Distributed power optimization method and system of wind-solar power generation energy storage system

Abstract

The invention relates to the technical field of power optimization, and discloses a distributed power optimization method and system for a wind-solar power generation and energy storage system, and the method comprises the steps: calculating a daily power generation correlation coefficient between historical power generation data in a historical power generation data set of a target wind-solar power generation and energy storage system; performing data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind and light power generation data set, and performing data optimization on the wind and light power generation data set to obtain a target power generation data set; performing parameter optimization on the power prediction model according to the target power generation data set to obtain a target prediction model; calculating a predicted wind power sequence and a predicted photoelectric power sequence by using the target prediction model, and calculating power change probability distribution of the wind power sequence and the predicted photoelectric power sequence; and performing power adjustment according to the power change probability distribution pair to obtain the optimized output power of the target wind and light power generation energy storage system. According to the invention, the accuracy of distributed power optimization of the wind-solar power generation energy storage system can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power optimization, and in particular to a distributed power optimization method and system for a wind-solar power generation and energy storage system. Background Art

[0002] Distributed generation (DG) such as photovoltaic and wind power has developed rapidly due to its advantages of environmental protection, high efficiency and cleanliness. The wind-solar power generation and energy storage system is an energy system that combines wind and solar power generation. It uses energy storage technology to solve the instability problem of wind and photovoltaic power generation, and can improve energy utilization and the absorption capacity of the power grid. At the same time, through distributed power optimization of wind-solar power generation and energy storage systems, the power output of energy storage equipment can be finely managed and controlled to achieve efficient distribution and utilization of electric energy in the entire power grid or energy network, and realize more efficient, more economical and more environmentally friendly power system operation.

[0004] The method of distributed power optimization in the prior art is mainly to take the minimum cost as the distribution network operation target, design model constraints, establish a real-time economic dispatch model for the distribution network, and adjust the unbalanced power under the real-time economic dispatch plan within the shortest time scale. However, this method only takes the minimum cost as the constraint of the model, and does not consider the influence of distribution network planning factors, which makes the prediction error analysis accuracy not high. And by dividing the wind power interval, predicting the probability distribution of wind power, designing the dispatch model constraints, and establishing a random optimization dispatch model, there is a problem of poor effectiveness when optimizing the distributed power of wind and solar power generation and energy storage systems. Therefore, how to improve the accuracy of distributed power optimization of wind and solar power generation and energy storage systems has become an urgent problem to be solved. Summary of the invention

[0005] The present invention provides a distributed power optimization method and system for a wind-solar power generation and energy storage system, the main purpose of which is to solve the problem of poor accuracy of distributed power optimization of the wind-solar power generation and energy storage system.

[0006] To achieve the above object, the present invention provides a distributed power optimization method for a wind-solar power generation and energy storage system, comprising:

[0007] Obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between the historical power generation data in the historical power generation data set;

[0008] Performing data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and performing data optimization on the wind-solar power generation data set to obtain a target power generation data set;

[0009] Optimizing parameters of a pre-built power prediction model according to the target power generation data set to obtain a target prediction model;

[0010] Calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system by using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence;

[0011] The wind power sequence and the predicted photovoltaic power sequence are power-regulated according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

[0012] Optionally, the calculating a daily power generation correlation coefficient between historical power generation data in the historical power generation data set includes:

[0013] Constructing a daily power generation data matrix of the target wind-solar power generation and energy storage system according to the historical power generation data set;

[0014] Decentralizing the daily power generation data matrix to obtain a centralized matrix, and calculating a semi-positive definite matrix of the centralized matrix;

[0015] Calculate the daily power generation correlation coefficient between the daily power generation data matrices according to the semi-positive definite matrix;

[0016] The daily power generation correlation coefficient between the historical power generation data in the historical power generation data set is calculated using the following formula:

[0017]

[0018] Among them, RV represents the daily power generation correlation coefficient, V x represents the semi-positive definite matrix corresponding to the power generation data matrix of the xth day, T represents the matrix transpose, V y It represents the semi-positive definite matrix corresponding to the power generation data matrix of the yth day, and tr() represents the trace of the matrix.

[0019] Optionally, clustering the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set includes:

[0020] Selecting a preset number of cluster centers from the daily power generation data matrix of the historical power generation data set;

[0021] Performing initial clustering on the daily power generation data matrix according to the daily power generation correlation coefficient and the cluster center to obtain an initial clustering data matrix set;

[0022] Calculating a mean matrix of the initial clustering data matrix set, and iterating the cluster centers according to the mean matrix to obtain updated cluster centers;

[0023] Calculating an updated daily power generation correlation coefficient between the updated cluster center and the daily power generation data matrix, and determining a target cluster center according to the updated daily power generation correlation coefficient;

[0024] Clustering the daily power generation data matrix according to the target cluster center to obtain a target cluster matrix set;

[0025] A wind-solar power generation data set is extracted from the historical power generation data set according to the target clustering matrix set.

[0026] Optionally, the optimizing the wind-solar power generation data set to obtain a target power generation data set includes:

[0027] Generate a wind power data sequence and a photovoltaic data sequence according to the wind and solar power generation data set;

[0028] Performing feature decomposition on the wind power data sequence and the photovoltaic data sequence to obtain an intrinsic modal component and a residual component;

[0029] Performing component inspection on the inherent modal component to obtain a high-frequency component and a low-frequency component, and performing wavelet threshold denoising on the high-frequency component to obtain a denoised high-frequency component;

[0030] The denoised high-frequency component, the low-frequency component and the residual component are superimposed to obtain a target power generation data set.

[0031] Optionally, the performing parameter optimization on a pre-built power prediction model according to the target power generation data set to obtain a target prediction model includes:

[0032] Acquiring initial parameters of the power prediction model, generating an initial particle population according to the initial parameters; performing normalization processing on the initial particle population to obtain a target particle population;

[0033] Performing chaotic mapping on the target particle population to obtain a mapped population, and calculating a prediction error value of the target power generation data set according to the mapped population;

[0034] The mapping population is optimized according to the prediction error value to obtain target parameters, and a target prediction model is generated according to the target parameters.

[0035] Optionally, performing chaotic mapping on the target particle population to obtain a mapped population includes:

[0036] The target particle population is chaotically mapped using the following formula to obtain a mapped population:

[0037]

[0038] Among them, Bi+1 Represents the i-th target particle B in the target particle population i Corresponding mapping particles, rand(0,1) represents a random number in [0,1], and N represents the total number of target particles in the target particle population.

[0039] Optionally, the step of calculating the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system by using the target prediction model includes:

[0040] Acquire the historical data and weather information to be predicted of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system in the target wind-solar power generation and energy storage system;

[0041] Utilizing the bidirectional long short-term memory network in the target prediction model to calculate the data characteristics of the historical data to be predicted and the weather characteristics of the weather information;

[0042] The feature weights of the data feature and the weather feature are calculated respectively, and the data feature and the weather feature are weighted summed using the feature weights to obtain the target feature;

[0043] The feature weights of the data features and the weather features are calculated using the following formula:

[0044] e t =ρtanh(θh t +B)

[0045] Among them, μ t Represents the tth data feature or weather feature e t The characteristic weight of , ρ, θ represent the preset first weight coefficient and the second weight coefficient, and B represents the preset bias coefficient;

[0046] The target features are fully connected to obtain a predicted wind power sequence and a predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system.

[0047] Optionally, the calculating the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence includes:

[0048] Calculating the maximum power fluctuation of the predicted wind power sequence and the predicted photovoltaic power sequence respectively;

[0049] The maximum power fluctuation is calculated using the following formula:

[0050] ΔP T =maxP(τ)-minP(τ),τ∈[HT,H)

[0051] Where ΔP Trepresents the maximum power fluctuation, maxP(τ) represents the maximum predicted wind power and the maximum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the preset time scale T, minP(τ) represents the minimum predicted wind power and the minimum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the time scale T, and H represents the total time scale of the predicted wind power sequence and the predicted photovoltaic power sequence;

[0052] The power variation probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence is calculated according to the maximum power fluctuation amount.

[0053] Optionally, the power regulation of the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system includes:

[0054] Determining whether the power change probability distribution is greater than a preset probability threshold;

[0055] When the power change probability distribution is less than or equal to the probability threshold, using the wind power sequence and the predicted photovoltaic power sequence as optimized output power;

[0056] When the power change probability distribution step is greater than the probability threshold, power distribution optimization is performed on the wind power sequence and the predicted photovoltaic power sequence until the power change probability distribution is less than or equal to or greater than the probability threshold, thereby obtaining the optimized output power.

[0057] In order to solve the above problems, the present invention also provides a distributed power optimization system for a wind-solar power generation and energy storage system, the system comprising:

[0058] A daily power generation correlation coefficient calculation module is used to obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between historical power generation data in the historical power generation data set;

[0059] A data optimization module, configured to perform data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and perform data optimization on the wind-solar power generation data set to obtain a target power generation data set;

[0060] A model parameter optimization module, used to optimize the parameters of the pre-built power prediction model according to the target power generation data set to obtain a target prediction model;

[0061] A power change probability distribution calculation module is used to calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence;

[0062] A power regulation module is used to perform power regulation on the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

[0063] The embodiment of the present invention calculates the daily power generation correlation coefficient between the historical power generation data in the historical power generation data set, which is conducive to data clustering of the historical power generation data set according to the daily power generation correlation coefficient, and obtains a representative wind and solar power generation data set, and optimizes the wind and solar power generation data set to obtain a more accurate target power generation data set, which helps to improve the accuracy of subsequent power prediction model parameter optimization and obtain a more accurate target prediction model; the target prediction model is used to calculate the predicted wind power sequence and the predicted photovoltaic power sequence, and the power change probability distribution of the wind power sequence and the predicted photovoltaic power sequence is calculated, and the output power can be complemented by wind power and photovoltaic power to meet the output requirements of the target wind and solar power generation and energy storage system, avoid the safety impact of power fluctuations on the target wind and solar power generation and energy storage system, and effectively improve the accuracy of distributed power optimization of the wind and solar power generation and energy storage system. Therefore, the distributed power optimization method and system of the wind and solar power generation and energy storage system proposed in the present invention can solve the problem of poor distributed power optimization accuracy of the wind and solar power generation and energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic flow chart of a distributed power optimization method for a wind-solar power generation and energy storage system provided in one embodiment of the present invention;

[0065] Figure 2 A schematic diagram of a process for optimizing a wind and solar power generation data set according to an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of a process for optimizing parameters of a pre-built power prediction model provided by an embodiment of the present invention;

[0067] Figure 4 A functional module diagram of a distributed power optimization system for a wind-solar power generation and energy storage system provided by an embodiment of the present invention;

[0068] Figure 5 An electronic device diagram of a distributed power optimization system for a wind-solar power generation and energy storage system provided in one embodiment of the present invention.

[0069] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0070] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0071] The embodiment of the present application provides a distributed power optimization method for a wind-solar power generation and energy storage system. The execution subject of the distributed power optimization method for the wind-solar power generation and energy storage system includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the distributed power optimization method for the wind-solar power generation and energy storage system can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0072] Reference Figure 1 As shown, it is a schematic diagram of the process of a distributed power optimization method of a wind-solar power generation and energy storage system provided by an embodiment of the present invention. In this embodiment, the distributed power optimization method of the wind-solar power generation and energy storage system includes:

[0073] S1. Obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between the historical power generation data in the historical power generation data set.

[0074] In an embodiment of the present invention, the target wind-solar power generation and energy storage system includes multiple wind power generation and energy storage systems and photovoltaic power generation and energy storage systems. The daily power generation data of the wind power generation and energy storage systems and the photovoltaic power generation and energy storage systems within a preset time period and the weather characteristics corresponding to each power generation data are counted, wherein the daily power generation data can be composed of the hourly power generation data.

[0075] In an embodiment of the present invention, the calculating of the daily power generation correlation coefficient between the historical power generation data in the historical power generation data set includes:

[0076] Constructing a daily power generation data matrix of the target wind-solar power generation and energy storage system according to the historical power generation data set;

[0077] Decentralizing the daily power generation data matrix to obtain a centralized matrix, and calculating a semi-positive definite matrix of the centralized matrix;

[0078] The daily power generation correlation coefficient between the daily power generation data matrices is calculated according to the semi-positive definite matrix.

[0079] In an embodiment of the present invention, the daily power generation data of the wind power generation energy storage system and the photovoltaic power generation energy storage system are constructed into a 1x24-dimensional matrix to obtain a daily power generation data matrix. The data mean of the daily power generation data matrix is ​​subtracted from each power generation data in the daily power generation data matrix to obtain a centralized matrix. The centralized matrix and the centralized matrix device are matrix multiplied to obtain a semi-positive definite matrix of each daily power generation data matrix.

[0080] Specifically, the daily power generation correlation coefficient between the historical power generation data in the historical power generation data set is calculated using the following formula:

[0081]

[0082] Among them, RV represents the daily power generation correlation coefficient, V x represents the semi-positive definite matrix corresponding to the power generation data matrix of the xth day, T represents the matrix transpose, V y It represents the semi-positive definite matrix corresponding to the power generation data matrix of the yth day, and tr() represents the trace of the matrix.

[0083] In the embodiment of the present invention, the correlation between the daily power generation data of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system can be calculated respectively through the daily power generation correlation coefficient, and then representative power generation data can be selected from the historical power generation data set, which is conducive to subsequent data optimization and obtaining a more accurate target power generation data set.

[0084] S2. Clustering the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and optimizing the wind-solar power generation data set to obtain a target power generation data set.

[0085] In the embodiment of the present invention, data clustering is to extract representative power generation data from the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind and solar power generation data set to reduce data redundancy.

[0086] In the embodiment of the present invention, clustering the historical power generation data set according to the daily power generation correlation coefficient to obtain the wind and solar power generation data set includes:

[0087] Selecting a preset number of cluster centers from the daily power generation data matrix of the historical power generation data set;

[0088] Performing initial clustering on the daily power generation data matrix according to the daily power generation correlation coefficient and the cluster center to obtain an initial clustering data matrix set;

[0089] Calculating a mean matrix of the initial clustering data matrix set, and iterating the cluster centers according to the mean matrix to obtain updated cluster centers;

[0090] Calculating an updated daily power generation correlation coefficient between the updated cluster center and the daily power generation data matrix, and determining a target cluster center according to the updated daily power generation correlation coefficient;

[0091] Clustering the daily power generation data matrix according to the target cluster center to obtain a target cluster matrix set;

[0092] A wind-solar power generation data set is extracted from the historical power generation data set according to the target clustering matrix set.

[0093] In an embodiment of the present invention, a preset number of cluster centers are randomly selected from a daily power generation data matrix constructed from a historical power generation data set, and the daily power generation data matrix is ​​classified into a cluster cluster corresponding to a cluster center with a minimum daily power generation correlation coefficient between the cluster centers to obtain an initial cluster data matrix set, and the mean of the daily power generation data matrix in each initial cluster data matrix set is calculated to obtain a mean matrix, and the mean matrix is ​​used as an updated cluster center, and the updated daily power generation correlation coefficient is calculated to update the clustering of the daily power generation data matrix until the distance between the updated cluster center and the current cluster center is less than a preset distance threshold, and a target cluster center is obtained.

[0094] Furthermore, the daily power generation data matrix is ​​clustered through the target clustering center to obtain a target clustering matrix set with different characteristics, and a preset number of daily power generation data matrices are selected from each target clustering matrix set, and the historical power generation data in the daily power generation data matrix is ​​used as a wind and solar power generation data set, wherein the wind and solar power generation data set includes the power generation power data of the wind power generation energy storage system and the photovoltaic power generation energy storage system.

[0095] In the embodiment of the present invention, refer to Figure 2 As shown, the data optimization of the wind and solar power generation data set to obtain the target power generation data set includes:

[0096] S21, generating a wind power data sequence and a photovoltaic data sequence according to the wind and solar power generation data set;

[0097] S22, performing feature decomposition on the wind power data sequence and the photovoltaic data sequence to obtain an intrinsic modal component and a residual component;

[0098] S23, performing component inspection on the intrinsic modal component to obtain a high-frequency component and a low-frequency component, and performing wavelet threshold denoising on the high-frequency component to obtain a denoised high-frequency component;

[0099] S24, superimposing the denoised high-frequency component, the low-frequency component and the residual component to obtain a target power generation data set.

[0100] In the embodiment of the present invention, the power generation data of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system in the photovoltaic power generation data set are sorted by time to obtain a wind power data sequence and a photovoltaic data sequence.

[0101] Furthermore, CEEMDAN (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) is performed on the wind power data sequence and the photovoltaic data sequence. A series of white noise sequences with different amplitudes are added to the wind power data sequence and the photovoltaic data sequence, and then empirical mode decomposition is performed to obtain a set of intrinsic mode functions, i.e., multiple intrinsic mode components, and the remaining signal after removing all identifiable intrinsic mode components, i.e., the residual component, which may contain non-periodic trends or noise that cannot be further decomposed.

[0102] In detail, a t-test can be performed on multiple intrinsic modal components, and the multiple intrinsic modal components can be further divided into high-frequency components and low-frequency components. The high-frequency components are denoised using a preset wavelet threshold algorithm, which can effectively improve the accuracy of the high-frequency components. The denoised high-frequency components, low-frequency components and residual components after denoising are superimposed to reconstruct the wind power data sequence and the photovoltaic data sequence to obtain a more accurate power generation data signal, that is, a target power generation data set.

[0103] In the embodiment of the present invention, the characteristics of the wind and solar power generation data set can be optimized through data optimization to obtain a more accurate target power generation data set, which helps to improve the accuracy of subsequent power prediction model parameter optimization and obtain a more accurate target prediction model.

[0104] S3. Optimize the parameters of the pre-built power prediction model according to the target power generation data set to obtain a target prediction model.

[0105] In an embodiment of the present invention, the pre-built power prediction model can be a neural model composed of BiLSTM (Bidirectional LongShort-Term Memory) and an attention mechanism, which can enable the power prediction model to memorize historical data information for a long time while also being able to determine important information in the data, thereby improving the prediction accuracy of the power prediction model.

[0106] In the embodiment of the present invention, refer to Figure 3 As shown, the parameter optimization of the pre-built power prediction model is performed according to the target power generation data set to obtain the target prediction model, including:

[0107] S31, obtaining initial parameters of the power prediction model, generating an initial particle population according to the initial parameters; performing normalization processing on the initial particle population to obtain a target particle population;

[0108] S32, performing chaos mapping on the target particle population to obtain a mapped population, and calculating a prediction error value of the target power generation data set according to the mapped population;

[0109] S33. Perform population optimization on the mapping population according to the prediction error value to obtain target parameters, and generate a target prediction model according to the target parameters.

[0110] In an embodiment of the present invention, the initial parameters are model parameters in a pre-constructed power prediction model. The initial parameters are expanded within a preset range of the initial parameters to obtain an initial particle population. The parameter combination in the initial particle population is normalized so that the range of the parameter combination is (0, 1), and a target particle population is subjected to chaotic mapping, which can make the distribution of the target particle population more uniform and improve the search speed and global optimization capability of a subsequent population optimization algorithm.

[0111] Furthermore, the target particle population is chaotically mapped using the following formula to obtain a mapped population:

[0112]

[0113] Among them, B i+1 Represents the i-th target particle B in the target particle population i Corresponding mapping particles, rand(0,1) represents a random number in [0,1], and N represents the total number of target particles in the target particle population.

[0114] In an embodiment of the present invention, the prediction error value of the power prediction model when predicting the power of the target power generation data set is calculated by mapping the parameter combination in the population, and the minimum prediction error value is used as the fitness function for population optimization of the mapping population, and then the particle population after population optimization that meets the prediction error value requirement is calculated, and the parameter combination with the minimum prediction error value is selected as the target parameter in the particle population after population optimization to obtain the target prediction model.

[0115] Among them, the prediction error value of the target power generation data set can be calculated using error functions such as the root mean square error or logarithmic error, and then the parameters of the power prediction model can be optimized to obtain a target prediction model with more accurate parameters, thereby improving the accuracy of subsequent predictions of wind power series and photovoltaic power series.

[0116] S4. Calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence.

[0117] In an embodiment of the present invention, the output power of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system of the target wind-solar power generation and energy storage system within a preset time period is calculated by a target prediction model, for example, the predicted wind power and predicted photovoltaic power per hour within 24 hours, so as to further optimize the power of the target wind-solar power generation and energy storage system in real time.

[0118] In an embodiment of the present invention, the method of calculating the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system by using the target prediction model includes:

[0119] Acquire the historical data and weather information to be predicted of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system in the target wind-solar power generation and energy storage system;

[0120] Utilizing the bidirectional long short-term memory network in the target prediction model to calculate the data characteristics of the historical data to be predicted and the weather characteristics of the weather information;

[0121] The feature weights of the data feature and the weather feature are calculated respectively, and the data feature and the weather feature are weighted summed using the feature weights to obtain the target feature;

[0122] The target features are fully connected to obtain a predicted wind power sequence and a predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system.

[0123] In an embodiment of the present invention, the historical data to be predicted may be the historical power generation data of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system in the latest time period. For example, the power generation data of the latest 24 hours is used as the historical data to be predicted to ensure the real-time nature of the power prediction. The weather information is the wind information of the environment where the wind power generation and energy storage system is located and the light information of the environment where the photovoltaic power generation and energy storage system is located, which can be obtained through the meteorological information released by the state.

[0124] Furthermore, the BiLSTM (Bidirectional Long Short-Term Memory) in the target prediction model can be used to extract characteristic information from the historical data and weather information to be predicted, and perform selective memory and forgetting, so as to extract the intrinsic connection and time series change law between the data, and realize the accurate prediction of future predicted wind power series and predicted photovoltaic power series.

[0125] Furthermore, the attention mechanism in the target prediction model can calculate the different weights corresponding to the data features and weather features, selectively learn key feature information, and further explore the deep feature information in the historical data and weather information to be predicted through the obtained target features.

[0126] Specifically, the feature weights of the data features and the weather features are calculated using the following formula:

[0127] e t =ρtanh(θh t +B)

[0128] Among them, μ t Represents the tth data feature or weather feature e t , ρ and θ represent the preset first weight coefficient and the second weight coefficient, and B represents the preset bias coefficient.

[0129] In an embodiment of the present invention, the target features can be activated and budgeted through full connection to obtain the predicted power probability corresponding to the target wind-solar power generation and energy storage system at multiple prediction time points, and the predicted power corresponding to the maximum value of the predicted power probability at each prediction time point is selected as the predicted wind power and predicted photovoltaic power at the prediction time point, thereby obtaining the predicted wind power sequence and the predicted photovoltaic power sequence.

[0130] In the embodiment of the present invention, the power change probability distribution is to calculate the fluctuation probability of the power output of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system within the prediction time scale. If the power change probability distribution does not meet the requirements, the power optimization adjustment of the target wind and photovoltaic power generation and energy storage system is further performed to improve the accuracy of the distributed power optimization of the wind and photovoltaic power generation and energy storage system.

[0131] In the embodiment of the present invention, the calculating of the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence includes:

[0132] Calculating the maximum power fluctuation of the predicted wind power sequence and the predicted photovoltaic power sequence respectively;

[0133] The power variation probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence is calculated according to the maximum power fluctuation amount.

[0134] Specifically, the maximum power fluctuation is calculated using the following formula:

[0135] ΔP T =maxP(τ)-minP(τ),τ∈[HT,H)

[0136] Where ΔP T represents the maximum power fluctuation, maxP(τ) represents the maximum predicted wind power and the maximum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the preset time scale T, minP(τ) represents the minimum predicted wind power and the minimum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the time scale T, and H represents the total time scale of the predicted wind power sequence and the predicted photovoltaic power sequence.

[0137] Specifically, the predicted wind power sequence and the predicted photovoltaic power sequence are predicted powers every hour or every half hour, then T can be every hour, every half hour, every two hours, etc., and H represents the total time of the predicted wind power sequence and the predicted photovoltaic power sequence, for example, 24 hours, 36 hours, 48 ​​hours, etc.

[0138] Furthermore, the difference between two adjacent predicted powers in the predicted wind power sequence and the predicted photovoltaic power sequence can be taken as the adjacent fluctuation value, and the ratio of the number of adjacent fluctuation values ​​equal to the maximum fluctuation amount to the number of adjacent fluctuation values ​​can be taken as the power change probability distribution.

[0139] S5. Power-regulate the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

[0140] In the embodiment of the present invention, power regulation is to ensure that the power fluctuations of the predicted wind power sequence and the predicted photovoltaic power sequence at different time scales do not exceed a threshold, and to regulate the power exceeding the threshold to obtain a more accurate output power.

[0141] In an embodiment of the present invention, the power adjustment of the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system includes:

[0142] Determining whether the power change probability distribution is greater than a preset probability threshold;

[0143] When the power change probability distribution is less than or equal to the probability threshold, using the wind power sequence and the predicted photovoltaic power sequence as optimized output power;

[0144] When the power change probability distribution step is greater than the probability threshold, power distribution optimization is performed on the wind power sequence and the predicted photovoltaic power sequence until the power change probability distribution is less than or equal to or greater than the probability threshold, thereby obtaining the optimized output power.

[0145] In an embodiment of the present invention, when the power change probability distribution is greater than the probability threshold, the predicted power in the wind power sequence and the predicted photovoltaic power sequence is adjusted, for example, the larger predicted power in adjacent fluctuation values ​​is lowered, or other energy storage devices are called to supplement the smaller predicted power in adjacent fluctuation values, and the output power is complemented by wind power and photovoltaic power to meet the output requirements of the target wind-solar power generation and energy storage system, avoid the impact of power fluctuations of the target wind-solar power generation and energy storage system on the safety of the power generation system, and further improve the accuracy of distributed power optimization of the wind-solar power generation and energy storage system.

[0146] like Figure 4 , which is a functional module diagram of a distributed power optimization system of a wind-solar power generation and energy storage system provided by an embodiment of the present invention.

[0147] The distributed power optimization system 400 of the wind-solar power generation and energy storage system of the present invention can be installed in an electronic device. According to the functions to be implemented, the distributed power optimization system 400 of the wind-solar power generation and energy storage system can include a daily power generation correlation coefficient calculation module 401, a data optimization module 402, a model parameter optimization module 403, a power change probability distribution calculation module 404 and a power regulation module 405. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor 501 and can complete fixed functions, which are stored in the memory 502 of the electronic device.

[0148] In this embodiment, the functions of each module / unit are as follows:

[0149] The daily power generation correlation coefficient calculation module 401 is used to obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between historical power generation data in the historical power generation data set;

[0150] The data optimization module 402 is used to perform data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and perform data optimization on the wind-solar power generation data set to obtain a target power generation data set;

[0151] The model parameter optimization module 403 is used to optimize the parameters of the pre-built power prediction model according to the target power generation data set to obtain a target prediction model;

[0152] The power change probability distribution calculation module 404 is used to calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence;

[0153] The power regulation module 405 is used to perform power regulation on the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

[0154] In detail, each module described in the distributed power optimization system 400 of the wind-solar power generation energy storage system described in the embodiment of the present invention is used in the same manner as described above. Figures 1 to 3 The distributed power optimization method of wind-solar power generation and energy storage system described in the invention has the same technical means and can produce the same technical effects, so it will not be repeated here.

[0155] like Figure 5 As shown, the present invention also provides an electronic device 500, which may include a processor 501, a memory 502, a communication bus 503 and a communication interface 504, and may also include a computer program stored in the memory 502 and executable on the processor 501, such as a distributed power optimization method program for a wind-solar power generation and energy storage system.

[0156] Among them, the processor 501 can be composed of an integrated circuit in some embodiments, for example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits packaged with the same function or different functions, including one or more central processing units 501 (Central Processing Unit, CPU), microprocessors 501, digital processing chips, graphics processors 501 and a combination of various control chips, etc.

[0157] The memory 502 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory 502 (for example, an SD or DX memory 502, etc.), a magnetic memory 502, a disk, an optical disk, etc. In some embodiments, the memory 502 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device.

[0158] The communication bus 503 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 502 and at least one processor 501, etc.

[0159] The communication interface 504 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices.

[0160] Figure 5 Only an electronic device with components is shown. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0161] For example, although not shown, the electronic device 500 may also include a power supply (such as a battery) for supplying power to each component. Preferably, the power supply may be logically connected to the at least one processor 501 through a power management system, so that the power management system can realize functions such as charging management, discharging management, and power consumption management. The power supply may also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0162] Specifically, the specific implementation method of the processor 501 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0163] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0164] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0166] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0167] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0168] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0169] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A distributed power optimization method for a wind-solar power generation and energy storage system, characterized in that: The method comprises: Obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between the historical power generation data in the historical power generation data set; Performing data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and performing data optimization on the wind-solar power generation data set to obtain a target power generation data set; Optimizing parameters of a pre-built power prediction model according to the target power generation data set to obtain a target prediction model; Calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system by using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence; The wind power sequence and the predicted photovoltaic power sequence are power-regulated according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

2. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The calculating of the daily power generation correlation coefficient between the historical power generation data in the historical power generation data set includes: Constructing a daily power generation data matrix of the target wind-solar power generation and energy storage system according to the historical power generation data set; Decentralizing the daily power generation data matrix to obtain a centralized matrix, and calculating a semi-positive definite matrix of the centralized matrix; Calculate the daily power generation correlation coefficient between the daily power generation data matrices according to the semi-positive definite matrix; The daily power generation correlation coefficient between the historical power generation data in the historical power generation data set is calculated using the following formula: Among them, RV represents the daily power generation correlation coefficient, V x represents the semi-positive definite matrix corresponding to the power generation data matrix of the xth day, T represents the matrix transpose, V y It represents the semi-positive definite matrix corresponding to the power generation data matrix of the yth day, and tr() represents the trace of the matrix.

3. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The data clustering of the historical power generation data set according to the daily power generation correlation coefficient to obtain the wind and solar power generation data set includes: Selecting a preset number of cluster centers from the daily power generation data matrix of the historical power generation data set; Performing initial clustering on the daily power generation data matrix according to the daily power generation correlation coefficient and the cluster center to obtain an initial clustering data matrix set; Calculating a mean matrix of the initial clustering data matrix set, and iterating the cluster centers according to the mean matrix to obtain updated cluster centers; Calculating an updated daily power generation correlation coefficient between the updated cluster center and the daily power generation data matrix, and determining a target cluster center according to the updated daily power generation correlation coefficient; Clustering the daily power generation data matrix according to the target cluster center to obtain a target cluster matrix set; A wind-solar power generation data set is extracted from the historical power generation data set according to the target clustering matrix set.

4. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The step of optimizing the wind and solar power generation data set to obtain a target power generation data set includes: Generate a wind power data sequence and a photovoltaic data sequence according to the wind and solar power generation data set; Performing feature decomposition on the wind power data sequence and the photovoltaic data sequence to obtain intrinsic modal components and residual components; Performing component inspection on the inherent modal component to obtain a high-frequency component and a low-frequency component, and performing wavelet threshold denoising on the high-frequency component to obtain a denoised high-frequency component; The denoised high-frequency component, the low-frequency component and the residual component are superimposed to obtain a target power generation data set.

5. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The step of optimizing parameters of a pre-built power prediction model according to the target power generation data set to obtain a target prediction model includes: Acquiring initial parameters of the power prediction model, generating an initial particle population according to the initial parameters; performing normalization processing on the initial particle population to obtain a target particle population; Performing chaotic mapping on the target particle population to obtain a mapped population, and calculating a prediction error value of the target power generation data set according to the mapped population; The mapping population is optimized according to the prediction error value to obtain target parameters, and a target prediction model is generated according to the target parameters.

6. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 5, characterized in that: The step of performing chaotic mapping on the target particle population to obtain a mapped population includes: The target particle population is chaotically mapped using the following formula to obtain a mapped population: Among them, B i+1 Represents the i-th target particle B in the target particle population i Corresponding mapping particles, rand(0,1) represents a random number in [0,1], and N represents the total number of target particles in the target particle population.

7. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The method of calculating the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system by using the target prediction model includes: Acquire the historical data and weather information to be predicted of the wind power generation and energy storage system and the photovoltaic power generation and energy storage system in the target wind-solar power generation and energy storage system; Utilizing the bidirectional long short-term memory network in the target prediction model to calculate the data characteristics of the historical data to be predicted and the weather characteristics of the weather information; The feature weights of the data feature and the weather feature are calculated respectively, and the data feature and the weather feature are weighted summed using the feature weights to obtain the target feature; The feature weights of the data features and the weather features are calculated using the following formula: Among them, μ t Represents the tth data feature or weather feature e t The characteristic weight of , ρ, θ represent the preset first weight coefficient and the second weight coefficient, and B represents the preset bias coefficient; The target features are fully connected to obtain a predicted wind power sequence and a predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system.

8. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The calculating of the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence includes: Calculating the maximum power fluctuation of the predicted wind power sequence and the predicted photovoltaic power sequence respectively; The maximum power fluctuation is calculated using the following formula: ΔP T =maxP(τ)-minP(τ),τ∈[HT,H) Among them, ΔP T represents the maximum power fluctuation, maxP(τ) represents the maximum predicted wind power and the maximum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the preset time scale T, minP(τ) represents the minimum predicted wind power and the minimum predicted photovoltaic power of the predicted wind power sequence and the predicted photovoltaic power sequence within the time scale T, and H represents the total time scale of the predicted wind power sequence and the predicted photovoltaic power sequence; The power variation probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence is calculated according to the maximum power fluctuation amount.

9. The distributed power optimization method of the wind-solar power generation and energy storage system according to claim 1, characterized in that: The power regulation of the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system includes: Determining whether the power change probability distribution is greater than a preset probability threshold; When the power change probability distribution is less than or equal to the probability threshold, using the wind power sequence and the predicted photovoltaic power sequence as optimized output power; When the power change probability distribution step is greater than the probability threshold, power distribution optimization is performed on the wind power sequence and the predicted photovoltaic power sequence until the power change probability distribution is less than or equal to or greater than the probability threshold, thereby obtaining the optimized output power.

10. A distributed power optimization system for a wind-solar power generation and energy storage system, characterized in that: The system comprises: A daily power generation correlation coefficient calculation module is used to obtain a historical power generation data set of a target wind-solar power generation and energy storage system, and calculate a daily power generation correlation coefficient between historical power generation data in the historical power generation data set; A data optimization module, configured to perform data clustering on the historical power generation data set according to the daily power generation correlation coefficient to obtain a wind-solar power generation data set, and perform data optimization on the wind-solar power generation data set to obtain a target power generation data set; A model parameter optimization module, used to optimize the parameters of the pre-built power prediction model according to the target power generation data set to obtain a target prediction model; A power change probability distribution calculation module is used to calculate the predicted wind power sequence and predicted photovoltaic power sequence of the target wind-solar power generation and energy storage system using the target prediction model, and calculate the power change probability distribution of the predicted wind power sequence and the predicted photovoltaic power sequence; A power regulation module is used to perform power regulation on the wind power sequence and the predicted photovoltaic power sequence according to the power change probability distribution to obtain the optimized output power of the target wind-solar power generation and energy storage system.

Images