Energy storage control strategy generation method and system for light storage and charging integrated power station

Abstract

The invention discloses an energy storage control strategy generation method and system for a light storage and charging integrated power station, and particularly relates to the technical field of energy storage capacity optimization. The method is technically characterized by comprising the following steps: carrying out clustering analysis on photovoltaic output sample data generated by simulation, and clustering to obtain a plurality of typical scenes; based on a preset constraint condition, an objective function is constructed with the minimum energy storage charging and discharging power, historical load data and photovoltaic power generation data under multiple typical scenes are input into an improved particle swarm algorithm, an optimal energy storage power set value is obtained, the energy storage capacity is updated, and then a historical energy storage control strategy is generated; inputting the historical load data, the photovoltaic power generation data in the plurality of typical scenes and the corresponding historical optimal energy storage charging and discharging power into the initial support vector machine model for training, and selecting a kernel function to optimize model parameters of the initial support vector machine model to obtain a trained and optimized support vector machine model; and inputting the load active power, the load reactive power and the photovoltaic power generation power at the current moment into the trained and optimized support vector machine model to obtain the current optimal energy storage charging and discharging power, and updating the energy storage capacity based on the current optimal energy storage charging and discharging power so as to obtain the current energy storage control strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage capacity optimization, and in particular to a method and system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station. Background Art

[0002] In response to the climate crisis and the rapid growth of energy demand, "photovoltaic + energy storage + charging pile" has become an effective green charging mode. It improves energy efficiency and system stability by realizing the coordinated regulation of new energy generation, power storage and intelligent charging. Adopting the operation strategy of "green charging, using light to support charging piles", the photovoltaic storage and charging system can not only meet the charging needs of electric vehicles, but also reduce energy waste and energy costs by storing surplus electricity, thus realizing the self-generation and self-use of photovoltaic power generation.

[0003] For the construction of integrated photovoltaic storage and charging electric vehicle charging stations in distribution network areas, it is necessary to design a reasonable energy storage control strategy to dynamically balance the charging and discharging operations under different photovoltaic output conditions. When the photovoltaic output is large, the system will give priority to using photovoltaic power to meet charging needs and store the remaining power in the energy storage system; when the photovoltaic output weakens or the charging load increases, the energy storage system will coordinate with the photovoltaic output to ensure continuous power supply. Every time the photovoltaic output changes, complex calculation methods are required to recalculate the energy storage charging and discharging power, which may cause the energy storage device to respond untimely.

[0004] Therefore, the present invention aims to provide a method and system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, so as to solve the above-mentioned related problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is that a complex calculation method is required to recalculate the energy storage charging and discharging power every time the photovoltaic output changes, which may cause the energy storage device to respond untimely. The purpose is to provide a method and system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station. The energy storage control strategy of the optimal energy storage charging and discharging power generation process of the particle swarm algorithm is learned through an SVM model to obtain a model that can be applied to the current energy storage charging and discharging power calculation, and the energy storage capacity is adjusted using the energy storage charging and discharging power; according to the historical data of different energy storage charging and discharging powers corresponding to different photovoltaic conditions of the photovoltaic storage charging station, the energy storage power demand of the photovoltaic storage charging station is quickly responded to. This energy storage control strategy can cope with the situation where the photovoltaic output fluctuates with the weather, effectively improve the absorption rate of photovoltaic power, and ensure the charging stability of electric vehicles; at the same time, the SVM model trained by clustering and scene-based annual photovoltaic data is adopted to configure the energy storage capacity according to different weather conditions and load fluctuations, without the need to repeatedly run complex optimization algorithms each time, which greatly improves the adaptability and economic benefits of the system in multiple scenarios, ensures the optimal energy storage utilization rate of the photovoltaic storage charging station under various operation modes, and promotes the efficient use of renewable energy.

[0006] The present invention is achieved through the following technical solutions:

[0007] A method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, the method comprising:

[0008] Obtain historical photovoltaic power generation data and historical load data of the power station, use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data, and generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions;

[0009] The K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuations of photovoltaic power generation under different weather conditions, and multiple typical scenarios representing the photovoltaic power generation level are obtained through clustering.

[0010] Based on the preset constraints, the objective function is constructed with the minimum energy storage charging and discharging power, and the historical load data and photovoltaic power generation data under multiple typical scenarios are input into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power;

[0011] The historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power are input into the initial support vector machine model for training, and the kernel function is selected to optimize the model parameters of the initial support vector machine model to obtain the trained and optimized support vector machine model;

[0012] The current load active power, load reactive power and photovoltaic power generation power are input into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and the energy storage capacity is updated using the energy storage charging and discharging power output, and the energy storage control strategy is generated at the same time.

[0013] Furthermore, the Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data to generate photovoltaic output sample data that reflects the photovoltaic output fluctuations under different weather conditions, specifically:

[0014] The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically:

[0015]

[0016] in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

[0017] Furthermore, the K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions. Multiple typical scenarios representing the photovoltaic power generation level are obtained by clustering, specifically:

[0018] The elbow method is used to determine the number of clusters, and the K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuations of photovoltaic power generation under different weather conditions, and a typical scenario characterizing the photovoltaic power generation level is obtained to determine the number of clusters.

[0019] Furthermore, based on the preset constraints, the objective function is constructed with the minimum energy storage charging and discharging power, and the historical load data and photovoltaic power generation data under multiple typical scenarios are input into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power, specifically:

[0020] Set the initial parameters of the particle swarm algorithm, and input the historical load data and photovoltaic power generation data under different typical scenarios into the particle swarm algorithm to obtain the initial energy storage charging and discharging power;

[0021] The objective function is constructed with the minimum energy storage charging and discharging power, and the constraint conditions are established;

[0022] The historical voltage data, historical load data, and historical photovoltaic power of each node are input into the distribution network model, and the initial power flow calculation is performed to obtain the initial voltage distribution. The energy storage active power value is calculated using the quadratic programming algorithm combined with the dynamic power flow equation to obtain the final energy storage active power value.

[0023] The final energy storage active power value is used to adjust the initial energy storage charging and discharging power to obtain the adjusted historical optimal energy storage charging and discharging power.

[0024] Furthermore, the objective function is constructed with the minimum energy storage charging and discharging power, and the constraint conditions are established, specifically:

[0025]

[0026] Among them, p i,d (t) represents the discharge power of the energy storage device at node i at time t; p i,c (t) represents the charging power of the energy storage device at node i at time t; V i min and V i max Respectively represent the minimum and maximum voltages at the ith node; V = [V1, V2, ..., V N ] T ,p=[p1,p2,...,p N ] T , q=[q1,q2,...,q N ] T They represent the node voltage vector, the injected node active power vector, and the injected node reactive power vector respectively; V0 is the voltage value of node 0; 1 N is a vector of all 1s; R and X are the sensitivity matrices of voltage with respect to active and reactive power injection, respectively; η min and η max They are respectively represented by the minimum working capacity ratio and the maximum working capacity ratio of the energy storage device; μ represents the working efficiency of the energy storage device; E i (t) represents the capacity of the energy storage device at time t; Δt represents the unit time of the data; Indicates the energy storage capacity of the energy storage device at the initial moment.

[0027] The present invention also provides a system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, which is used in any one of the above-mentioned methods for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, and the system includes:

[0028] The first module is used to obtain the historical photovoltaic power generation data and historical load data of the power station, use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data, and generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions;

[0029] The second module is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions using the K-means clustering algorithm, and cluster to obtain multiple typical scenarios that characterize the photovoltaic power generation level;

[0030] The third module is used to construct an objective function based on preset constraints with the minimum energy storage charging and discharging power, and input historical load data and photovoltaic power generation data under multiple typical scenarios into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power;

[0031] The fourth module is used to input historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power into the initial support vector machine model for training, and select a kernel function to optimize the model parameters of the initial support vector machine model to obtain a trained and optimized support vector machine model;

[0032] The fifth module is used to input the current load active power, load reactive power and photovoltaic power generation power into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and use the energy storage charging and discharging power output to update the energy storage capacity and generate an energy storage control strategy.

[0033] Furthermore, the Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data to generate photovoltaic output sample data that reflects the photovoltaic output fluctuations under different weather conditions, specifically:

[0034] The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically:

[0035]

[0036] in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

[0037] The present invention also provides a computer device, comprising a system memory and a processor, wherein the system memory stores a computer program, and wherein the processor implements the steps of the method described in any one of claims 1 to 5 when executing the computer program.

[0038] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any one of the methods described above when executed by a processor.

[0039] The present invention also provides a computer program product comprising instructions, wherein when the instructions are executed by a computer device cluster, the computer device cluster executes any of the above methods.

[0040] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0041] In the present invention, the energy storage control strategy of the optimal energy storage charging and discharging power generation process of the particle swarm algorithm is learned through the SVM model to obtain a model that can be applied to the current energy storage charging and discharging power calculation, and the energy storage capacity is adjusted by using the energy storage charging and discharging power; according to the historical data of different energy storage charging and discharging powers corresponding to different photovoltaic conditions of the photovoltaic charging station, the energy storage power demand of the photovoltaic charging station is quickly responded to. This energy storage control strategy can cope with the situation where the photovoltaic output fluctuates with the weather, effectively improve the absorption rate of photovoltaic power, and ensure the charging stability of electric vehicles; at the same time, the SVM model trained by clustering and sub-scenario of annual photovoltaic data is adopted to configure the energy storage capacity according to different weather conditions and load fluctuations, without the need to repeatedly run complex optimization algorithms each time, which greatly improves the adaptability and economic benefits of the system in multiple scenarios, ensures the optimal energy storage utilization rate of the photovoltaic charging station under various operation modes, and promotes the efficient use of renewable energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0043] Figure 1 Flow chart of a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0044] Figure 2 A test system node diagram used in a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0045] Figure 3 A one-day load and photovoltaic output fluctuation diagram in the verification of a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0046] Figure 4 This is a 34-node annual photovoltaic output fluctuation diagram being verified in a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0047] Figure 5 This is a typical daily photovoltaic output fluctuation diagram used in the verification of a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0048] Figure 6 It is a target solar-storage-charging area energy storage power setting diagram in the verification of an energy storage control strategy generation method for a solar-storage-charging integrated power station in this embodiment;

[0049] Figure 7 It is an energy storage power diagram of each photovoltaic storage and charging area in the verification of an energy storage control strategy generation method for a photovoltaic storage and charging integrated power station in this embodiment;

[0050] Figure 8 a is a schematic diagram of energy storage capacity of each photovoltaic storage and charging area under the application of improved particle swarm algorithm in the verification of energy storage control strategy generation method for photovoltaic storage and charging integrated power station in this embodiment;

[0051] Figure 8 b is a schematic diagram of the energy storage capacity of each photovoltaic storage and charging area in the verification of the energy storage control strategy generation method for a photovoltaic storage and charging integrated power station in this embodiment;

[0052] Fig. 9 This is a schematic diagram of system module connections for a system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station in this embodiment;

[0053] Fig.10 A schematic diagram of the structure of a computer device. DETAILED DESCRIPTION

[0054] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0055] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0056] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0057] Example 1

[0058] See also Figure 1 As shown, this embodiment provides a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, the method comprising:

[0059] S1: Obtain the historical photovoltaic power generation data and historical load data of the power station, use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data, and generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions, specifically:

[0060] Obtain historical photovoltaic power generation data and historical load data of the power station. It should be noted that in this embodiment, the historical data is the photovoltaic power generation data and load data of the past year; then pre-process the acquired historical photovoltaic power generation data and historical load data, mainly to fill in abnormal values ​​and data to ensure the integrity and accuracy of the data;

[0061] The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically:

[0062]

[0063] in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

[0064] S2: Based on the fluctuation of photovoltaic power generation under different weather conditions, the K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data. Multiple typical scenarios representing the photovoltaic power generation level are obtained through clustering, specifically:

[0065] The elbow method is used to determine the number of clusters. The K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions, and the typical scenario characterizing the photovoltaic power generation level is obtained to determine the number of clusters.

[0066] Specifically, in this embodiment, the number of clusters k is gradually increased, and the intra-cluster sum of square errors under each k value is calculated; for each k, the calculation formula of the intra-cluster sum of square errors (WCSS) is:

[0067]

[0068] Among them, ||x i -c j || 2 Represents the sample point x i With cluster center c j The square of the Euclidean distance between j represents the set of sample points in the jth cluster, and WCSS is the sum of the squared errors in all clusters. The goal of the elbow method is to select k so that WCSS clearly shows an "elbow" shape when k increases, that is, the rate of decline of WCSS slows down significantly, which is usually expressed as:

[0069] ΔS(k)=S(k-1)-S(k)

[0070] It is believed that when ΔS(k)≤0.01, the optimal k is found;

[0071] In cluster analysis, the goal of the K-means algorithm is to minimize the sum of squared errors within a cluster. The K-means algorithm divides the data set into k clusters, each with a center. The goal of the K-means algorithm is to find the k cluster centers c1, c2, …, c k And the cluster assignment of each sample to minimize the sum of squared errors within the cluster. For each sample x, calculate its distance to each cluster center and assign it to the nearest cluster center, specifically:

[0072]

[0073] Among them, r ij It is represented as an indicator variable, indicating that the sample x i Whether it is assigned to cluster j;

[0074] The objective function can be expressed as:

[0075]

[0076] Specifically, in this embodiment, based on the collected historical photovoltaic power generation data, a large amount of photovoltaic output sample data is generated through the Monte Carlo simulation method to cover the output fluctuations under different weather conditions, ensuring that the model can optimize the energy storage capacity under various weather scenarios; then the K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data, and it is divided into multiple typical scenarios according to the fluctuations of photovoltaic power generation, where each scenario represents a different photovoltaic output level.

[0077] S3: Based on the preset constraints, the objective function is constructed with the minimum energy storage charging and discharging power, and the historical load data and photovoltaic power generation data under multiple typical scenarios are input into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power, specifically:

[0078] The initial parameters of the particle swarm algorithm are set. It should be noted that, in this embodiment, the initial parameters include parameters such as the number of particles, the number of iterations, and the speed; and the historical load data and the photovoltaic power generation data under different typical scenarios are input into the particle swarm algorithm to obtain the initial energy storage charging and discharging power, based on which the energy storage capacity is updated;

[0079] Specifically, in this embodiment, a population of 100 particles is initialized, each particle represents a different energy storage capacity configuration, and the current optimal solution is found during initialization assuming that the number of positions is fixed.

[0080] The objective function is constructed with the minimum energy storage charging and discharging power, and the constraints are established, specifically:

[0081]

[0082] Among them, p i,d (t) represents the discharge power of the energy storage device at node i at time t; p i,c (t) represents the charging power of the energy storage device at node i at time t; V i min and V i max Respectively represent the minimum and maximum voltages at the ith node; V = [V1, V2, ..., V N ] T ,p=[p1,p2,...,p N ] T , q=[q1,q2,...,q N ] TThey represent the node voltage vector, the injected node active power vector, and the injected node reactive power vector respectively; V0 is the voltage value of node 0; 1 N is a vector of all 1s; R and X are the sensitivity matrices of voltage with respect to active and reactive power injection, respectively; η min and η max They are respectively represented by the minimum working capacity ratio and the maximum working capacity ratio of the energy storage device; μ represents the working efficiency of the energy storage device; E i (t) represents the capacity of the energy storage device at time t; Δt represents the unit time of the data; Indicates the energy storage capacity of the energy storage device at the initial moment;

[0083] The historical voltage data, historical load data, and historical photovoltaic power of each node are input into the distribution network model, and the initial power flow calculation is performed to obtain the initial voltage distribution. The energy storage active power value is calculated using the quadratic programming algorithm combined with the dynamic power flow equation to obtain the final energy storage active power value.

[0084] It should be noted that, in this embodiment, the historical voltage data, historical load data, and historical photovoltaic power of each node are input into the distribution network model, and the power demand of the distribution network is calculated by the dynamic power flow equation and the quadratic programming algorithm, and the response capacity of the energy storage under the load demand is calculated. The linearized DistFlow equation of the distribution network power flow is:

[0085] V=Rp+Xq+V01 N

[0086] Where V = [V1, V2, ..., V N ] T ,p=[p1,p2,...,p N ] T , q=[q1,q2,...,q N ] T They represent the node voltage vector, the injected node active power vector, and the injected node reactive power vector respectively. V0 is the voltage value of node 0; 1 N is a vector of all 1s; R and X are the sensitivity matrices of voltage with respect to active and reactive power injection, respectively;

[0087] Then, the quadratic programming method is used to calculate the optimal energy storage active power setting value of the energy storage device compensation at time t.

[0088] The final energy storage active power value is used to adjust the initial energy storage charging and discharging power, obtain the adjusted historical optimal energy storage charging and discharging power, and update the energy storage capacity.

[0089] It should be noted that in this embodiment, the fitness value of each particle (the balance between economic benefits and costs) is calculated again to find the global optimal solution and the individual optimal solution; the speed and position parameters of the particle swarm are updated, and the energy storage power output is adjusted, the energy storage capacity is updated, and the iterative adjustment is continued to obtain the adjusted optimal energy storage power setting value, determine the energy storage capacity, and generate a historical energy storage control strategy.

[0090] S4: inputting historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power into the initial support vector machine model for training, and selecting a kernel function to optimize the model parameters of the initial support vector machine model to obtain a trained and optimized support vector machine model;

[0091] It should be noted that, in this embodiment, the training formula of the SVM model is:

[0092] f(p load (t),q load (t),p pv (t))=<ω,Φ>+b

[0093] Among them, Φ is the input feature vector, including the load active power p load (t), reactive power q load (t) and photovoltaic power generation power p pv (t); ω is the weight vector of the model; b is the bias term;

[0094] At the same time, it should be noted that in this embodiment, the best kernel function is selected by K-fold cross validation; in order to optimize the model, a suitable kernel function is selected, and a linear kernel, a polynomial kernel or a radial basis function (RBF) kernel can be used. The best kernel function is selected by cross validation to minimize the training error and improve the generalization ability of the model. The formulas of the three kernel functions are specifically as follows:

[0095] Linear kernel: K(x i ,x j )= <x i ,x j >

[0096] Polynomial kernel: K(x i ,x j )=(γ <x i ,x j >+r) d

[0097] Radial basis function kernel: K(x i ,x j )=exp(-βx i -x j 2 )

[0098] Where γ is the kernel coefficient, r is a constant, d is the order of the polynomial, and β controls the width of the Gaussian function; these kernel functions are used to calculate the eigenvector x i and x j The similarity between them helps the model find the best decision boundary, specifically:

[0099] First, the data set will contain historical data such as load active power, load reactive power, and photovoltaic power. These data are divided into K subsets, and the performance of different kernel functions in energy storage power prediction is evaluated through cross-validation. Each round of training uses K-1 of them as training sets, and the remaining 1 as a validation set. In this way, the entire data set will be fully trained and validated K times to ensure that each subset is used as a validation set. The optimization objective function of SVM is:

[0100]

[0101] Among them, the weight vector ω is the weight of each control variable that affects the energy storage power regulation; since the energy storage capacity regulation in the integrated photovoltaic storage and charging power station may be affected by a variety of uncertain factors (such as load fluctuations, photovoltaic output changes), ξ i It is used to indicate the extent to which the energy storage allowed exceeds the target range. The larger the ξ i It means that the energy storage regulation is relatively loose and allows a certain deviation; the smaller ξ i This indicates that the energy storage power needs to be controlled more accurately. The penalty parameter C is used to balance the accuracy and flexibility of energy storage control and can control the degree of penalty for energy storage power deviation. A larger C value means that the system is more sensitive to energy storage power deviation and requires that the energy storage power must be strictly controlled within the specified range; a smaller C value means that the system has a higher tolerance for energy storage power deviation and can allow a certain degree of overshoot or undershoot. In energy storage capacity regulation, the ||ω|| 2 To ensure that the charging and discharging power of the energy storage system is minimized during the voltage regulation process, thereby improving the economic benefits of the system.

[0102] Secondly, the linear kernel, polynomial kernel and RBF kernel are used to train the SVM model respectively, and these models are used to predict the power storage power response of the power station and evaluate its error. After each round of training, the model is tested using the validation set and the error value of each kernel function is calculated. Common error measurement indicators include mean square error (MSE) or misclassification rate. The error value of each round of verification is recorded, and finally the errors of all K rounds are averaged. The verification error formula is:

[0103]

[0104] where y i is the true value, is the predicted value of the model, and n is the number of samples in the validation set.

[0105] For each kernel function, the average error of K-fold validation is calculated as follows:

[0106]

[0107] Among them, MSE k is the validation error of the k-th fold.

[0108] Finally, the kernel function with the smallest error is selected to optimize the energy storage control strategy, so that the charging and discharging power of the energy storage system of the photovoltaic storage charging station is close to the historical optimal energy storage charging and discharging power obtained based on the particle swarm algorithm, thereby improving the economic benefits of the photovoltaic storage and charging integrated power station.

[0109] S5: Input the current load active power, load reactive power and photovoltaic power generation power into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and use the energy storage charging and discharging power output to update the energy storage capacity, and generate the energy storage control strategy.

[0110] Specifically, in this embodiment, the energy storage control strategy of the optimal energy storage charging and discharging power generation process of the particle swarm algorithm is learned through the SVM model to obtain a model that can be applied to the current energy storage charging and discharging power calculation, and the energy storage capacity is adjusted using the energy storage charging and discharging power; according to the historical data of different energy storage charging and discharging powers corresponding to different photovoltaic conditions of the photovoltaic charging station, the energy storage power demand of the photovoltaic charging station is quickly responded to. This energy storage control strategy can cope with the situation where photovoltaic output fluctuates with weather, effectively improve the absorption rate of photovoltaic power, and ensure the charging stability of electric vehicles; at the same time, the SVM model trained by clustering and sub-scenario of annual photovoltaic data is adopted to configure energy storage capacity according to different weather conditions and load fluctuations, without the need to repeatedly run complex optimization algorithms each time, which greatly improves the adaptability and economic benefits of the system in multiple scenarios, ensures the optimal energy storage utilization rate of the photovoltaic charging station under various operating modes, and promotes the efficient use of renewable energy.

[0111] At the same time, in this embodiment, in order to verify the effectiveness of the provided method, an improved IEEE34-node system is used as a test system for simulation analysis. The adjustable distributed resources of the system include 33 photovoltaics and 4 energy storage devices, such as Figure 2As shown. All nodes except node 0 are equipped with photovoltaic inverters, and their capacity is set to 12kVA. The power factor of each photovoltaic power generation device is 0.97; nodes 4, 11, 20 and 28 are set as photovoltaic storage and charging areas, and the energy storage related configuration is shown in Table 1. At the same time, the weather probability table for generating photovoltaic data based on seasons and weather probabilities is also given, as shown in Table 2. For ease of calculation, all energy storage device parameters are set to the same, assuming that the voltage of substation node 0 is fixed to the reference voltage per unit value 1.

[0112] Table 1 Energy storage device parameters

[0113]

[0114]

[0115] Table 2 Season and weather probability coefficients

[0116]

[0117] In the photovoltaic storage and charging integrated power station, the photovoltaic output peak is set to 11kW, and the sampling time is set to 1min. The specific waveform changes of the load and photovoltaic output in a random day are shown in the figure below: Figure 3 As shown; the photovoltaic fluctuation diagram of the annual photovoltaic data used is as follows Figure 4 The 6 typical days of photovoltaic output fluctuations obtained after clustering are shown in Figure 5 shown.

[0118] In the present invention, some parameters in K-fold cross validation during SVM model training are selected as shown in Table 3 below:

[0119] Table 3 SVM model training parameter selection

[0120]

[0121] After optimizing the energy storage capacity of the integrated photovoltaic power station, the energy storage power distribution of the test system is obtained through simulation. Figure 6 The energy storage control strategy generation method for a photovoltaic storage and charging integrated power station designed by the present invention is used to quickly approach the optimal capacity configuration plan by learning the historical data of photovoltaic power generation and load demand. The energy storage power distribution obtained is as follows Figure 7 In order to further verify the effectiveness of the provided method, the energy storage control strategy generation method designed by the present invention is compared with the improved particle swarm algorithm, and the energy storage capacity distribution is obtained as follows Figure 8As shown. By comparing the two capacity configurations, it can be found that the energy storage control strategy generation method of the present invention can achieve similar effects as the improved particle swarm algorithm. While greatly reducing the computational complexity, the SVM strategy has better real-time performance and can obtain similar energy storage power distribution and energy storage capacity configuration, that is, it can achieve the economic benefits of the photovoltaic charging station similar to the improved particle swarm algorithm, and has stronger flexibility and robustness.

[0122] Example 2

[0123] The present invention also provides a system for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, which is used in any one of the above-mentioned methods for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, and the system includes:

[0124] The first module 100 is used to obtain historical photovoltaic power generation data and historical load data of the power station, and use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data to generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions;

[0125] The second module 200 is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions using a K-means clustering algorithm, and cluster to obtain multiple typical scenes representing the photovoltaic power generation level;

[0126] The third module 300 is used to construct an objective function based on preset constraints with minimum energy storage charging and discharging power, and input historical load data and photovoltaic power generation data under multiple typical scenarios into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power;

[0127] The fourth module 400 is used to input historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power into the initial support vector machine model for training, and select a kernel function to optimize the model parameters of the initial support vector machine model to obtain a trained and optimized support vector machine model;

[0128] The fifth module 500 is used to input the current load active power, load reactive power and photovoltaic power generation power into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and use the energy storage charging and discharging power output to update the energy storage capacity, and generate an energy storage control strategy.

[0129] Furthermore, the Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data to generate photovoltaic output sample data that reflects the photovoltaic output fluctuations under different weather conditions, specifically:

[0130] The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically:

[0131]

[0132] in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

[0133] It should be noted that the modules in the system of Example 2 correspond to the steps in the method of Example 1. The steps in the method of Example 1 have been described in detail in Example 1. In this Example 2, the contents of the modules in the system will not be described in detail.

[0134] Example 3

[0135] This embodiment further provides a computer device, including a system memory 1005 and a processor 1001 . The system memory 1005 stores a computer program. When the processor 1001 executes the computer program, the steps of any one of the above methods are implemented.

[0136] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, the processor 1001 implements the functions of each module / unit in the above system / device embodiments when executing the computer program.

[0137] Specifically, in this embodiment, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the system memory 1005, and are executed by the processor 1001 to complete the present application. One or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0138] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0139] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors 1001, digital signal processors 1001 (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 1001 may be a microprocessor or any conventional processor, etc.

[0140] The system memory 1005 may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The system memory 1005 may also be a storage device 1004 of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the terminal device. Further, the system memory 1005 may also include both the internal storage unit of the terminal device and the storage device 1004. The system memory 1005 is used to store computer programs and other programs and data required by the terminal device. The system memory 1005 may also be used to temporarily store data that has been output or is to be output.

[0141] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0142] Example 4

[0143] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0144] Among them, the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, system or device, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk. Random Access Memory (RAM), Read-Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM), registers, hard disks, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above, or any other form of computer-readable storage medium known in the art.

[0145] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In an embodiment of the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system, or device.

[0146] Example 5

[0147] This embodiment also provides a computer program product including instructions. When the instructions are executed by a computer device cluster, the computer device cluster executes the method described in Embodiment 1.

[0148] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station, characterized in that the method include: Obtain historical photovoltaic power generation data and historical load data of the power station, use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data, and generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions; The K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuations of photovoltaic power generation under different weather conditions, and multiple typical scenarios representing the photovoltaic power generation level are obtained through clustering. Based on the preset constraints, the objective function is constructed with the minimum energy storage charging and discharging power, and the historical load data and photovoltaic power generation data under multiple typical scenarios are input into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power; The historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power are input into the initial support vector machine model for training, and the kernel function is selected to optimize the model parameters of the initial support vector machine model to obtain the trained and optimized support vector machine model; The current load active power, load reactive power and photovoltaic power generation power are input into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and the energy storage capacity is updated using the energy storage charging and discharging power output, and the energy storage control strategy is generated at the same time.

2. A method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station according to claim 1, characterized in that: The Monte Carlo simulation method is used to simulate historical photovoltaic power generation data to generate photovoltaic output sample data that reflects the photovoltaic output fluctuations under different weather conditions. Specifically: The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically: in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

3. The method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station according to claim 1, characterized in that: The K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions. The clustering results in multiple typical scenarios that characterize the photovoltaic power generation level, specifically: The elbow method is used to determine the number of clusters, and the K-means clustering algorithm is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuations of photovoltaic power generation under different weather conditions, and a typical scenario characterizing the photovoltaic power generation level is obtained to determine the number of clusters.

4. The method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station according to claim 1, characterized in that: Based on the preset constraints, the objective function is constructed with the minimum energy storage charging and discharging power, and the historical load data and photovoltaic power generation data under multiple typical scenarios are input into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power, which is: Set the initial parameters of the particle swarm algorithm, and input the historical load data and photovoltaic power generation data under different typical scenarios into the particle swarm algorithm to obtain the initial energy storage charging and discharging power; The objective function is constructed with the minimum energy storage charging and discharging power, and the constraint conditions are established; The historical voltage data, historical load data, and historical photovoltaic power of each node are input into the distribution network model, and the initial power flow calculation is performed to obtain the initial voltage distribution. The energy storage active power value is calculated using the quadratic programming algorithm combined with the dynamic power flow equation to obtain the final energy storage active power value. The final energy storage active power value is used to adjust the initial energy storage charging and discharging power to obtain the adjusted historical optimal energy storage charging and discharging power.

5. The method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station according to claim 4, characterized in that: The objective function is constructed with the minimum energy storage charging and discharging power, and the constraints are established, specifically: s.t.V i min ≤V i (t)≤V i max p i,e (t)=-p i,c (t)+p i,d (t) Among them, p i,d (t) represents the discharge power of the energy storage device at node i at time t; p i,c (t) represents the charging power of the energy storage device at node i at time t; V i min and V i max Respectively represent the minimum and maximum voltages at the ith node; V = [V1, V2, ..., V N ] T ,p=[p1,p2,...,p N ] T , q=[q1,q2,...,q N ] T They represent the node voltage vector, the injected node active power vector, and the injected node reactive power vector respectively; V0 is the voltage value of node 0; 1 N is a vector of all 1s; R and X are the sensitivity matrices of voltage with respect to active and reactive power injection, respectively; η min and η max They are respectively represented by the minimum working capacity ratio and the maximum working capacity ratio of the energy storage device; μ represents the working efficiency of the energy storage device; E i (t) represents the capacity of the energy storage device at time t; Δt represents the unit time of the data; Indicates the energy storage capacity of the energy storage device at the initial moment.

6. A system for generating energy storage control strategies for a photovoltaic storage and charging integrated power station, characterized in that: The system is used in a method for generating an energy storage control strategy for a photovoltaic storage and charging integrated power station as described in any one of claims 1 to 5, and the system includes: The first module is used to obtain the historical photovoltaic power generation data and historical load data of the power station, use the Monte Carlo simulation method to simulate the historical photovoltaic power generation data, and generate photovoltaic output sample data that maps the output fluctuations of photovoltaic power generation under different weather conditions; The second module is used to perform cluster analysis on the simulated photovoltaic output sample data according to the output fluctuation of photovoltaic power generation under different weather conditions using the K-means clustering algorithm, and cluster to obtain multiple typical scenarios that characterize the photovoltaic power generation level; The third module is used to construct an objective function based on preset constraints with the minimum energy storage charging and discharging power, and input historical load data and photovoltaic power generation data under multiple typical scenarios into the particle swarm algorithm to obtain the historical optimal energy storage charging and discharging power; The fourth module is used to input historical load data, photovoltaic power generation data under multiple typical scenarios and the corresponding historical optimal energy storage charging and discharging power into the initial support vector machine model for training, and select a kernel function to optimize the model parameters of the initial support vector machine model to obtain a trained and optimized support vector machine model; The fifth module is used to input the current load active power, load reactive power and photovoltaic power generation power into the trained and optimized support vector machine model to obtain the energy storage charging and discharging power output, and use the energy storage charging and discharging power output to update the energy storage capacity and generate an energy storage control strategy.

7. The energy storage control strategy generation system for a photovoltaic storage and charging integrated power station according to claim 6, characterized in that: The Monte Carlo simulation method is used to simulate historical photovoltaic power generation data to generate photovoltaic output sample data that reflects the photovoltaic output fluctuations under different weather conditions. Specifically: The Monte Carlo simulation method is used to simulate the historical photovoltaic power generation data, and the seasonal probability coefficient and weather condition probability coefficient are introduced in the simulation process to generate photovoltaic output sample data that maps the photovoltaic output fluctuations under different weather conditions. Specifically: in, represents the photovoltaic output sample data at time t under the conditions of season s and weather ω; P s represents the seasonal probability coefficient of season s; P ω,s represents the weather condition probability coefficient of weather ω in season s; Represents simulated PV output data in any season and weather.

8. A computer device comprising a system memory and a processor, wherein the system memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product comprising instructions, characterized in that When the instructions are executed by a computer device cluster, the computer device cluster executes the method according to any one of claims 1 to 5.

Images