic diagram of VMD decomposition results provided in an embodiment of the present application;
fig. 8 is a schematic diagram of a VMD decomposition spectrum according to an embodiment of the disclosure;
FIG. 9 is a schematic diagram of approximate entropy values of modal components provided in an embodiment of the present application;
FIG. 10 is a schematic diagram of the reconstructed modal components provided by the embodiments of the present application;
FIGS. 11 and 12 are schematic diagrams of a different algorithm optimized LSTM network prediction model provided in an embodiment of the present application;
FIGS. 13 and 14 are schematic diagrams of a comparison of curves provided in embodiments of the present application;
FIG. 15 is a schematic structural diagram of a short-term wind power prediction device according to an embodiment of the present disclosure;
fig. 16 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. Wherein the terms "system" and "network" in embodiments of the present application may be used interchangeably. Unless otherwise indicated, "/" indicates that the associated object is an "or" relationship, e.g., A/B may represent A or B; the term "and/or" in this application is merely an association relation describing an association object, and means that three kinds of relations may exist, for example, a and/or B may mean: there are three cases, a alone, a and B together, and B alone, wherein a, B may be singular or plural. Also, in the description of the present application, unless otherwise indicated, "a plurality" means two or more than two. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be one or more. In addition, in order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first", "second", etc. are used to distinguish the network element from the same item or similar items having substantially the same effect. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ.
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
The following detailed description is provided for further details of the objects, technical solutions and advantageous effects of the present application, and it should be understood that the following description is only a detailed description of the present application, and is not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present application should be included in the scope of the present application.
The following description of technical terms that may appear in embodiments of the present application is given first, and the terms used in the implementation section of the present application are only used to explain specific embodiments of the present application and are not intended to limit the present application. In the various embodiments of the application, if there is no specific description or logical conflict, terms and/or descriptions between the various embodiments are consistent and may reference each other, and features of the various embodiments may be combined to form new embodiments according to their inherent logical relationships.
(1) Whale optimization algorithm
The whale optimization algorithm is a novel heuristic optimization algorithm which simulates hunting behavior of whales with the head. In the whale optimization algorithm, the position of each whale can represent a feasible solution. The method has the advantages of simple operation, few adjusted parameters and strong capability of jumping out of local optimum.
(2) Variational Modal Decomposition (VMD)
The variation modal decomposition can be applied to nonlinear time series signals, and the signals can be extracted mainly by utilizing the idea of solving the variation problem, so that one original signal is decomposed into a plurality of signals with different center frequencies under the condition that the characteristics of the original signal are not lost, namely, the signals are not in the same modulation signal.
(3)LSTM
LSTM is a special variant of a recurrent neural network (recurrent neural network, RNN), a time-recurrent neural network. LSTM is suitable for processing and predicting important events with very long intervals and delays in a time series. As a nonlinear model, LSTM can be used as a complex nonlinear unit to construct larger deep neural networks.
Referring to fig. 1, fig. 1 is a schematic diagram of a network architecture according to an embodiment of the present application. As shown in fig. 1, the network architecture may include a server 10d and a terminal cluster, which may include one or more terminal devices, without limiting the number of terminal devices included in the terminal cluster. As shown in fig. 1, the terminal cluster may specifically include a terminal device 10a, a terminal device 10b, a terminal device 10c, and the like; all terminal devices in the terminal cluster (which may include, for example, terminal device 10a, terminal device 10b, and terminal device 10c, etc.) may be in network connection with the server 10d, so that each terminal device may interact with the server 10d through the network connection.
The terminal devices of the terminal cluster may include, but are not limited to: electronic devices such as smart phones, tablet computers, notebook computers, palm computers, mobile internet devices (mobile internet device, MID), wearable devices (such as smart watches, smart bracelets and the like), intelligent voice interaction devices, intelligent household appliances (such as smart televisions and the like), vehicle-mounted devices, aircrafts and the like, and the type of terminal devices is not limited. It will be appreciated that each terminal device in the terminal cluster shown in fig. 1 may be provided with an application client, and when the application client runs in each terminal device, data interaction may be performed between the application client and the server 10d shown in fig. 1. The application client running in each terminal device may be an independent client, or may be an embedded sub-client integrated in a certain client, which is not limited in this application.
The application client may specifically include, but is not limited to: a client having video-text processing functions such as a browser, a vehicle-mounted client, a smart home client, an entertainment client (e.g., a game client), a multimedia client (e.g., a video client, a short video client), a conference client, and a social client. If the terminal device included in the terminal cluster is a vehicle-mounted device, the vehicle-mounted device may be an intelligent terminal in an intelligent traffic scene, and an application client running in the vehicle-mounted device may be referred to as a vehicle-mounted client.
The server 10d may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, a content delivery network (Content Delivery Network, CDN), basic cloud computing services such as big data and an artificial intelligence platform, and the type of the server is not limited in this application.
It should be noted that, the short-term wind power prediction method provided in the embodiment of the present application may be executed by a computer device, where the computer device may be a server 10d (the server 10d may be a background server corresponding to an application client) in a network architecture shown in fig. 1, or any one of terminal devices in a terminal cluster, or may be a computer program (including program code, for example, an application client integrated by the terminal device), and the embodiment of the present application is not limited to this.
In combination with the above network architecture, a short-term wind power prediction method provided in the embodiments of the present application is described below. Referring to fig. 2, fig. 2 is a flowchart of a short-term wind power prediction method according to an embodiment of the present application. It will be appreciated that the short-term wind power prediction method may be performed by a computer device, which may be a server, or may be a terminal device, which is not limited in this application. As shown in fig. 2, the short-term wind power prediction method may include the following steps S201 to S206:
s201: historical data is collected.
The computer device may collect historical data, wherein the historical data may include wind speed. Further alternatively, the historical data may also include wind direction, yaw angle, fan data power, and the like.
S202: and carrying out modal decomposition on the historical data through the VMD to obtain P modal components.
After the computer equipment collects the historical data, the historical data can be subjected to modal decomposition through the VMD to obtain P modal components. That is, the computer device may decompose the history data as an input signal into modal components of different characteristics, the estimated bandwidth and minimum constraint of each modal component being the sum of all modalities; estimating component bandwidths by using all modal components and equal input signals as constraint conditions and utilizing Gaussian smoothing and gradient square norms to obtain a VMD variation model with constraint; converting the constrained VMD variational model into an unconstrained VMD variational model by using a quadratic penalty factor and a Lagrangian multiplier; p modal components are determined according to the unconstrained VMD variational model and convergence criteria. Specifically:
The VMD decomposition process may include the following five steps:
step 1: and constructing a constrained problem model by using a variational method.
Decomposing the history data as an input signal f (t) into P modal components u p Ensuring that the decomposition sequence is a finite eigenmode component with a center frequency, and simultaneously ensuring that the estimated bandwidth and minimum constraint condition of each mode are the sum of all modesIts Hilbert transform, and frequency modulation:
and estimating component bandwidths by using all modal components and equal input signals as constraint conditions and utilizing Gaussian smoothing and gradient square norms to obtain a VMD variation model with constraint:
where δ (t) represents a dirac function, x represents a convolution operator, ω p Representing the center frequency corresponding to the p-th component.
Step 2: and (3) introducing a penalty term alpha and a Lagrange operator into the formula (2) to obtain an unconstrained problem model, wherein the formula (3) is shown.
In order to obtain an optimal solution of the VMD model, reconstructing data by using a secondary penalty factor alpha, introducing a Lagrange multiplier operator lambda, and converting a constrained VMD variational model problem into an unconstrained problem to obtain an augmented Lagrange function:
step 3: initialization ofλ 1 And the number of iterations n.
Iterative solution of the saddle point by using the alternating direction method of the multiplication operator to obtain a modal component u p With center frequency omega p ：
In the method, in the process of the invention,for the current->Least squares filtering of ∈j->Corresponding to the frequency center of the power spectrum of the current mode function, after the iteration is completed, for +.>Performing Fourier transform to obtain a real part of u p (t)。
Step 4: updating parametersAnd lambda.
Step 5: judging convergence criterionIf the requirement is satisfied, returning to Step4 if the condition is not satisfied, and outputting +.>And->
In one possible embodiment, referring to fig. 3, fig. 3 is a flowchart illustrating a VMD decomposition according to an embodiment of the present application. As shown in fig. 3, the number of decomposition layers P may be first determined, where the number of iterations n=0, and initializedλ 1 P=1, and the modal component u is determined using equation (4) above p Judging whether p isEqual to P, if yes, p=1, continuing to determine the center frequency ω according to equation (5) above p If not, p=p+1, continuing to determine the modal component u according to equation (4) above p . Determining the center frequency ω according to the above (5) p Afterwards, judging whether P is equal to P, if so, updating +.>If not, p=p+1, continuing to determine the center frequency ω according to equation (5) above p . Update->After that, judge->If less than ε, if not, then n=n+1, perform initialization ∈ ->λ 1 P=1, and if so, P modal components can be obtained.
S203: the P modal components are reconstructed into N subsequences based on the approximate entropy.
The computer device may reconstruct the P modal components into N subsequences based on the approximate entropy. That is, for P modal components after VMD decomposition, the complexity of the P modal components may be estimated using approximate entropy to obtain N subsequences after reconstruction, where the N subsequences are subcomponents with typical characteristics. Specifically:
for a given time sequence x (1), x (2), x (N), the input integer m and the positive real number r represent the embedding dimension and the similarity tolerance. And construct an m-dimensional vector sequence X (i) = [ X (1), X (2), …, X (N-m+1)]Where i=1, 2,..n-m+1, for a certain fixed i, d between x (i) and the remaining vectors is calculated ij ：
d ij ＝f m (X(i),X(j))＝max 0≤k≤m-1 (|x(i+k)-x(j+k)|) (6)
Wherein: j=1, 2,..n-m+1, j+.i
Calculate d ij Number p and distance of < hRatio of the total number of leaves
For a pair ofTaking the logarithm and calculating the average value phi of all the i m (h)：
Increasing the embedding dimension to m+1 to obtain a new oneAnd phi is phi m (h) When N takes a finite value, an approximate entropy estimated value is obtained:
AppEn＝Φ m (r)-Φ m+1 (r) (9)
the approximate entropy is a dimensionless scalar, the value of which is related to m and h, and m= 2,h =0.25d S Wherein D is S Is the standard deviation of the time series x (i).
S204: and carrying out parameter optimization on the LSTM neural network model based on the improved WOA to obtain the optimal LSTM parameters.
The computer device may perform a parameter optimization on the LSTM neural network model based on the improved WOA to obtain optimal LSTM parameters. That is, the computer device may take the parameters to be optimized of the LSTM neural network model as an initialization solution to improve WOA; and optimizing the parameters to be optimized of the LSTM neural network model by adopting the improved WOA to obtain the optimal LSTM parameters. Initializing parameters of the improved WOA, determining the current optimal value and the optimal solution of the improved WOA, and obtaining the optimal LSTM parameters according to the current optimal value and the optimal solution of whales. Specifically:
as an improved recurrent neural network (recurrent neural network, LSTM), the LSTM overcomes the gradient disappearance problem of the traditional RNN during back propagation, improves the short-time memory defect and the like, enhances the long-sequence processing capability, and can overcome the prediction accuracy problem under the influence of severe environment. Referring to fig. 4, fig. 4 is a schematic diagram of an LSTM model provided in an embodiment of the present application. The mathematical model expression may be as follows:
where σ is the sigmod activation function, W f 、W i 、W C And W is O B is a weight matrix of a forgetting gate, an input gate, a memory node and an output gate at the moment t respectively f 、b i 、b C 、b O To correspond to the bias parameters of the gate, x t 、h t 、C t The input information, the output information and the state information at the current t moment. In order to reduce the influence of random weight and bias parameters on the model prediction effect, intelligent algorithm optimization parameters can be quoted to determine the model structure and improve the generalization capability of the model in the wind power output prediction process training.
In one possible embodiment, please refer to fig. 5, fig. 5 is a schematic flow chart of a short-term wind power prediction model according to an embodiment of the present application. As shown in fig. 5, the WOA may represent a solution of the problem to be optimized as a feasible solution, and perform location optimization according to three strategies of local search, spiral search and random search, so as to finally converge on the optimal solution.
1) Local optimization
The algorithm assumes that the individual closest to the position of the prey is the current local optimal solution, other individuals walk around the position of the current optimal solution, gradually surround the prey in a local optimizing mode, and finish position updating.
D＝|CX * (t)-X(t)| (12)
Wherein t represents the current iteration number, X (t) represents the position at the time of the t-th iteration, X * (t) represents the current local optimal solution, D represents the distance of the whale position from the current local optimal position, c=2r, a=2a (r-1), r represents the random number in (0, 1), and a decreases linearly with the number of iterations from 2 to 0.b is a constant, l represents a random number in (0, 1), P.epsilon.0, 1 ]。
2) Random optimization
Besides the local optimizing mode, partial whales can randomly search one whale according to the position of the whale to update the position.
X(t+1)＝X rand (t)-AD (13)
D＝|CX rand (t)-X(t)| (14)
X rand (t) is a random position. When |A|>1, whale will update randomly with random optimizing formula, when |A| < 1, further determine the next walk mode with probability P, as shown in formula (11), thus the influence of control parameter a on the value of parameter |A| is obvious. The method can effectively overcome the influence of the control parameter a on convergence accuracy, reduce the control parameter a in a cosine curve form, and effectively solve the problem of early-maturing in the later stage of the algorithm.
When the traditional WOA algorithm initializes the population, certain randomness exists, the search space cannot be traversed, and the optimal point set method is utilized to initialize the population in order to effectively improve the overall convergence accuracy of the algorithm. That is, let Gs be the unit cube of the S-dimensional European space, if r ε Gs, it is shaped as:
wherein P is n (k) For the best point set, the deviationIs->Wherein C (r, ε) n -1+ε Is a constant that is related solely to the sweet spot r and any integer epsilon. />The decimal part n represents the point number, and
p is the minimum prime number satisfying (p-3)/2. Gtoreq.s, mapped to search spaceWherein ub is j Represents the upper bound of the j-th dimension, lb j Representing the lower bound of dimension j.
The method has the advantages that the problem of dependence of an initialization population is overcome by traversing the search space through the optimal point set, the control convergence factor is reduced in a cosine curve mode, the algorithm local optimizing capability is further improved, a neural network model for optimizing LSTM parameters based on an improved WOA algorithm is established, and the method has the advantages of simplicity in operation, strong robustness, few control parameters and the like, and can achieve remarkable effect in solving the complex engineering problem, so that the WOA is used for optimizing and training parameters in the LSTM neural network, the optimal weight and the threshold can be automatically selected, the randomness obtained by experience selection is reduced, the accuracy of the model is improved, and the accuracy of short-term wind power prediction is improved.
S205: and constructing a short-term wind power prediction model based on the N subsequences after reconstruction and the optimal LSTM parameters.
That is, the N subsequences after reconstruction are respectively used as the input of an improved LSTM model, and the improved LSTM model is trained; and obtaining the short-term wind power prediction model based on the trained improved LSTM model and the optimal LSTM parameters.
S206: and inputting the prediction data into a short-term wind power prediction model to obtain a short-term wind power prediction result.
In one possible implementation, the computer device may superimpose the predictions of all modal components to obtain a final prediction.
With reference to S201-S206 above, one possible embodiment is described below.
For example, 2000 sets of meteorological data and measured power data of a certain wind farm can be selected, wherein wind speed is taken as a model input, wind power is taken as a model output, and a data sampling interval is 1h. The time series of the original wind speed and the wind power can be shown in fig. 6 (a) and fig. 6 (b), and it can be seen that the original wind speed signal has obvious nonlinearity and fluctuation.
To get a more accurate prediction result, it can be decomposed using VMD-approximated entropy. First, an input wind speed signal is decomposed into a series of modal components with different characteristics as shown in fig. 7, and a spectrum diagram of each component after decomposition is shown in fig. 8. For each sub-sequence after VMD decomposition, the complexity is estimated using the approximate entropy, with the component approximate entropy values shown in FIG. 9. As can be seen from FIG. 9, the similar entropy values of the modes IMF1 and IMF2 are different, which indicates that the complexity of the two modes is high, the two modes are respectively defined as a sub-sequence 1 and a sub-sequence 2, the difference of the entropy values of the modes IMF 3-5 is smaller than 0.1, the maximum difference is 0.016, and certain similarity is presented, which indicates that the complexity of the three modes is low, the three groups of mode components are combined into a sub-sequence 3, the difference of the entropy values of the mode 6 and the mode 7 is equal to 0.022, the modes IMF 6-7 are defined as a sub-sequence 4, the mode 8 is defined as a sub-sequence 5, the obtained sub-sequence after recombination is shown in FIG. 10, and the sub-sequences after recombination are respectively used as the model input of the improved LSTM.
In order to better show the discrete degree of the predicted data change, the fluctuation condition and the mean value of the errors, the embodiment of the application can select root mean square error (root mean square error, RMSE) and mean absolute error (mean absolute error, MAE) as evaluation indexes of a prediction model, and the formula is as follows:
where vi' represents a predicted power value and vi represents an actual power value.
And (3) obtaining new 5 subsequences through the signal decomposition and recombination, respectively training the recombined 5 subsequences, predicting one by one, and then superposing predicted values of all components to obtain a final predicted result.
Further optionally, in order to further test the accuracy and effectiveness of the model provided by the embodiment of the present application, an LSTM model, a WOA-LSTM model and a VMD-LSTM model may be set as reference models for comparison, the WOA-LSTM model is used to optimize weights and biases of gates of the LSTM neural network, in order to eliminate the influence of dimensions, normalization processing may be performed on input data, and supposing that the maximum iteration number of WOA is 25, population size 20 is provided, different algorithms optimize LSTM network prediction model curves as shown in fig. 11 and 12, it may be seen that a single LSTM model can track power output, but the difference between the prediction curves and actual curves is larger, after the WOA algorithm is used alone to optimize LSTM model parameters, the wind power curves may be accurately fitted in case of small wind power variation, but in case of a weather mutation, the prediction curves cannot be accurately tracked, so that the problem of insufficient control of wind power prediction error exists.
Referring to fig. 13 and 14, fig. 13 and 14 are graphs comparing curves provided in the embodiments of the present application. As shown in fig. 13 and 14, a graph of the change curve of the difference between the predicted value and the true value in the four models of 5 months and 11 months can be shown. It can be seen that the single LSTM network error fluctuation is larger, the error fluctuation range of the WOA-LSTM model and the VMD-LSTM model is smaller, but the single-point error maximum condition exists, and the error curve of the model provided by the embodiment of the application is closer to a value of 0, and no obvious fluctuation exists.
Tables 1 and 2 below can represent the comparison of prediction errors for the four models for 5 months and 11 months, respectively.
Table 1 comparison of 5 month prediction errors for four models
Predictive model MAX MIN MAE RMSE
LSTM 17.5700 0.1158 8.4152 8.3287
WOA-LSTM 18.5350 0.1163 5.0715 4.9370
VMD-LSTM 17.6900 0.0736 4.1023 3.8614
Improved VMD-LSTM 2.7070 0.0075 1.3057 1.2154
Table 2 comparison of the 10 month prediction errors for the four models
As can be seen from tables 1 and 2, taking the 5-month prediction result error as an example, the root mean square error and the standard error of the LSTM model are large, the root mean square error is 8.1452, and the standard error is 8.3287. After model parameters are optimized by using an algorithm, the error is obviously reduced, the root mean square error of the prediction model provided by the embodiment of the application is reduced to 1.2154, and the standard error is reduced to 1.3057. Better than the other three models, indicating that predictions within this range are relatively stable, and that both MSE and RSME are lower than the other three models.
In the embodiment of the application, the LSTM model can be improved based on the improved VMD and the approximate entropy combined with a whale optimization algorithm and is applied to short-term wind power prediction. Because the whale optimization algorithm has the advantages of few parameters, simple structure, high prediction precision and the like, the improved whale optimization algorithm is utilized to perform optimization training on the parameters in the LSTM neural network, the optimal weight and the threshold can be automatically selected, the randomness obtained by experience selection is reduced, the precision of the model is improved, and the precision of short-term wind power prediction is improved.
The foregoing describes a method provided by the present application, and in order to facilitate better implementation of the foregoing solutions of the embodiments of the present application, the embodiments of the present application further provide corresponding apparatuses.
The embodiment of the application may divide the functional modules of the apparatus according to the above method example, for example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present application, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.
Referring to fig. 15, fig. 15 is a schematic structural diagram of a short-term wind power prediction apparatus according to an embodiment of the present application. The short-term wind power prediction device may be a computer device, a module (e.g., a chip or a processor) in the computer device, or a logic module or software that can implement all or part of the functions of the computer device. As shown in fig. 15, the short-term wind power prediction apparatus 1500 includes at least: the device comprises an acquisition unit 1501, a decomposition unit 1502, a reconstruction unit 1503, an optimization unit 1504, a construction unit 1505 and an input unit 1506; wherein:
an acquisition unit 1501 for acquiring history data including wind speed;
a decomposition unit 1502, configured to perform modal decomposition on the history data by using a variant modal decomposition VMD to obtain P modal components, where P is a positive integer greater than or equal to 1;
a reconstruction unit 1503, configured to reconstruct the P modal components into N subsequences based on approximate entropy, where N is less than or equal to P;
the optimizing unit 1504 is used for carrying out parameter optimization on the LSTM neural network model of the long-term and short-term memory network based on the WOA (WOA) of the improved whale optimizing algorithm to obtain optimal LSTM parameters;
a construction unit 1505, configured to construct a short-term wind power prediction model based on the reconstructed N subsequences and the optimal LSTM parameter;
The input unit 1506 is configured to input the prediction data into the short-term wind power prediction model, to obtain a prediction result of the short-term wind power.
In one embodiment, the optimization unit 1504 performs parameter optimization on the long-short-term memory network LSTM neural network model based on the improved whale optimization algorithm WOA to obtain optimal LSTM parameters, which is specifically configured to:
taking parameters to be optimized of the LSTM neural network model as an initialization solution of the improved WOA;
and optimizing the parameters to be optimized of the LSTM neural network model by adopting the improved WOA to obtain the optimal LSTM parameters.
In one embodiment, the optimizing unit 1504 uses the improved WOA to optimize the parameters to be optimized of the LSTM neural network model to obtain the optimal LSTM parameters, which is specifically configured to:
initializing parameters of the improved WOA;
determining the current optimal value and the optimal solution of the improved WOA;
and obtaining the optimal LSTM parameter according to the current optimal value and the optimal solution of the whale.
In one embodiment, the decomposing unit 1502 performs modal decomposition on the historical data through the VMD to obtain P modal components, which is specifically configured to:
the historical data is used as an input signal and is decomposed into modal components with different characteristics, and the estimated bandwidth and the minimum constraint condition of each modal component are the sum of all modalities;
Estimating component bandwidths by using the modal components and the input signals as constraint conditions and using Gaussian smoothing and gradient square norms to obtain a VMD variation model with constraint;
converting the constrained VMD variational model into an unconstrained VMD variational model by using a quadratic penalty factor and a Lagrangian multiplier;
and determining the P modal components according to the unconstrained VMD variational model and convergence criteria.
In one embodiment, the reconstruction unit 1503 reconstructs the P modal components into N subsequences based on approximate entropy, specifically for:
and for the P modal components after VMD decomposition, estimating the complexity of the P modal components by adopting approximate entropy to obtain N subsequences after reconstruction, wherein the N subsequences are the subsequences with typical characteristics.
In one embodiment, the construction unit 1505 constructs a short-term wind power prediction model based on the reconstructed N subsequences and the optimal LSTM parameters, specifically for:
respectively taking the N subsequences after reconstruction as the input of an improved LSTM model, and training the improved LSTM model;
and obtaining the short-term wind power prediction model based on the trained improved LSTM model and the optimal LSTM parameters.
In one embodiment, the historical data further includes wind direction, yaw angle, and fan output power.
For more detailed descriptions of the above-mentioned acquisition unit 1501, decomposition unit 1502, reconstruction unit 1503, optimization unit 1504, construction unit 1505 and input unit 1506, reference may be made directly to the relevant descriptions of the computer devices in the above-mentioned method embodiments shown in fig. 2-14, which are not repeated here.
Further, referring to fig. 16, fig. 16 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 16, the computer device 1000 may be a terminal device, for example, the terminal device 10a in the embodiment corresponding to fig. 1, or a server, for example, the server 10d in the embodiment corresponding to fig. 1, which is not limited herein. For ease of understanding, the present application takes a computer device as an example of a terminal device, and the computer device 1000 may include: processor 1001, network interface 1004, and memory 1005, in addition, the computer device 1000 may further comprise: a user interface 1003, and at least one communication bus 1002. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may also include a standard wired interface, a wireless interface, among others. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 1005 may also optionally be at least one storage device located remotely from the processor 1001. As shown in fig. 16, an operating system, a network communication module, a user interface module, and a device control application program may be included in a memory 1005, which is one type of computer-readable storage medium.
The network interface 1004 in the computer device 1000 may also provide network communication functions, and the optional user interface 1003 may also include a Display screen (Display) and a Keyboard (Keyboard). In the computer device 1000 shown in fig. 16, the network interface 1004 may provide network communication functions; while user interface 1003 is primarily used as an interface for providing input to a user; and the processor 1001 may be used to invoke a device control application stored in the memory 1005 to implement:
collecting historical data, wherein the historical data comprises wind speed;
performing modal decomposition on the historical data through a variational modal decomposition VMD to obtain P modal components, wherein P is a positive integer greater than or equal to 1;
reconstructing the P modal components into N subsequences based on approximate entropy, wherein N is less than or equal to P;
performing parameter optimization on the LSTM neural network model of the long-term and short-term memory network based on an improved whale optimization algorithm WOA to obtain optimal LSTM parameters;
constructing a short-term wind power prediction model based on the N subsequences after reconstruction and the optimal LSTM parameters;
and inputting the prediction data into the short-term wind power prediction model to obtain a short-term wind power prediction result.
It should be appreciated that the computer device 1000 described in embodiments of the present application may perform the description of the short-term wind power prediction method in any of the embodiments of fig. 2-14 above.
Furthermore, it should be noted here that: the embodiments of the present application further provide a computer readable storage medium, where a computer program is stored, and the computer program includes program instructions, when executed by a processor, can perform the description of the short-term wind power prediction method in any of the foregoing embodiments of fig. 2 to 14, and therefore, a detailed description will not be given here. In addition, the description of the beneficial effects of the same method is omitted. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like. For technical details not disclosed in the embodiments of the computer-readable storage medium according to the present application, please refer to the description of the method embodiments of the present application. As an example, program instructions may be deployed to be executed on one computing device or on multiple computing devices at one site or, alternatively, across multiple computing devices distributed across multiple sites and interconnected by a communication network, where the multiple computing devices distributed across multiple sites and interconnected by the communication network may constitute a blockchain system.
In addition, it should be noted that: embodiments of the present application also provide a computer program product or computer program that may include computer instructions that may be stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor may execute the computer instructions, so that the computer device performs the foregoing description of the short-term wind power prediction method in any of the embodiments of fig. 2 to 14, and thus, a detailed description thereof will not be provided herein. In addition, the description of the beneficial effects of the same method is omitted. For technical details not disclosed in the computer program product or the computer program embodiments related to the present application, please refer to the description of the method embodiments of the present application.
The terms first, second and the like in the description and in the claims and drawings of the embodiments of the present application are used for distinguishing between different media content and not for describing a particular sequential order. Furthermore, the term "include" and any variations thereof is intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or device that comprises a list of steps or elements is not limited to the list of steps or modules but may, in the alternative, include other steps or modules not listed or inherent to such process, method, apparatus, article, or device.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The methods and related devices provided in the embodiments of the present application are described with reference to the method flowcharts and/or structure diagrams provided in the embodiments of the present application, and each flowchart and/or block of the method flowcharts and/or structure diagrams may be implemented by computer program instructions, and combinations of flowcharts and/or blocks in the flowchart and/or block diagrams. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or structural diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or structures.
The foregoing disclosure is only illustrative of the preferred embodiments of the present application and is not intended to limit the scope of the claims herein, as the equivalent of the claims herein shall be construed to fall within the scope of the claims herein.

Claims (10)

1. A short-term wind power prediction method, comprising:
collecting historical data, wherein the historical data comprises wind speed;
performing modal decomposition on the historical data through a variational modal decomposition VMD to obtain P modal components, wherein P is a positive integer greater than or equal to 1;
reconstructing the P modal components into N subsequences based on approximate entropy, wherein N is less than or equal to P;
performing parameter optimization on the LSTM neural network model of the long-term and short-term memory network based on an improved whale optimization algorithm WOA to obtain optimal LSTM parameters;
constructing a short-term wind power prediction model based on the N subsequences after reconstruction and the optimal LSTM parameters;
and inputting the prediction data into the short-term wind power prediction model to obtain a short-term wind power prediction result.
2. The method of claim 1, wherein the performing parameter optimization on the LSTM neural network model based on the improved WOA to obtain the optimal LSTM parameters comprises:
taking parameters to be optimized of the LSTM neural network model as an initialization solution of the improved WOA;
And optimizing the parameters to be optimized of the LSTM neural network model by adopting the improved WOA to obtain the optimal LSTM parameters.
3. The method of claim 2, wherein optimizing the parameters to be optimized of the LSTM neural network model using the improved WOA to obtain the optimal LSTM parameters, comprises:
initializing parameters of the improved WOA;
determining the current optimal value and the optimal solution of the improved WOA;
and obtaining the optimal LSTM parameter according to the current optimal value and the optimal solution of the whale.
4. A method according to any one of claims 1-3, wherein the performing modal decomposition on the history data by the VMD to obtain P modal components includes:
the historical data is used as an input signal and is decomposed into modal components with different characteristics, and the estimated bandwidth and the minimum constraint condition of each modal component are the sum of all modalities;
estimating component bandwidths by using the modal components and the input signals as constraint conditions and using Gaussian smoothing and gradient square norms to obtain a VMD variation model with constraint;
converting the constrained VMD variational model into an unconstrained VMD variational model by using a quadratic penalty factor and a Lagrangian multiplier;
And determining the P modal components according to the unconstrained VMD variational model and convergence criteria.
5. The method of claim 4, wherein reconstructing the P modal components into N subsequences based on approximate entropy comprises:
and for the P modal components after VMD decomposition, estimating the complexity of the P modal components by adopting approximate entropy to obtain N subsequences after reconstruction, wherein the N subsequences are the subsequences with typical characteristics.
6. The method of claim 5, wherein constructing a short-term wind power prediction model based on the reconstructed N subsequences and the optimal LSTM parameters comprises:
respectively taking the N subsequences after reconstruction as the input of an improved LSTM model, and training the improved LSTM model;
and obtaining the short-term wind power prediction model based on the trained improved LSTM model and the optimal LSTM parameters.
7. The method of any of claims 1-6, wherein the historical data further includes wind direction, yaw angle, and fan output power.
8. A short-term wind power prediction device, characterized by comprising means for performing the method according to any of claims 1-7.
9. A short-term wind power prediction device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the short-term wind power prediction method of any one of claims 1-7 when executing the computer program.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program or computer instructions, which, when executed by a processor, implement the method of any of claims 1-7.
CN202310555581.4A 2023-05-17 2023-05-17 Short-term wind power prediction method and device and computer readable storage medium Pending CN116562650A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310555581.4A CN116562650A (en) 2023-05-17 2023-05-17 Short-term wind power prediction method and device and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310555581.4A CN116562650A (en) 2023-05-17 2023-05-17 Short-term wind power prediction method and device and computer readable storage medium

Publications (1)

Publication Number Publication Date
CN116562650A true CN116562650A (en) 2023-08-08

Family

ID=87494370

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310555581.4A Pending CN116562650A (en) 2023-05-17 2023-05-17 Short-term wind power prediction method and device and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN116562650A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117252448A (en) * 2023-11-20 2023-12-19 华东交通大学 Wind power prediction method based on space-time feature extraction and secondary decomposition polymerization

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110009125A (en) * 2019-01-23 2019-07-12 华南理工大学 Short-term power load prediction model establishment method based on EMD-VMD-PSO-BPNN
CN115310669A (en) * 2022-07-06 2022-11-08 国家能源集团新能源技术研究院有限公司 Ultra-short-term wind power prediction method based on secondary decomposition and IWOA-LSSVM

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110009125A (en) * 2019-01-23 2019-07-12 华南理工大学 Short-term power load prediction model establishment method based on EMD-VMD-PSO-BPNN
CN115310669A (en) * 2022-07-06 2022-11-08 国家能源集团新能源技术研究院有限公司 Ultra-short-term wind power prediction method based on secondary decomposition and IWOA-LSSVM

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
DR.JIN: "鲸鱼优化数学原理及ptyhon实现", pages 1 - 4, Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/347741616> *
TONGLIN FU: "Hybrid the long short-term memory with whale optimization algorithm and variational mode decomposition for monthly evapotranspiration estimation", 《SCIENTIFIC REPORTS》, vol. 12, pages 1 - 9 *
岳晓宇;彭显刚;林俐;: "鲸鱼优化支持向量机的短期风电功率预测", 电力系统及其自动化学报, no. 02, pages 151 - 155 *
王俊;李霞;周昔东;张凯;: "基于VMD和LSTM的超短期风速预测", 电力系统保护与控制, no. 11, pages 51 - 58 *
蔡小鹏: "基于 VMD-IWOA-LSTM 的短期电力负荷预测研究", 《中国优秀博硕士学位论文全文数据库（硕士）工程科技Ⅱ辑》, no. 3, pages 042 - 868 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117252448A (en) * 2023-11-20 2023-12-19 华东交通大学 Wind power prediction method based on space-time feature extraction and secondary decomposition polymerization
CN117252448B (en) * 2023-11-20 2024-02-20 华东交通大学 Wind power prediction method based on space-time feature extraction and secondary decomposition polymerization

Similar Documents

Publication Publication Date Title
Ferreira et al. Multi-step ahead forecasting of daily reference evapotranspiration using deep learning
Liang et al. A novel wind speed prediction strategy based on Bi-LSTM, MOOFADA and transfer learning for centralized control centers
Kavousi-Fard A hybrid accurate model for tidal current prediction
US11861474B2 (en) Dynamic placement of computation sub-graphs
CN111406264A (en) Neural architecture search
CN110222838B (en) Document sorting method and device, electronic equipment and storage medium
Guinness et al. Compression and conditional emulation of climate model output
CN116562650A (en) Short-term wind power prediction method and device and computer readable storage medium
Acquaviva et al. SED fitting with MCMC: methodology and application to large galaxy surveys
CN115238855A (en) Completion method of time sequence knowledge graph based on graph neural network and related equipment
Peng et al. A cloud load forecasting model with nonlinear changes using whale optimization algorithm hybrid strategy
CN115221396A (en) Information recommendation method and device based on artificial intelligence and electronic equipment
CN113836823A (en) Load combination prediction method based on load decomposition and optimized bidirectional long-short term memory network
Maulik et al. Efficient high-dimensional variational data assimilation with machine-learned reduced-order models
Said et al. AI-based solar energy forecasting for smart grid integration
CN116090536A (en) Neural network optimization method, device, computer equipment and storage medium
Eban et al. Learning the experts for online sequence prediction
Yu et al. Analysis and Application of the Spatio-Temporal Feature in Wind Power Prediction.
JP2024504179A (en) Method and system for lightweighting artificial intelligence inference models
CN113822291A (en) Image processing method, device, equipment and storage medium
Zhao et al. Spatio-temporal Model Combining VMD and AM for Wind Speed Prediction
Kummer Dynamic Neural Network Architectural and Topological Adaptation and Related Methods--A Survey
CN117150326B (en) New energy node output power prediction method, device, equipment and storage medium
CN115114345B (en) Feature representation extraction method, device, equipment, storage medium and program product
CN115392594B (en) Electrical load model training method based on neural network and feature screening

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