A new energy storage configuration method and system

Abstract

The present invention belongs to the field of new energy technology, and discloses a new energy storage configuration method and system. The method comprises the following steps: constructing a regional integrated new energy generation and storage distribution map; classifying several new energy storage systems in the new energy generation and storage distribution map; collecting real-time operation data of new energy power stations, and using a power generation prediction model to predict power generation; constructing a consumption target function of the new energy power station and a cost target function of the attached energy storage cluster; according to the consumption target function and the corresponding cost target function, using a new energy storage configuration model, performing new energy storage configuration, and obtaining a new energy storage configuration plan; visually displaying several new energy storage configuration plans in combination with the new energy generation and storage distribution map. The present invention solves the problems of poor energy storage configuration effect, low accuracy, lack of value of the plan, poor real-time performance, and poor timeliness in the prior art.

Description

Technical Field

[0001] The present invention belongs to the field of new energy technology, and specifically relates to a new energy storage configuration method and system. Background Art

[0002] At present, global energy shortages and environmental pollution are becoming increasingly serious. With the development of the energy revolution, the proportion of clean energy used by loads is increasing. New energy power generation has the characteristics of intermittency, randomness, and uncertainty, which brings new challenges to the operation of the power grid. Among them, with its fast power regulation and flexible energy management characteristics, the new energy storage system has shown great potential in smoothing the output fluctuations of new energy power stations, improving voltage quality, improving power supply reliability, and optimizing system energy management. Therefore, studying the energy storage configuration of the energy storage system is of great research significance for the planning and operation of new energy power stations.

[0003] In the prior art, the energy storage configuration of new energy power stations is mostly analyzed based on a single power station. In actual use, the new energy storage system often provides energy storage services for multiple new energy power stations, resulting in poor existing energy storage configuration effects and low accuracy, which cannot meet the energy storage needs of new energy power stations in multiple regions. In addition, there are information islands between new energy power stations and new energy storage systems in different regions, and there is a lack of information interaction, which leads to the lack of value of energy storage configuration plans and the inability to provide guidance for subsequent energy storage configurations. Finally, the existing energy storage configurations are mostly based on the historical power generation data of the previous quarter, and the real-time performance of the plans is poor, and it is impossible to make timely energy storage configuration adjustments according to real-time grid fluctuations. Summary of the invention

[0004] In order to solve the problems of poor energy storage configuration effect, low accuracy, lack of solution value, poor real-time performance and poor timeliness in the prior art, the present invention aims to provide a new energy storage configuration method and system.

[0005] The technical solution adopted by the present invention is:

[0006] A new energy storage configuration method comprises the following steps:

[0007] Based on the location data of renewable energy power stations, renewable energy storage systems and transmission lines in several regions, a regional integrated renewable energy generation-storage distribution map is constructed;

[0008] Classify the several new energy storage systems in the new energy generation-storage distribution diagram to obtain the subsidiary energy storage cluster of each new energy power station;

[0009] Collect the real-time operation data of the new energy power station, and use the power generation prediction model to predict the power generation according to the real-time operation data to obtain the power generation prediction data of the new energy power station;

[0010] According to the power generation forecast data, the consumption objective function of the new energy power station is constructed, and according to the system cost data of the new energy storage system, the cost objective function of the corresponding subsidiary energy storage cluster is constructed;

[0011] According to the consumption objective function and the corresponding cost objective function, a new energy storage configuration model is used to configure new energy storage and obtain a new energy storage configuration plan;

[0012] Traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration plans, and visualize the several new energy storage configuration plans in combination with the new energy generation-storage distribution map.

[0013] Furthermore, the new energy power station includes wind power station, photovoltaic power station and wind-solar complementary power station;

[0014] A wind power station includes several wind turbines, a photovoltaic power station includes several photovoltaic units, and a wind-solar hybrid power station includes several wind turbines and several photovoltaic units;

[0015] New energy storage systems include battery energy storage systems, power battery energy storage systems, compressed air energy storage systems, supercapacitor energy storage systems and thermal energy storage systems.

[0016] Furthermore, based on the location data of renewable energy power stations, renewable energy storage systems and transmission lines in several regions, a regional integrated renewable energy generation-storage distribution map is constructed, including the following steps:

[0017] According to the location data of renewable energy power stations in several regions, several master nodes are set up;

[0018] According to the location data of the new energy storage systems in several areas, several secondary nodes are set;

[0019] According to the location data of the power transmission lines in several regions, several discrete nodes are set, and according to the primary nodes / secondary nodes connected at both ends of the power transmission lines, several discrete nodes of the same power transmission line are connected and smoothed to obtain several edges;

[0020] Based on several main nodes, several secondary nodes and several edges, a regional integrated new energy generation-storage distribution map is constructed.

[0021] Furthermore, the several new energy storage systems in the new energy generation-storage distribution diagram are classified and assigned to obtain the subsidiary energy storage cluster of each new energy power station, including the following steps:

[0022] Based on the new energy generation-storage distribution map, several main nodes are used as several cluster centers, and the Euclidean distances from several secondary nodes to several cluster centers are obtained;

[0023] If the Euclidean distance from the current sub-node to any cluster center is the shortest, and there is a corresponding edge between the sub-node and the cluster center, the current sub-node is divided into the corresponding cluster center. Otherwise, find the next cluster center with a shorter Euclidean distance and repeat this step.

[0024] Traverse all secondary nodes to obtain the affiliated energy storage cluster of each new energy power station.

[0025] Furthermore, the power generation prediction model is constructed based on the CNN-BiGRU-Attention-KELM algorithm, and the power generation prediction model includes an input layer connected in sequence, a spatial feature extraction layer constructed based on the CNN algorithm, a spatial feature learning layer constructed based on the BiGRU algorithm, an attention weight layer constructed based on the Attention algorithm, an error prediction layer constructed based on the KELM algorithm, and an output layer;

[0026] The method for constructing a power generation prediction model includes the following steps:

[0027] Collect historical operation data sets of renewable energy power stations, use improved grey correlation analysis method to cluster historical operation data sets, and obtain similar data sets under different unit configurations;

[0028] Use PCA analysis method to reduce the dimension of similar data sets under different unit configurations, and obtain historical model training data sets under different unit configurations;

[0029] According to the historical model training data set, the CNN-BiGRU-Attention-KELM algorithm is used to build a power generation prediction model.

[0030] Furthermore, real-time operation data of the new energy power station is collected, and based on the real-time operation data, a power generation prediction model is used to perform power generation prediction to obtain power generation prediction data of the new energy power station, including the following steps:

[0031] Collect the real-time operation data of the new energy power station, and use the PCA analysis method to reduce the data dimension of the real-time operation data under the current unit configuration to obtain the real-time operation key data, and input the real-time operation key data into the input layer of the power generation prediction model;

[0032] Use the spatial feature extraction layer to extract the spatial features of key data in real-time operation, and input the spatial features into the feature learning layer;

[0033] Use the attention weight layer to set the output attention weight value of the feature learning layer;

[0034] Based on the output attention weight values, the feature learning layer is used to obtain the original prediction values ​​of the key data for real-time operation;

[0035] Obtain the original error value between the original predicted value and the true value, and use the error prediction layer to correct the original error value to obtain the predicted error value;

[0036] According to the original prediction value and prediction error value of the real-time operation key data, the output layer is used to output the final prediction value, that is, the power generation prediction data of the new energy power station.

[0037] Furthermore, the formula of the consumption objective function of the new energy power station is:

[0038]

[0039] Where f1 is the consumption objective function of the new energy power station; P wt,i is the power generation prediction data of wind turbine i; P pv,j is the power generation prediction data of photovoltaic unit j; i is the wind turbine unit indication value of the new energy power station; j is the photovoltaic unit indication value of the new energy power station.

[0040] Furthermore, the formula of the cost objective function of the attached energy storage cluster is:

[0041]

[0042] Where f2 is the cost objective function of the attached energy storage cluster; F in,k is the investment cost of the new energy storage system k; c is the equal-year coefficient; F on,k is the operation and maintenance cost of the new energy storage system k; F ex,k is the capacity expansion cost of the new energy storage system k; α is the capacity expansion ratio coefficient of the new energy storage system k; k is the indicator of the new energy storage system; K is the total number of new energy storage systems in the affiliated energy storage cluster.

[0043] Furthermore, the new energy storage configuration model is constructed based on the IFWA optimization algorithm;

[0044] According to the consumption objective function and the corresponding cost objective function, the new energy storage configuration model is used to configure the new energy storage and obtain the new energy storage configuration plan, including the following steps:

[0045] According to the consumption objective function of the new energy power station and the cost objective function of the attached energy storage cluster, the optimization objective of the IFWA optimization algorithm is set;

[0046] The new energy storage configuration scheme is used as the position of the IFWA individual, and the fitness function of the IFWA individual is set;

[0047] Set the IFWA population parameters and maximum number of iterations of the IFWA optimization algorithm;

[0048] According to the IFWA population parameters, the IFWA population is initialized using the Tent-Logistic-Cosine chaotic mapping sequence to obtain the initialized IFWA population;

[0049] According to the fitness function, the fitness value of each IFWA individual in the initialized IFWA population is obtained;

[0050] Get the explosion radius and number of sparks of each IFWA individual in the initialized IFWA population;

[0051] According to the explosion radius and the number of sparks of each initial IFWA individual in the initialized IFWA population, fireworks are exploded to obtain an updated IFWA population;

[0052] Use the Gaussian mutation algorithm to perform Gaussian mutation on the initialized IFWA population to generate a Gaussian mutated IFWA population;

[0053] Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the initialized IFWA population to obtain the reverse IFWA population;

[0054] Obtain the fitness values ​​of the IFWA individuals of the updated IFWA population, the Gaussian-mutated IFWA population, and the reverse IFWA population, and take the IFWA individual with the minimum fitness value as the optimal individual;

[0055] If the number of iterations reaches the threshold or the fitness value of the optimal individual meets the requirements, the optimal solution corresponding to the current optimal individual is output to obtain the optimal new energy storage configuration plan.

[0056] A new energy storage configuration system is used to implement a new energy storage configuration method, the system includes a distribution map construction unit, an energy storage system attribution division unit, a power generation prediction unit, an objective function construction unit, a new energy storage configuration unit and an energy storage configuration scheme visualization unit connected in sequence;

[0057] A distribution map construction unit, used to construct a regional integrated new energy generation-storage distribution map based on the location data of new energy power stations, new energy storage systems and transmission lines in several regions;

[0058] The energy storage system attribution division unit is used to classify a number of new energy storage systems in the new energy generation-storage distribution diagram to obtain the affiliated energy storage clusters of each new energy power station;

[0059] The power generation prediction unit is used to collect the real-time operation data of the new energy power station, and use the power generation prediction model to perform power generation prediction based on the real-time operation data to obtain the power generation prediction data of the new energy power station;

[0060] An objective function construction unit is used to construct an absorption objective function of a new energy power station according to power generation prediction data, and to construct a cost objective function of a corresponding subsidiary energy storage cluster according to system cost data of a new energy energy storage system;

[0061] A new energy storage configuration unit is used to configure new energy storage according to a consumption objective function and a corresponding cost objective function using a new energy storage configuration model to obtain a new energy storage configuration plan;

[0062] The energy storage configuration scheme visualization unit is used to traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration schemes, and visualize the several new energy storage configuration schemes in combination with the new energy generation-storage distribution map.

[0063] The beneficial effects of the present invention are:

[0064] The invention provides a new energy storage configuration method and system. The method and system construct a new energy generation-storage distribution map through location data, strengthen the spatial analysis of the new energy storage configuration, and the regional integrated new energy generation-storage distribution map strengthens the information interaction between new energy power stations and new energy storage systems in different regions, avoids information islands, improves the value of the energy storage configuration plan, and provides guidance for subsequent energy storage configuration. Based on the new energy generation-storage distribution map, the new energy storage system is classified and assigned, and the principle of geographical priority and the nearest configuration with minimum line loss is adopted to reduce damage and improve the effect and accuracy of the energy storage configuration. The power generation prediction model is used to perform short-term power generation prediction, and real-time data is used to provide data support for the subsequent new energy storage configuration, thereby improving the real-time performance and being able to adjust the energy storage configuration in time according to the real-time power grid fluctuation. The consumption target of the new energy power station and the cost target function of the attached energy storage cluster are considered, the economic benefit and voltage quality are taken into account, and the practicality and economy of the new energy storage configuration are improved. A number of new energy storage configuration plans are combined with the new energy generation-storage distribution map for visual display, and the new energy storage configuration results are intuitively and clearly displayed, thereby improving the practicality and implementation efficiency of the plan.

[0065] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a flowchart of the new energy storage configuration method in the present invention.

[0067] Figure 2 It is a structural block diagram of the new energy storage configuration system in the present invention. DETAILED DESCRIPTION

[0068] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0069] Embodiment 1:

[0070] like Figure 1 As shown, this embodiment provides a new energy storage configuration method, including the following steps:

[0071] Based on the location data of renewable energy power stations, renewable energy storage systems and transmission lines in several regions, a regional integrated renewable energy generation-storage distribution map is constructed, including the following steps:

[0072] According to the location data of renewable energy power stations in several regions, several master nodes are set up;

[0073] According to the location data of the new energy storage systems in several areas, several secondary nodes are set;

[0074] According to the location data of the power transmission lines in several regions, several discrete nodes are set, and according to the primary nodes / secondary nodes connected at both ends of the power transmission lines, several discrete nodes of the same power transmission line are connected and smoothed to obtain several edges;

[0075] Based on several main nodes, several secondary nodes and several edges, a regional integrated new energy generation-storage distribution map is constructed;

[0076] Eliminate geographical and administrative barriers between different regions, strengthen information exchange, and display the new energy power stations, new energy storage systems and transmission lines in several regions in a two-dimensional and intuitive way from the perspective of spatial location, providing support for subsequent attribution division;

[0077] New energy power stations include wind power stations, photovoltaic power stations and wind-solar hybrid power stations;

[0078] A wind power station includes several wind turbines, a photovoltaic power station includes several photovoltaic units, and a wind-solar hybrid power station includes several wind turbines and several photovoltaic units;

[0079] New energy storage systems include battery energy storage systems, power battery energy storage systems, compressed air energy storage systems, supercapacitor energy storage systems, and thermal energy storage systems;

[0080] It eliminates the barriers between the types of power generation and storage in the configuration of new energy storage, analyzes from a macro perspective, and improves the efficiency and effect of the configuration of new energy storage;

[0081] The following steps are included to classify several new energy storage systems in the new energy generation-storage distribution diagram to obtain the subsidiary energy storage cluster of each new energy power station:

[0082] Based on the new energy generation-storage distribution map, several main nodes are used as several cluster centers, and the Euclidean distances from several secondary nodes to several cluster centers are obtained;

[0083] If the Euclidean distance from the current sub-node to any cluster center is the shortest, and there is a corresponding edge between the sub-node and the cluster center, the current sub-node is divided into the corresponding cluster center. Otherwise, find the next cluster center with a shorter Euclidean distance and repeat this step.

[0084] Traverse all secondary nodes to obtain the affiliated energy storage cluster of each new energy power station;

[0085] The subsidiary energy storage cluster is set up according to the principle of nearby energy storage, which avoids power loss caused by long-distance transmission, avoids data waste, and improves energy storage efficiency;

[0086] A power generation prediction model is constructed based on the CNN-BiGRU-Attention-KELM algorithm. The power generation prediction model includes a sequentially connected input layer, a spatial feature extraction layer constructed based on the Convolutional Neural Network (CNN) algorithm, a spatial feature learning layer constructed based on the Bidirectional Recurrent Neural Network (BiGRU) algorithm, an attention weight layer constructed based on the Attention algorithm, an error prediction layer constructed based on the Kernel Based Extreme Learning Machine (KELM) algorithm, and an output layer;

[0087] The spatial feature extraction layer is mainly used to extract spatial features from the running data. It consists of two convolutional layers and one pooling layer. Due to the characteristics of the running data, the two convolutional layers are designed as one-dimensional convolutions. The activation function of the spatial feature extraction layer is sigmoid, and the filling is the same. In order to save more fluctuation information, the pooling layer uses the maximum pooling.

[0088] The attention weight layer imitates the internal process of biological observation activities. It is a mechanism that imitates cognitive attention and quickly filters out high-value information from a large amount of information. Its core idea is to reasonably allocate weights, that is, to give greater weights to important information so as to more reasonably change the external attention to information, ignore irrelevant information, and amplify the required information. The Attention mechanism is introduced to give weights to the hidden layer output of the BiGRU network, reduce information loss caused by too long time series, and highlight the influence of strongly correlated features and reduce the influence of weakly correlated features.

[0089] The feature learning layer uses the forward and reverse BiGRU networks to learn the features extracted by the CNN network. By building a two-layer BiGRU structure, it fully learns the extracted features, captures the bidirectional information flow of the data set, and learns the dynamic changes of the running data.

[0090] The error prediction layer is an improved algorithm based on the Extreme Learning Machine (ELM) combined with the kernel function. KELM can improve the prediction performance of the model while retaining the advantages of ELM and optimize the prediction error. Compared with traditional neural networks, it can effectively avoid the generation of local optimal solutions and further improve the prediction accuracy.

[0091] The steps include:

[0092] The historical operation data set of the renewable energy power station is collected, and the improved grey correlation analysis method is used to cluster the historical operation data set to obtain similar data sets under different unit configurations, including the following steps:

[0093] Collect historical operation data sets of renewable energy power stations;

[0094] Construct a comparison data series X based on the historical running data set i' ={x i' (k'),i'=1,2,...,n-1,k'=1,2,...,m} and the reference data sequence X obj ={x obj (k'), k'=1,2,...,m}, where i' is the operating data indicator, n is the total number of operating data, k' is the operating parameter indicator, and m is the total number of operating parameters;

[0095] According to the comparison data sequence and the reference data sequence, the grey correlation coefficient of the comparison data sequence and the reference data sequence is obtained. The formula is:

[0096]

[0097] Δ i' (k)=|x obj (k')-xi' (k')|

[0098] In the formula, c i' (k') is the grey correlation coefficient; Δ i' (k') is the absolute difference; Δ i' (k') min , Δ i' (k') max is the extreme value of the absolute difference; ρ is the resolution coefficient;

[0099] Get the information entropy value of the comparison data sequence, the formula is:

[0100]

[0101]

[0102] In the formula, E i' is the information entropy value; B i' (k') is the comparison data x i' (k') is the percentage of the sum of the comparison data series;

[0103] According to the grey relational coefficient and information entropy value, the improved grey relational degree is obtained, and the formula is:

[0104]

[0105] In the formula, C i' To improve the grey relational degree;

[0106] According to ρ=0.5, C i >0.85 comparison data sequence, divide the historical operation data set to obtain similar data sets under different unit configurations;

[0107] Principal Component Analysis (PCA) was used to reduce the dimension of similar data sets under different unit configurations to obtain historical model training data sets under different unit configurations. The original redundant data was fitted into a low-dimensional coordinate system to establish a new data set, which not only retained the relative relationship of the data in the original space, but also improved the training speed of the model.

[0108] Based on the historical model training data set, the CNN-BiGRU-Attention-KELM algorithm is used to build a power generation prediction model. Compared with the prediction of a single neural network, the combined model is more comprehensive, realizes the complementary advantages of the models, further improves the prediction accuracy and model prediction effect, and the prediction error and error fluctuation range of the combined model are smaller than those of a single model.

[0109] Collecting real-time operation data of the new energy power station, and using the power generation prediction model to perform power generation prediction based on the real-time operation data to obtain power generation prediction data of the new energy power station, including the following steps:

[0110] Collect the real-time operation data of the new energy power station, and use the PCA analysis method to reduce the data dimension of the real-time operation data under the current unit configuration to obtain the real-time operation key data, and input the real-time operation key data into the input layer of the power generation prediction model;

[0111] Use the spatial feature extraction layer to extract the spatial features of key data in real-time operation, and input the spatial features into the feature learning layer;

[0112] Use the attention weight layer to set the output attention weight value of the feature learning layer;

[0113] Based on the output attention weight values, the feature learning layer is used to obtain the original prediction values ​​of the key data for real-time operation;

[0114] Obtain the original error value between the original predicted value and the true value, and use the error prediction layer to correct the original error value to obtain the predicted error value;

[0115] According to the original prediction value and prediction error value of the real-time operation key data, the output layer is used to output the final prediction value, that is, the power generation prediction data of the new energy power station;

[0116] According to the power generation forecast data, the consumption objective function of the new energy power station is constructed, and according to the system cost data of the new energy storage system, the cost objective function of the corresponding subsidiary energy storage cluster is constructed;

[0117] The formula of the consumption objective function of the new energy power station is:

[0118]

[0119] Where f1 is the consumption objective function of the new energy power station; P wt,i is the power generation prediction data of wind turbine i; P pv,j is the power generation prediction data of photovoltaic unit j; i is the wind turbine unit indication value of the new energy power station; j is the photovoltaic unit indication value of the new energy power station;

[0120] The formula of the cost objective function of the attached energy storage cluster is:

[0121]

[0122] Where f2 is the cost objective function of the attached energy storage cluster; F in,k is the investment cost of the new energy storage system k; c is the equal-year coefficient; F on,kis the operation and maintenance cost of the new energy storage system k; F ex,k is the capacity expansion cost of the new energy storage system k; α is the capacity expansion ratio coefficient of the new energy storage system k; k is the indicator of the new energy storage system; K is the total number of new energy storage systems in the affiliated energy storage cluster;

[0123] A new energy storage configuration model is constructed based on the Improved Fireworks Optimization Algorithm (IFWA);

[0124] According to the consumption objective function and the corresponding cost objective function, the new energy storage configuration model is used to configure the new energy storage, and the new energy storage configuration scheme includes the following steps:

[0125] According to the consumption objective function of the new energy power station and the cost objective function of the attached energy storage cluster, the optimization objective of the IFWA optimization algorithm is set as follows:

[0126]

[0127] In the formula, f s is the optimization objective function of the IFWA optimization algorithm; f1 is the consumption objective function of the new energy power station; f2 is the cost objective function of the attached energy storage cluster;

[0128] The new energy storage configuration scheme is used as the position of the IFWA individual, and the fitness function of the IFWA individual is set as follows:

[0129]

[0130] In the formula, f(q l' ) is the initial IFWA individual q l' The fitness value; MSE is the prediction mean square error function; Y' l' , Y l' are the predicted value and the true value; L is the total number of samples;

[0131] Set the IFWA population parameters and maximum number of iterations of the IFWA optimization algorithm;

[0132] According to the IFWA population parameters, the IFWA population is initialized using the Tent-Logistic-Cosine chaotic mapping sequence to obtain the initialized IFWA population;

[0133]

[0134] In the formula, q' i"The initial position of the IFWA individuals generated by the Tent-Logistic-Cosine chaotic mapping sequence; q' i is the initial position of the randomly generated IFWA individual; r is a preset parameter of 0-1; i" is the IFWA individual indicator;

[0135] According to the fitness function, the fitness value of each IFWA individual in the initialized IFWA population is obtained;

[0136] Get the explosion radius and number of sparks of each IFWA individual in the initialized IFWA population. The formula is:

[0137]

[0138] In the formula, S l' is the initial IFWA individual q l' The number of sparks; M' is a constant; f max is the maximum fitness value in the initialized IFWA population; f(q l' ) is the initial IFWA individual q l' The fitness value of ; τ is an infinitesimal constant;

[0139]

[0140] In the formula, R l' is the initial IFWA individual q l' Explosion radius; f is the explosion radius adjustment constant; min is the minimum fitness value in the initialized IFWA population;

[0141] According to the explosion radius and number of sparks of each initial IFWA individual in the initialized IFWA population, fireworks are exploded to obtain an updated IFWA population. The formula is:

[0142] q' l' =q l' +S l' ×rand(-1,1)

[0143] In the formula, q' l' is the updated IFWA individual; rand(-1,1) is a random number between -1 and 1;

[0144] Use the Gaussian mutation algorithm to perform Gaussian mutation on the initialized IFWA population to generate a Gaussian mutated IFWA population. The formula is:

[0145] q" l' =q l' +S l' ×G(1,1)

[0146] In the formula, q" l' is an IFWA individual with Gaussian variation; G(1,1) is a random number with Gaussian distribution whose mean and variance are both 1;

[0147] Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the initialized IFWA population to obtain the reverse IFWA population. The formula is:

[0148] q"' l' =γ(L max +L min )-q l'

[0149] In the formula, q"' l' is the reverse IFWA individual; γ is the decreasing inertia coefficient; L max , L min are the maximum and minimum values ​​of the vector space respectively;

[0150] Obtain the fitness values ​​of the IFWA individuals of the updated IFWA population, the Gaussian-mutated IFWA population, and the reverse IFWA population, and take the IFWA individual with the minimum fitness value as the optimal individual;

[0151] If the number of iterations reaches the threshold or the fitness value of the optimal individual meets the requirements, the optimal solution corresponding to the current optimal individual is output to obtain the optimal new energy storage configuration plan;

[0152] Traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration plans, and visualize the several new energy storage configuration plans in combination with the new energy generation-storage distribution map.

[0153] Embodiment 2:

[0154] like Figure 2 As shown, this embodiment provides a new energy storage configuration system for implementing a new energy storage configuration method, the system includes a distribution map construction unit, an energy storage system attribution division unit, a power generation prediction unit, an objective function construction unit, a new energy storage configuration unit, and an energy storage configuration scheme visualization unit connected in sequence;

[0155] A distribution map construction unit, used to construct a regional integrated new energy generation-storage distribution map based on the location data of new energy power stations, new energy storage systems and transmission lines in several regions;

[0156] The energy storage system attribution division unit is used to classify a number of new energy storage systems in the new energy generation-storage distribution diagram to obtain the affiliated energy storage clusters of each new energy power station;

[0157] The power generation prediction unit is used to collect the real-time operation data of the new energy power station, and use the power generation prediction model to perform power generation prediction based on the real-time operation data to obtain the power generation prediction data of the new energy power station;

[0158] An objective function construction unit is used to construct an absorption objective function of a new energy power station according to power generation prediction data, and to construct a cost objective function of a corresponding subsidiary energy storage cluster according to system cost data of a new energy energy storage system;

[0159] A new energy storage configuration unit is used to configure new energy storage according to a consumption objective function and a corresponding cost objective function using a new energy storage configuration model to obtain a new energy storage configuration plan;

[0160] The energy storage configuration scheme visualization unit is used to traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration schemes, and visualize the several new energy storage configuration schemes in combination with the new energy generation-storage distribution map.

[0161] The invention provides a new energy storage configuration method and system. The method and system construct a new energy generation-storage distribution map through location data, strengthen the spatial analysis of the new energy storage configuration, and the regional integrated new energy generation-storage distribution map strengthens the information interaction between new energy power stations and new energy storage systems in different regions, avoids information islands, improves the value of the energy storage configuration plan, and provides guidance for subsequent energy storage configuration. Based on the new energy generation-storage distribution map, the new energy storage system is classified and assigned, and the principle of geographical priority and the nearest configuration with minimum line loss is adopted to reduce damage and improve the effect and accuracy of the energy storage configuration. The power generation prediction model is used to perform short-term power generation prediction, and real-time data is used to provide data support for the subsequent new energy storage configuration, thereby improving the real-time performance and being able to adjust the energy storage configuration in time according to the real-time power grid fluctuation. The consumption target of the new energy power station and the cost target function of the attached energy storage cluster are considered, the economic benefit and voltage quality are taken into account, and the practicality and economy of the new energy storage configuration are improved. A number of new energy storage configuration plans are combined with the new energy generation-storage distribution map for visual display, and the new energy storage configuration results are intuitively and clearly displayed, thereby improving the practicality and implementation efficiency of the plan.

[0162] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. A new energy storage configuration method, characterized in that: The steps include: Based on the location data of renewable energy power stations, renewable energy storage systems and transmission lines in several regions, a regional integrated renewable energy generation-storage distribution map is constructed; Classify the several new energy storage systems in the new energy generation-storage distribution diagram to obtain the subsidiary energy storage cluster of each new energy power station; Collect the real-time operation data of the new energy power station, and use the power generation prediction model to predict the power generation according to the real-time operation data to obtain the power generation prediction data of the new energy power station; According to the power generation forecast data, the consumption objective function of the new energy power station is constructed, and according to the system cost data of the new energy storage system, the cost objective function of the corresponding subsidiary energy storage cluster is constructed; The formula of the consumption objective function of the new energy power station is: Where f1 is the consumption objective function of the new energy power station; P wt,i is the power generation prediction data of wind turbine i; P pv,j is the power generation prediction data of photovoltaic unit j; i is the wind turbine unit indication value of the new energy power station; j is the photovoltaic unit indication value of the new energy power station; The formula for the cost objective function of the attached energy storage cluster is: Where f2 is the cost objective function of the attached energy storage cluster; F in,k is the investment cost of the new energy storage system k; c is the equal-year coefficient; F on,k is the operation and maintenance cost of the new energy storage system k; F ex,k is the capacity expansion cost of the new energy storage system k; α is the capacity expansion ratio coefficient of the new energy storage system k; k is the indicator of the new energy storage system; K is the total number of new energy storage systems in the affiliated energy storage cluster; Construct a new energy storage configuration model based on the IFWA optimization algorithm; According to the consumption objective function and the corresponding cost objective function, the new energy storage configuration model is used to configure the new energy storage and obtain the new energy storage configuration plan, including the following steps: According to the consumption objective function of the new energy power station and the cost objective function of the attached energy storage cluster, the optimization objective of the IFWA optimization algorithm is set as follows: In the formula, f s is the optimization objective function of the IFWA optimization algorithm; f1 is the consumption objective function of the new energy power station; f2 is the cost objective function of the attached energy storage cluster; The new energy storage configuration scheme is used as the position of the IFWA individual, and the fitness function of the IFWA individual is set as follows: In the formula, f(q l' ) is the initial IFWA individual q l' The fitness value; MSE is the prediction mean square error function; Y' l' , Y l' are the predicted value and the true value; L is the total number of samples; Set the IFWA population parameters and maximum number of iterations of the IFWA optimization algorithm; According to the IFWA population parameters, the IFWA population is initialized using the Tent-Logistic-Cosine chaotic mapping sequence to obtain the initialized IFWA population; In the formula, q' i" The initial position of the IFWA individuals generated by the Tent-Logistic-Cosine chaotic mapping sequence; q' i is the initial position of the randomly generated IFWA individual; r is a preset parameter of 0-1; i" is the IFWA individual indicator; According to the fitness function, the fitness value of each IFWA individual in the initialized IFWA population is obtained; Get the explosion radius and number of sparks of each IFWA individual in the initialized IFWA population. The formula is: In the formula, S l' is the initial IFWA individual q l' The number of sparks; M' is a constant; f max is the maximum fitness value in the initialized IFWA population; f(q l' ) is the initial IFWA individual q l' The fitness value of ; τ is an infinitesimal constant; In the formula, R l' is the initial IFWA individual q l' Explosion radius; f is the explosion radius adjustment constant; min is the minimum fitness value in the initialized IFWA population; According to the explosion radius and number of sparks of each initial IFWA individual in the initialized IFWA population, fireworks are exploded to obtain an updated IFWA population. The formula is: q' l' =q l' +S l' ×rand(-1,1) In the formula, q' l' is the updated IFWA individual; rand(-1,1) is a random number between -1 and 1; Use the Gaussian mutation algorithm to perform Gaussian mutation on the initialized IFWA population to generate a Gaussian mutated IFWA population. The formula is: q" l' =q l' +S l' ×G(1,1) In the formula, q" l' is an IFWA individual with Gaussian variation; G(1,1) is a random number with Gaussian distribution whose mean and variance are both 1; Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the initialized IFWA population to obtain the reverse IFWA population. The formula is: what"' l' =γ(L max +L min )-q l' In the formula, q"' l' is the reverse IFWA individual; γ is the decreasing inertia coefficient; L max , L min are the maximum and minimum values ​​of the vector space respectively; Obtain the fitness values ​​of the IFWA individuals of the updated IFWA population, the Gaussian-mutated IFWA population, and the reverse IFWA population, and take the IFWA individual with the minimum fitness value as the optimal individual; If the number of iterations reaches the threshold or the fitness value of the optimal individual meets the requirements, the optimal solution corresponding to the current optimal individual is output to obtain the optimal new energy storage configuration plan; Traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration plans, and visualize the several new energy storage configuration plans in combination with the new energy generation-storage distribution map.

2. A new energy storage configuration method according to claim 1, characterized in that: The new energy power stations include wind power stations, photovoltaic power stations and wind-solar complementary power stations; The wind power station includes several wind turbines, the photovoltaic power station includes several photovoltaic units, and the wind-solar complementary power station includes several wind turbines and several photovoltaic units; The new energy storage system includes a battery energy storage system, a power battery energy storage system, a compressed air energy storage system, a supercapacitor energy storage system and a thermal energy storage system.

3. A new energy storage configuration method according to claim 1, characterized in that: Based on the location data of renewable energy power stations, renewable energy storage systems and transmission lines in several regions, a regional integrated renewable energy generation-storage distribution map is constructed, including the following steps: According to the location data of renewable energy power stations in several regions, several master nodes are set up; According to the location data of the new energy storage systems in several areas, several secondary nodes are set; According to the location data of the power transmission lines in several regions, several discrete nodes are set, and according to the primary nodes / secondary nodes connected at both ends of the power transmission lines, several discrete nodes of the same power transmission line are connected and smoothed to obtain several edges; Based on several main nodes, several secondary nodes and several edges, a regional integrated new energy generation-storage distribution map is constructed.

4. A new energy storage configuration method according to claim 3, characterized in that: The following steps are included to classify several new energy storage systems in the new energy generation-storage distribution diagram to obtain the subsidiary energy storage cluster of each new energy power station: Based on the new energy generation-storage distribution map, several main nodes are used as several cluster centers, and the Euclidean distances from several secondary nodes to several cluster centers are obtained; If the Euclidean distance from the current sub-node to any cluster center is the shortest, and there is a corresponding edge between the sub-node and the cluster center, the current sub-node is divided into the corresponding cluster center. Otherwise, find the next cluster center with a shorter Euclidean distance and repeat this step. Traverse all secondary nodes to obtain the affiliated energy storage cluster of each new energy power station.

5. A new energy storage configuration method according to claim 1, characterized in that: The power generation prediction model is constructed based on the CNN-BiGRU-Attention-KELM algorithm, and the power generation prediction model includes an input layer connected in sequence, a spatial feature extraction layer constructed based on the CNN algorithm, a spatial feature learning layer constructed based on the BiGRU algorithm, an attention weight layer constructed based on the Attention algorithm, an error prediction layer constructed based on the KELM algorithm, and an output layer; The method for constructing a power generation prediction model includes the following steps: Collect historical operation data sets of renewable energy power stations, use improved grey correlation analysis method to cluster historical operation data sets, and obtain similar data sets under different unit configurations; Use PCA analysis method to reduce the dimension of similar data sets under different unit configurations, and obtain historical model training data sets under different unit configurations; According to the historical model training data set, the CNN-BiGRU-Attention-KELM algorithm is used to build a power generation prediction model.

6. A new energy storage configuration method according to claim 5, characterized in that: Collecting real-time operation data of the new energy power station, and using the power generation prediction model to perform power generation prediction based on the real-time operation data to obtain power generation prediction data of the new energy power station, including the following steps: Collect the real-time operation data of the new energy power station, and use the PCA analysis method to reduce the data dimension of the real-time operation data under the current unit configuration to obtain the real-time operation key data, and input the real-time operation key data into the input layer of the power generation prediction model; Use the spatial feature extraction layer to extract the spatial features of key data in real-time operation, and input the spatial features into the feature learning layer; Use the attention weight layer to set the output attention weight value of the feature learning layer; Based on the output attention weight values, the feature learning layer is used to obtain the original prediction values ​​of the key data for real-time operation; Obtain the original error value between the original predicted value and the true value, and use the error prediction layer to correct the original error value to obtain the predicted error value; According to the original prediction value and prediction error value of the real-time operation key data, the output layer is used to output the final prediction value, that is, the power generation prediction data of the new energy power station.

7. A new energy storage configuration system, used to implement the new energy storage configuration method as described in any one of claims 1 to 6, characterized in that: The system comprises a distribution map construction unit, an energy storage system attribution division unit, a power generation prediction unit, an objective function construction unit, a new energy storage configuration unit and an energy storage configuration scheme visualization unit which are connected in sequence; A distribution map construction unit, used to construct a regional integrated new energy generation-storage distribution map based on the location data of new energy power stations, new energy storage systems and transmission lines in several regions; The energy storage system attribution division unit is used to classify a number of new energy storage systems in the new energy generation-storage distribution diagram to obtain the affiliated energy storage clusters of each new energy power station; The power generation prediction unit is used to collect the real-time operation data of the new energy power station, and use the power generation prediction model to perform power generation prediction based on the real-time operation data to obtain the power generation prediction data of the new energy power station; An objective function construction unit is used to construct an absorption objective function of a new energy power station according to power generation prediction data, and to construct a cost objective function of a corresponding subsidiary energy storage cluster according to system cost data of a new energy energy storage system; A new energy storage configuration unit is used to configure new energy storage according to a consumption objective function and a corresponding cost objective function using a new energy storage configuration model to obtain a new energy storage configuration plan; The energy storage configuration scheme visualization unit is used to traverse all the new energy power stations in the new energy generation-storage distribution map, obtain several new energy storage configuration schemes, and visualize the several new energy storage configuration schemes in combination with the new energy generation-storage distribution map.

Images