Photovoltaic power station energy storage optimal configuration method, computer equipment and storage medium

Abstract

The invention relates to the technical field of energy storage configuration, and discloses a photovoltaic power station energy storage optimal configuration method, computer equipment and a storage medium. The method comprises the following steps: acquiring first historical operation data of a photovoltaic power station; constructing a first multi-dimensional time sequence vector corresponding to the photovoltaic power station by using the first historical operation data; performing clustering analysis on the first multi-dimensional time sequence vector to generate typical scene data corresponding to the photovoltaic power station; and based on the typical scene data and a preset energy storage charging and discharging strategy, optimizing energy storage configuration parameters of the photovoltaic power station to obtain an optimal configuration scheme. By implementing the technical scheme, the energy storage configuration parameter configuration is dynamically optimized, and the power generation utilization rate of the photovoltaic power station and the operation economy of the energy storage system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage configuration, and in particular to a photovoltaic power station energy storage optimization configuration method, computer equipment, and storage medium. Background Art

[0002] Optimizing energy storage configuration parameters is crucial when deploying energy storage systems in photovoltaic power plants to increase photovoltaic absorption rates and improve economic efficiency. However, current optimization methods often fail to meet actual operational characteristics and project requirements, resulting in poor economic performance and impracticality. Summary of the Invention

[0003] In view of this, the present invention provides a photovoltaic power station energy storage optimization configuration method, computer equipment and storage medium to solve the problem that related optimization methods are divorced from actual operating characteristics and engineering requirements, resulting in poor economic efficiency and unimplementable optimization results.

[0004] In a first aspect, the present invention provides a method for optimizing energy storage configuration of a photovoltaic power station, comprising: obtaining first historical operating data of the photovoltaic power station; constructing a first multidimensional time series vector corresponding to the photovoltaic power station using the first historical operating data; performing cluster analysis on the first multidimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station; and optimizing energy storage configuration parameters of the photovoltaic power station based on the typical scenario data and a preset energy storage charging and discharging strategy to obtain an optimal configuration scheme.

[0005] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention constructs a first multidimensional time series vector reflecting the comprehensive operating characteristics of the photovoltaic power station and the characteristics of the electricity market through first historical operating data, eliminating the deviation of the theoretical model. Cluster analysis is used to extract representative typical scenario data, avoiding the configuration inaccuracy caused by missing scenarios. Parameter optimization is combined with preset charging and discharging strategies, and the parameters are forced to meet engineering constraints and market rules, thereby realizing a closed-loop optimization process that is data-driven, scenario-adaptive, and strategy-coordinated, and improving the practicality and feasibility of the configuration solution.

[0006] In an optional embodiment, a cluster analysis is performed on the first multidimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station, including: dividing the first multidimensional time series vector according to the time dimension to obtain multiple time series vector subsets; for any time series vector subset, using a time series clustering method to determine the optimal number of clusters corresponding to the time series vector subset; clustering the time series vector subsets according to the optimal number of clusters to generate typical scenario data corresponding to the photovoltaic power station.

[0007] The photovoltaic power plant energy storage optimization configuration method provided by the present invention specifically addresses the issue of data characteristic differences across time periods by partitioning the first multidimensional time series vector into subsets based on the time dimension. The optimal number of clusters is dynamically determined for each subset, avoiding the scene distortion caused by a fixed number of clusters. Based on customized clustering results, typical scenario data is generated, significantly improving the time series representativeness and clustering quality of these typical scenarios.

[0008] In an optional embodiment, a subset of time series vectors is clustered according to an optimal number of clusters to generate typical scenario data corresponding to a photovoltaic power station, including: clustering a subset of time series vectors according to an optimal number of clusters to obtain multiple cluster clusters; for any cluster cluster, determining the dynamic time warping distance between a first vector corresponding to a unit duration in the cluster and a second vector corresponding to a remaining duration; obtaining a third vector corresponding to the minimum value of the multiple dynamic time warping distances, and determining the third vector as a target vector of the cluster cluster; and using the target vector to determine the typical scenario data corresponding to the photovoltaic power station.

[0009] The photovoltaic power plant energy storage optimization configuration method provided by the present invention accurately quantifies the similarity of temporal morphology by calculating the mean dynamic time warping distance between each vector within a cluster and the remaining vectors. The method uses the minimum mean dynamic time warping distance as a criterion to select the most representative real-world date vectors within the cluster, ensuring that the target scenario accurately reflects the temporal fluctuation characteristics of the original data. Typical scenario data is constructed based on the representative vectors of each cluster, significantly improving the authenticity and interpretability of the typical scenarios.

[0010] In an optional embodiment, the target vector is used to determine the typical scene data corresponding to the photovoltaic power station, including: determining the ratio of the number of unit durations of each cluster to the corresponding target duration, and obtaining the target weight corresponding to each cluster; using the target weight corresponding to each cluster, performing weighted averaging on the target vectors of each cluster belonging to the target duration, and obtaining the duration scene data corresponding to the target duration; and splicing the various duration scene data in chronological order to obtain the typical scene data corresponding to the photovoltaic power station.

[0011] The photovoltaic power plant energy storage optimization configuration method provided by this embodiment uses the actual number of days in clusters as weights to perform a weighted average of representative target vectors, resulting in a probability distribution of duration scenario data that is highly consistent with historical reality. The weighted results for each time period are then concatenated in chronological order to construct typical scenario data for a complete cycle. This significantly improves the accuracy of the scenario set's statistical representation of actual operating conditions while preserving temporal continuity.

[0012] In an optional embodiment, based on typical scenario data and a preset energy storage charging and discharging strategy, the energy storage configuration parameters of the photovoltaic power station are optimized to obtain an optimal configuration scheme, including: obtaining the energy storage system parameters and decision variable parameters of the photovoltaic power station; under preset constraints, using the energy storage system parameters, typical scenario data and the energy storage charging and discharging strategy, tuning the resource consumption parameters and resource replacement parameters of the photovoltaic power station to obtain optimal resource consumption parameters and optimal resource replacement parameters; under the conditions of the optimal resource consumption parameters and optimal resource replacement parameters, using a preset genetic algorithm to tune the decision variable parameters to obtain target decision variable parameters; and using the target decision variable parameters to determine the optimal configuration scheme.

[0013] The photovoltaic power plant energy storage optimization configuration method provided by the present invention utilizes a hierarchical optimization framework to first independently optimize resource parameters under preset constraints based on system parameters and scenario data, ensuring economic viability. Using the optimized resource parameters as fixed conditions, a genetic algorithm is then employed to specifically tune decision variables, enabling efficient global search of complex decision spaces. Finally, the results of these two optimization phases are integrated to generate target parameters, significantly improving optimization efficiency and solution quality while ensuring project feasibility.

[0014] In an optional embodiment, second historical operating data of the photovoltaic power station under extreme weather scenarios is obtained; a second multidimensional time series vector of the photovoltaic power station under extreme weather scenarios is constructed using the second historical operating data; and the energy storage backup capacity of the photovoltaic power station is determined using the second multidimensional time series vector.

[0015] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention extracts key weather scenarios by analyzing historical weather data and calculates the energy storage backup capacity based on this. This enables the energy storage configuration solution to actively defend against abnormal weather conditions and enhances the system's operational robustness in real complex environments.

[0016] In an optional embodiment, the output capacity data, actual online power data and power resource data of the photovoltaic power station in the current time period are obtained; if the difference between the output capacity data and the online power data is greater than a first preset threshold, it is determined whether the energy storage capacity of the photovoltaic power station reaches a second preset threshold; if the energy storage capacity does not reach the second preset threshold, an energy storage charging operation is performed; if the energy storage capacity reaches the second preset threshold, a photovoltaic power curtailment operation is performed.

[0017] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention triggers energy storage operation by comparing the difference between the output capacity and the actual grid power in real time, and sets the energy storage capacity threshold as the basis for secondary decision-making, forming a dynamic response mechanism with dual threshold control. This not only effectively utilizes surplus photovoltaic power generation to reduce abandoned light, but also avoids the risk of energy storage overcharging, thereby improving the safety of system operation and resource utilization.

[0018] In an optional embodiment, if the difference between the output capacity data and the online power data is less than a first preset threshold, it is determined whether the remaining energy storage capacity of the photovoltaic power station reaches a third preset threshold; if the energy storage capacity does not reach the third preset threshold, the photovoltaic power abandonment operation is performed; if the energy storage capacity reaches the third preset threshold, it is determined whether the power resource data reaches a fourth preset threshold; if the power resource data reaches the fourth preset threshold, the energy storage discharge operation is performed; if the power resource data does not reach the fourth preset threshold, the energy storage static operation is performed.

[0019] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention uses a three-level condition-based hierarchical decision-making mechanism to first determine the output gap, then check the remaining energy storage capacity, and finally trigger discharge by coupling with power resource data. This achieves refined energy management under abnormal operating conditions, maximizes economic benefits while ensuring energy storage safety, and significantly improves the system's responsiveness and intelligence in low-output scenarios.

[0020] In a second aspect, the present invention provides a photovoltaic power station energy storage optimization configuration device, including: a first acquisition module, used to obtain first historical operation data of the photovoltaic power station; a first construction module, used to use the first historical operation data to construct a first multidimensional time series vector corresponding to the photovoltaic power station; a clustering module, used to perform cluster analysis on the first multidimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station; an optimization module, used to optimize the energy storage configuration parameters of the photovoltaic power station based on the typical scenario data and a preset energy storage charging and discharging strategy to obtain an optimal configuration scheme.

[0021] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to thereby execute the photovoltaic power station energy storage optimization configuration method of the first aspect or any corresponding embodiment thereof.

[0022] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the photovoltaic power station energy storage optimization configuration method of the first aspect or any corresponding embodiment thereof.

[0023] In a fifth aspect, the present invention provides a computer program product comprising computer instructions for causing a computer to execute the method for optimizing energy storage configuration of a photovoltaic power station according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 is a flow chart of a method for optimizing energy storage configuration in a photovoltaic power station according to an embodiment of the present invention;

[0026] Figure 2 is a flow chart of another photovoltaic power station energy storage optimization configuration method according to an embodiment of the present invention;

[0027] Figure 3 is a flow chart of another photovoltaic power station energy storage optimization configuration method according to an embodiment of the present invention;

[0028] Figure 4 is a schematic diagram of a photovoltaic power station energy storage charging and discharging strategy according to an embodiment of the present invention;

[0029] Figure 5 This is a structural block diagram of a photovoltaic power station energy storage optimization configuration device according to an embodiment of the present invention;

[0030] Figure 6 Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0032] Photovoltaic power generation, as an important form of clean energy, exhibits significant intermittency and volatility in its output, influenced by sunlight intensity, meteorological conditions, and seasonal variations. To improve the grid's capacity and increase the economic benefits of power plants, deploying energy storage systems has become a key technological path for photovoltaic power plants.

[0033] However, current methods for configuring energy storage for photovoltaic power plants are weak in modeling diverse characteristic data, such as climate conditions and load fluctuations. This results in the generation of typical scenarios that fail to capture all regional operational characteristics, impacting the rationality of subsequent optimization. Furthermore, extreme weather risk analysis is generally neglected, resulting in a lack of robustness in configuration results. Furthermore, current mathematical models fail to fully account for key engineering conditions such as energy storage charge and discharge times, power limits, and rate restrictions, and lack an economic closed-loop optimization mechanism centered on maximizing net returns. This ultimately results in low utilization of the intermittent nature of photovoltaic power generation and poor economic returns.

[0034] In view of this, the technical solution of the present invention constructs a multi-dimensional time series vector driven by historical operating data, and adopts refined time series clustering based on dynamic time warping distance to effectively capture the complex characteristics of photovoltaic power stations, generate target data covering all operating scenarios, and solve the problem of incomplete scene coverage caused by the weak modeling ability of traditional methods. At the same time, historical extreme weather data is introduced to quantify the energy storage backup capacity, explicitly enhancing the defense capability against extreme risks. In the optimization link, the engineering conditions are constrained by pre-set charging and discharging strategies, and a two-stage economic optimization mechanism of resource parameter tuning and genetic algorithm decision-making is established to achieve a closed-loop design that maximizes net benefits. Assisted by dynamic control logic based on output / power / energy storage, charging, power abandonment and discharging operations are accurately coordinated, ultimately systematically improving the efficiency and economy of light resource utilization.

[0035] According to an embodiment of the present invention, an embodiment of a method for optimizing energy storage configuration in a photovoltaic power station is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0036] In this embodiment, a photovoltaic power station energy storage optimization configuration method is provided, which can be used for computer equipment, such as servers, Figure 1 FIG. 1 is a flow chart of a method for optimizing energy storage configuration in a photovoltaic power station according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0037] Step S101: Acquire first historical operating data of a photovoltaic power station.

[0038] The first historical operating data refers to various types of relevant data generated by the photovoltaic power station during its past operation, which is used to reflect the operating status of the photovoltaic power station, the environmental conditions of the region where it is located, and the electricity market situation. Specifically, the historical data of the photovoltaic power station is collected according to a preset period and preset resolution. For example, it may include light intensity, temperature, predicted maximum output of the photovoltaic power station, actual grid power, electricity spot market price, etc. After collection, the historical data is reviewed, verified, and corrected. Upsampling and downsampling are used to unify the time resolution, duplicate data is deleted, missing data is interpolated using the Lagrange interpolation method, outlier data is filtered and restored to obtain the first historical operating data. The preset period can be, for example, 1 year, and the preset resolution can be, for example, 15 minutes.

[0039] Step S102: construct a first multi-dimensional time series vector corresponding to the photovoltaic power station using the first historical operation data.

[0040] The first multidimensional time series vector is a multidimensional vector used to characterize the operating characteristics of the photovoltaic power station and the characteristics of the electricity market, and is presented in the form of a time series. Specifically, using the first historical operating data, various types of data of the photovoltaic power station can be sorted in chronological order to form a data vector containing multiple features. The data at each time point constitutes a time series vector, and the vectors of multiple time points are synthesized into a multidimensional time series vector, which can reflect the operating characteristics of the photovoltaic power station in different time periods and the dynamic situation of the electricity market. For example, light intensity (S), temperature (T), and predicted maximum output of the photovoltaic power station (P) are extracted from the cleaned historical operating data. pv ), actual online power consumption (P grid ), electricity spot market price (J dj ), construct the first multidimensional time series vector X = [S, T, P pv 、P grid 、J dj ], which is used to characterize the weather conditions, output capacity, actual grid-connected power and market electricity prices of photovoltaic power stations.

[0041] Step S103 : performing cluster analysis on the first multi-dimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station.

[0042] Typical scenario data is obtained through cluster analysis and processing of multidimensional time series vectors. It is used to characterize the typical characteristics of PV power plant operations and the electricity market. Specifically, when performing cluster analysis on the first multidimensional time series vector, an appropriate clustering algorithm (such as K-means or DBSCAN) is preselected, and similar time series vectors are grouped into the same cluster based on the characteristics of the PV power plant operation data. Through cluster analysis, typical and representative operation scenarios can be extracted from a large amount of historical data. These clustering results are typical scenario data, reflecting the typical operation modes of PV power plants in different environments.

[0043] Step S104 , based on typical scenario data and preset energy storage charging and discharging strategies, the energy storage configuration parameters of the photovoltaic power station are optimized to obtain an optimal configuration solution.

[0044] Energy storage charging and discharging strategies refer to the rules and policies established for the charging and discharging operations of energy storage systems within photovoltaic power plants. Specifically, the preset energy storage charging and discharging strategy sets the photovoltaic system to operate in maximum power point tracking mode, with actual power consumption prioritizing grid connection demand, and the remaining power used to charge the energy storage. When the energy storage system falls below rated capacity and the electricity purchase price from the grid is below the purchase threshold, the energy storage system purchases electricity from the grid. When the photovoltaic output capacity reaches zero and the grid connection price is above the set price threshold, the energy storage system discharges.

[0045] Energy storage configuration parameters are various parameters used to describe the characteristics and operating status of the energy storage system. For example, they may include energy storage capacity, charging and discharging power, peak discharge price threshold, off-peak grid charging price threshold, etc. The target optimization parameters refer to the optimal parameter values ​​obtained after optimizing the energy storage configuration parameters. Specifically, based on typical scenario data, an optimization model can be established in combination with preset energy storage charging and discharging strategies to simulate the charging and discharging behavior of the energy storage system under different operating scenarios. Through optimization algorithms (such as linear programming, genetic algorithms, etc.), the various energy storage configuration parameters of the energy storage system are adjusted to enable the photovoltaic power station to achieve the best economic benefits and operating efficiency in different scenarios, and ultimately obtain the target optimization parameters.

[0046] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention constructs a first multidimensional time series vector reflecting the comprehensive operating characteristics of the photovoltaic power station and the characteristics of the electricity market through first historical operating data, eliminating the deviation of the theoretical model. Cluster analysis is used to extract representative typical scenario data, avoiding the configuration inaccuracy caused by missing scenarios. Parameter optimization is combined with preset charging and discharging strategies, and the parameters are forced to meet engineering constraints and market rules, thereby realizing a closed-loop optimization process that is data-driven, scenario-adaptive, and strategy-coordinated, and improving the practicality and feasibility of the configuration solution.

[0047] In this embodiment, a photovoltaic power station energy storage optimization configuration method is provided, which can be used for computer equipment, such as servers, Figure 2 FIG. 1 is a flow chart of a method for optimizing energy storage configuration in a photovoltaic power station according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0048] Step S201: Acquire the first historical operation data of the photovoltaic power station. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0049] Step S202: Using the first historical operation data, construct a first multi-dimensional time series vector corresponding to the photovoltaic power station. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0050] Step S203 : performing cluster analysis on the first multi-dimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station.

[0051] Specifically, the above step S203 includes:

[0052] Step S2031 : Divide the first multi-dimensional time series vector according to the time dimension to obtain multiple time series vector subsets.

[0053] The time dimension refers to the time unit for dividing the multidimensional time series vector. Multiple time series vector subsets refer to multiple data sets formed by the multidimensional time series vector data divided according to the time dimension. Specifically, the first multidimensional time series vector is divided according to the time dimension, and the first historical operation data after the division is respectively used as an independent unit. The multidimensional time series vector of each division result of the first historical operation data is normalized, thereby forming multiple data sets, that is, obtaining multiple time series vector subsets. For example, when the time dimension is monthly and the historical operation data is one year of data, the first multidimensional time series vector is divided according to the time dimension of the month, and the first historical operation data of the 12 months of the year are respectively used as independent units. The first multidimensional time series vector of each month is normalized, thereby forming 12 data sets, each data set corresponding to one month of time series vector data, that is, obtaining multiple time series vector subsets.

[0054] Step S2032: for any time series vector subset, determine the optimal number of clusters corresponding to the time series vector subset using a time series clustering method.

[0055] The time series clustering method refers to a method of clustering time series vector subsets using the improved time series clustering algorithm DTW-Kmeans. The optimal number of clusters refers to the optimal number of clusters for each time series vector subset determined by using the silhouette score (SSE) as the main evaluation indicator. Specifically, for any time series vector subset, the improved time series clustering algorithm DTW-Kmeans is used, with the silhouette coefficient as the main evaluation indicator. For each possible number of clusters k, after performing K-means clustering, the average distance a(i) of each sample point i in the subset to other samples in its own cluster and the average distance b(i) to the nearest cluster are calculated to obtain the sample's silhouette coefficient s(i), and then the overall silhouette coefficient S(k) is calculated, and the k corresponding to the maximum value of S(k) is selected as the optimal number of clusters for the time series vector subset. For example, for each sample point i, its silhouette coefficient s(i) is defined as:

[0056]

[0057] Where a(i) is the average distance from sample i to other samples in its own cluster, and b(i) is the average distance from sample i to its nearest neighbor. For each value of k, after performing K-means clustering, the mean of all samples s(i) is calculated and recorded as the overall silhouette coefficient S(k):

[0058]

[0059] Select k corresponding to the maximum value of the silhouette coefficient as the optimal number of clusters:

[0060]

[0061] Step S2033: clustering the time series vector subsets according to the optimal number of clusters to generate typical scenario data corresponding to the photovoltaic power station.

[0062] Based on the determined optimal number of clusters, DTW-Kmeans clustering is performed on the subset of time series vectors. During the clustering process, the DTW-Kmeans algorithm assigns each time series vector to a different cluster based on the similarity of the time series. Each cluster represents a time series dataset with similar operating modes, thereby generating typical scenario data corresponding to a PV power plant.

[0063] In some optional implementations, the above step S2033 includes:

[0064] Step a1: cluster the time series vector subset according to the optimal number of clusters to obtain multiple clusters.

[0065] Multiple clusters refer to the division of a subset of time series vectors into multiple different groups after DTW-Kmeans clustering according to the optimal number of clusters. The time series vectors within each group have high similarity, and these groups are multiple clusters. Specifically, the improved time series clustering algorithm DTW-Kmeans is used to iteratively cluster the normalized time series vector subsets based on the optimal number of clusters k' determined by the silhouette coefficient. The algorithm measures the similarity between time series vectors based on the dynamic time warping (DTW) distance and divides the time series vector subset into k' clusters, maximizing the similarity of the dynamic characteristics of the time series vectors within each cluster and maximizing the differences between different clusters.

[0066] Step a2: for any cluster, determine the dynamic time warping distance between the first vector corresponding to the unit duration and the second vector corresponding to the remaining duration within the cluster.

[0067] Unit duration refers to a preset time unit, for example, it can be any day in the cluster cluster, that is, each day in each cluster cluster can be regarded as an independent time unit. The first vector refers to the multidimensional time series vector corresponding to a unit duration in the cluster cluster. The remaining duration refers to all other unit durations in the cluster cluster except the current unit duration, for example, it can be other days in the cluster cluster except any day. The second vector refers to the multidimensional time series vector corresponding to each unit duration in the remaining duration. Dynamic time warping distance refers to the nonlinear similarity between vectors calculated using the dynamic time warping algorithm. Specifically, the nonlinear alignment distance between the first vector and each second vector in the time series is calculated by the DTW algorithm. This distance can measure the similarity between the two in time series characteristics such as light intensity, temperature, and photovoltaic output. The specific formula is as follows:

[0068]

[0069] Among them, C k is the total number of data points in class k, d represents the reference point, and d j represents the j data point, Indicates that from C k Select the reference point d to minimize the average dynamic time warping distance. This reference point d is the selected target point d. typical .

[0070] Step a3: Obtain a third vector corresponding to the minimum value among the multiple dynamic time warping distances, and determine the third vector as the target vector of the cluster.

[0071] The third vector refers to the unit duration vector with the smallest average dynamic time warping distance to all other unit durations in the cluster. The target vector refers to the third vector determined as the typical representative vector of the cluster. Specifically, for the first vector of each unit duration in the cluster, its dynamic time warping distance with all the second vectors in the cluster is calculated and the average is taken to obtain the average dynamic time warping distance of the unit duration. The first vector with the smallest average dynamic time warping distance is selected as the third vector, because it is most similar to the timing characteristics of the remaining durations in the cluster, so it is determined as the target vector of the cluster as a representative of the typical operating characteristics of the cluster.

[0072] Step a4: using the target vector to determine the typical scene data corresponding to the photovoltaic power station.

[0073] After clustering the time series vector subsets, the target vector of each cluster and the proportion of days in its corresponding cluster are extracted as weights. A weighted algorithm is then used to synthesize the final typical day vector for the target duration (e.g., a month). The typical day vectors of multiple target durations are concatenated to form the basic operating scenario for the entire year, i.e., the typical scenario data.

[0074] In this implementation, the mean dynamic time warping distance between each vector within a cluster and the remaining vectors is calculated to accurately quantify temporal morphological similarity. The most representative real-world date vectors within a cluster are selected using the minimum mean dynamic time warping distance as a criterion, ensuring that the target scenario accurately reflects the temporal fluctuation characteristics of the original data. This constructs representative scenario data based on representative vectors from each cluster, significantly improving the authenticity and interpretability of these typical scenarios.

[0075] In some optional implementations, the above step a4 includes:

[0076] Step a41 : determining the ratio of the unit duration of each cluster to the corresponding target duration, and obtaining the target weight corresponding to each cluster.

[0077] The number of unit durations refers to the number of unit durations contained in a single cluster, and the target duration refers to the total number of days in each time dimension. For example, when the month is the time dimension, the target duration is the total number of days in the month. The target weight refers to the ratio of the number of unit durations of each cluster to the target duration. Specifically, the number of unit durations (such as days) contained in each cluster is counted, and then the target duration (such as the total number of days) of the time dimension (such as the month) to which the cluster belongs is obtained, and the ratio is calculated using the formula "target weight = number of unit durations of clusters / target duration". For example, if the total number of days in a month is 30 and a cluster contains 6 days, then its target weight is 6 / 30 = 0.2, which represents the probability of occurrence of the cluster within the target duration.

[0078] Step a42 : Using the target weights corresponding to the clusters, weighted averaging is performed on the target vectors of the clusters belonging to the target duration to obtain duration scene data corresponding to the target duration.

[0079] Duration scenario data refers to the typical scenario data obtained by weighted averaging the target vectors of each cluster within the same target duration (such as one month). Specifically, for all clusters within the target duration (such as one month), the target vector (first multidimensional time series vector) of each cluster is multiplied by its target weight, and then all the product results are accumulated to obtain the weighted average comprehensive vector. For example, when the target duration is one month, the formula for determining the duration scenario data is:

[0080]

[0081] Among them, X weighted is the duration scene data (matrix form); |C m ∣ is the number of days in the mth cluster; n is the total number of days in the month; is the target vector of the mth cluster.

[0082] Step a43 : splicing the scene data of each duration in chronological order to obtain typical scene data corresponding to the photovoltaic power station.

[0083] The data of each duration scenario are spliced ​​in time series in chronological order to form typical scenario data. For example, when the target duration is one month, the duration scenario data of each of the 12 months are spliced ​​in time series in monthly order (January to December) to form the typical scenario data of regular operation throughout the year, that is, the typical scenario data.

[0084] In this implementation, by weighting the representative target vectors using the actual number of days in the cluster as weights, the resulting probability distribution of duration scenario data closely matches historical reality. The weighted results for each time period are then concatenated chronologically to construct representative scenario data for a complete cycle. This significantly improves the accuracy of the scenario set's statistical representation of actual operating conditions while preserving temporal continuity.

[0085] Step S204: Based on the typical scenario data and the preset energy storage charging and discharging strategy, the energy storage configuration parameters of the photovoltaic power station are optimized to obtain the optimal configuration solution. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0086] The photovoltaic power plant energy storage optimization configuration method provided by the present invention specifically addresses the issue of data characteristic differences across time periods by partitioning the first multidimensional time series vector into subsets based on the time dimension. The optimal number of clusters is dynamically determined for each subset, avoiding the scene distortion caused by a fixed number of clusters. Based on customized clustering results, typical scenario data is generated, significantly improving the time series representativeness and clustering quality of these typical scenarios.

[0087] In this embodiment, a photovoltaic power station energy storage optimization configuration method is provided, which can be used for computer equipment, such as servers, Figure 3 FIG. 1 is a flow chart of a method for optimizing energy storage configuration in a photovoltaic power station according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0088] Step S301: Acquire the first historical operation data of the photovoltaic power station. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.

[0089] Step S302: Using the first historical operation data, construct a first multi-dimensional time series vector corresponding to the photovoltaic power station. Figure 2 Step S202 of the illustrated embodiment will not be described in detail here.

[0090] Step S303: Perform cluster analysis on the first multi-dimensional time series vector to generate typical scene data corresponding to the photovoltaic power station. Figure 2 Step S203 of the illustrated embodiment will not be described in detail here.

[0091] Step S304 , based on typical scenario data and preset energy storage charging and discharging strategies, the energy storage configuration parameters of the photovoltaic power station are optimized to obtain an optimal configuration solution.

[0092] Specifically, the above step S304 includes:

[0093] Step S3041: Obtain energy storage system parameters and decision variable parameters of the photovoltaic power station.

[0094] Energy storage system parameters refer to the inherent technical parameters of the energy storage system itself, such as the maximum storage capacity of the energy storage system, the maximum charging and discharging power, the service life of the energy storage system, the charging and discharging efficiency and cost parameters. Decision variable parameters refer to variables that need to be dynamically adjusted during the optimization process. Specifically, energy storage system parameters can be obtained through engineering design data or historical operation data. The initial values ​​of decision variable parameters can be set based on industry standards or historical cases, for example, they may include energy storage capacity (MWh), charging and discharging power (MW), peak discharge price threshold (yuan / MWh), valley grid charging price threshold (yuan / MWh), etc., and then iteratively adjusted through the optimization algorithm.

[0095] Step S3042 , under preset constraints, using energy storage system parameters, typical scenario data, and energy storage charging and discharging strategies, optimizes resource consumption parameters and resource replacement parameters of the photovoltaic power station to obtain optimal resource consumption parameters and optimal resource replacement parameters.

[0096] The preset constraints refer to the physical and operational restrictions that must be met in the optimization model, such as energy balance constraints, charge and discharge mutual exclusion constraints, charge and discharge power constraints, etc. The resource consumption parameter refers to the economic target that needs to be optimized, that is, the average annual net benefit of energy storage. The resource replacement parameter refers to the technical target that needs to be optimized, that is, the photovoltaic absorption rate. Specifically, with the average annual net benefit of energy storage (resource consumption parameter) and the photovoltaic absorption rate (resource replacement parameter) as the optimization targets, under the constraints of energy balance, SOC, charge and discharge power, the energy storage charging and discharging process is simulated based on typical scenario data: photovoltaics are connected to the grid first, the remaining power is charged, the grid purchases electricity when the electricity price is low, and discharges when the electricity price is peak. The average annual net benefit N of energy storage under different parameter combinations is calculated through time series simulation. benefit and photovoltaic absorption rate λ pv , filter out the parameter combination that maximizes both, and obtain the optimal resource consumption parameters and the optimal resource replacement parameters.

[0097] Specifically, the average annual net benefit of energy storage = the average annual benefit from participating in peak-valley arbitrage - the average annual construction cost - the operation and maintenance cost, which can be expressed as:

[0098] maxNbenefit =R arbitrage -C CP -C OM ;

[0099] Among them, N benefit is the average annual net benefit of energy storage, C CP is the average annual construction cost of energy storage, C OM is the average annual operation and maintenance cost, R arbitrage The average annual return of energy storage arbitrage.

[0100] Specifically, the average annual construction cost of energy storage = (cell cost + system cost + EPC cost) / energy storage life, which is expressed as:

[0101] C CP =(C bat +C sys +C epc ) / design_years;

[0102] Among them, C bat is the cost of energy storage cells, C sys is the energy storage system cost, C epc is the EPC cost, and design_years is the energy storage life.

[0103] Specifically, the energy storage operation and maintenance cost = (replacement component cost + labor cost) / energy storage life, which is expressed as:

[0104] C OM =(C rep +C lab ) / design_years;

[0105] Among them, C rep C is the cost of replacing energy storage system components (such as battery replacement, etc.), lab For labor costs.

[0106] Specifically, the benefits of energy storage participating in peak-valley arbitrage are:

[0107]

[0108] Among them, J peak,t is the electricity selling price (yuan / MWh), that is, the price at which the energy storage system sells electricity when the electricity price is high, Q dis,t is the energy storage discharge capacity in time period t (MWh), and η is the energy storage energy conversion efficiency.

[0109] Specifically, the photovoltaic absorption rate λ pv for:

[0110]

[0111] Among them, Q pv,ab is the amount of abandoned light from the photovoltaic power station (MWh), Q pv is the total electricity generated by the photovoltaic power station (MWh).

[0112] Specifically, the constraints include:

[0113] (1) Energy balance constraints:

[0114]

[0115] Among them, E t-1 It refers to the energy storage system power at time t-1 (MWh), E t refers to the energy storage system power at time t (MWh), Q ch,t-1 refers to the charging capacity at time t-1 (MWh), Q dis,t-1 Refers to the discharge power at time t-1 (MWh), η ch Refers to the charging efficiency (for example, it can be 85%), η dis It refers to the discharge efficiency (for example, it can be 85%).

[0116] (2) Charge and discharge mutual exclusion constraints:

[0117] P ch,t ·P dis,t =0;

[0118] Among them, P ch,t is the charging power at time t (MW), P dis,t is the discharge power at time t (MW).

[0119] (3) SOC constraint:

[0120]

[0121] SOC min ≤SOC t ≤SOC max ;

[0122] Among them, SOC t is the state of charge (SOC) of the energy storage system at time t, E t is the energy storage system power at time t (MWh), E max is the maximum capacity of the energy storage system (MWh), SOC min is the minimum SOC (e.g. 5%), SOC max is the maximum SOC (which may be 97%, for example).

[0123] (4) Energy storage charging and discharging power constraints:

[0124] 0≤P ch,t ≤P ch,max ;

[0125] 0≤P dis,t ≤P dis,max ;

[0126] Among them, P ch,t is the charging power at time t (MW), P ch,max is the maximum charging power (MW), P dis,t is the discharge power at time t (MW), P dis,max is the maximum discharge power (MW).

[0127] (5) Constraints on photovoltaic power station output distribution:

[0128]

[0129] Among them, Q ch,t is the charging capacity at time t (MWh), Q pv,t is the available photovoltaic power generation at time t, Q grid,t is the amount of electricity consumed online at time t.

[0130] (6) Energy storage discharge triggering conditions:

[0131]

[0132] Among them, Q dis,t is the energy storage discharge capacity in time period t, J t is the electricity price in period t, J peak is the on-grid price threshold.

[0133] (7) Energy storage charging and discharging time constraints:

[0134] Energy storage capacity / power = {0.1, 0.2, 0, 25, 0.3, 0.4, 0.5, 1, 2, 2.5, 3, 4, 5}.

[0135] Step S3043 , under the conditions of the optimal resource consumption parameters and the optimal resource replacement parameters, the decision variable parameters are tuned using a preset genetic algorithm to obtain target decision variable parameters.

[0136] The target decision variable parameters refer to the optimal combination of decision variables obtained through optimization using a preset genetic algorithm, namely, parameters such as energy storage capacity, charge and discharge power, and peak-valley electricity price thresholds that simultaneously meet the requirements for maximizing both the average annual net return on energy storage and the highest photovoltaic absorption rate. Specifically, a two-stage collaborative optimization algorithm, NSGA-II-GA, is employed. The decision variable parameters are first defined, parameter ranges and constraints are set, and a mathematical model is constructed with the optimal net return and photovoltaic absorption rate as the objective functions. After initializing the population, selection is performed through non-dominated sorting and congestion calculations. A new population is generated using hybrid crossover and Gaussian mutation. After multiple generations of evolution, a Pareto optimal solution set is generated, ultimately outputting a combination of decision variable parameters that meets the engineering requirements.

[0137] For example, a four-dimensional decision space is constructed, representing energy storage capacity (MWh), charging and discharging power (MW), peak-time discharge price threshold (yuan / MWh), and off-peak grid charging price threshold (yuan / MWh). Parameter ranges and constraints are set. A mathematical model for a dual-objective optimization function is established to maximize net revenue and maximize the photovoltaic integration rate. The population is initialized. Fitness is evaluated, a penalty function mechanism is constructed, and target values ​​are calculated through time series simulation. Genetic operators are configured, using tools.cxBlend for hybrid crossover and constraint processing; tools.mutGaussian for Gaussian mutation and constraint processing; and tools.selNSGA2 for non-dominated sorting and congestion selection. The optimization algorithm is executed, gradually optimizing the population through multiple generations of evolution to generate a Pareto-optimal solution set. Constraint checking and processing are embedded in the population initialization, crossover, and mutation processes to ensure that all solutions meet engineering requirements. The optimal configuration solution is obtained, and the Pareto frontier is output after the number of iterations is reached.

[0138] Step S3044: Determine the optimal configuration solution using the target decision variable parameters.

[0139] When the genetic algorithm reaches a preset number of iterations, the optimal solution that takes into account both the average annual net benefit of energy storage and the photovoltaic absorption rate is selected from the Pareto frontier solution set. The decision variable parameters corresponding to this solution, such as energy storage capacity, charging and discharging power, peak discharge price threshold, and valley charging price threshold, are the target optimization parameters that meet the requirements of economy and robustness. These can be directly used in the configuration plan of the photovoltaic power station energy storage system to obtain the optimal configuration plan.

[0140] The photovoltaic power plant energy storage optimization configuration method provided by the present invention utilizes a hierarchical optimization framework to first independently optimize resource parameters under preset constraints based on system parameters and scenario data, ensuring economic viability. Using the optimized resource parameters as fixed conditions, a genetic algorithm is then employed to specifically tune decision variables, enabling efficient global search of complex decision spaces. Finally, the results of these two optimization phases are integrated to generate target parameters, significantly improving optimization efficiency and solution quality while ensuring project feasibility.

[0141] In some optional implementations, the above photovoltaic power station energy storage optimization configuration method further includes:

[0142] Step b1: obtaining second historical operating data of the photovoltaic power station under extreme weather scenarios.

[0143] The second type of historical operational data refers to the historical operational data of PV power plants under extreme weather conditions. Specifically, dates defined as extreme weather events (including sandstorms, heavy rain, and snowstorms) were screened from the historical meteorological database of the region where the PV power plants are located. The full-dimensional operational data corresponding to these dates was extracted. Data quality was ensured through data cleaning (outlier removal and missing value interpolation), resulting in an independent dataset containing only extreme weather days.

[0144] Step b2: constructing a second multi-dimensional time series vector of the photovoltaic power station under extreme weather scenarios using the second historical operation data.

[0145] The second multidimensional time series vector is a confounded time series vector constructed based on the second historical operating data under extreme weather scenarios. Specifically, the cleaned data from extreme weather days is organized chronologically to construct a second multidimensional time series vector with the same structure as the first multidimensional time series vector. The vector data only comes from extreme weather days, forming a set of time series vectors independent of conventional scenarios, which serves as the input for extreme scenario cluster analysis.

[0146] Step b3: Determine the energy storage backup capacity of the photovoltaic power station using the second multi-dimensional time series vector.

[0147] Energy storage backup capacity refers to the additional energy storage capacity configured on top of conventional energy storage capacity to cope with special situations such as extreme weather, and is used to ensure the stable operation of photovoltaic power stations under extreme weather conditions. Specifically, the second multidimensional time series vector is input into the same optimization process: DTW-Kmeans clustering is used to generate typical days for extreme scenarios; with extreme scenarios as input, a dual-objective optimization model (maximizing net income and absorption rate) is run, considering the same charging and discharging strategies and constraints (such as SOC, rate); the energy storage capacity value that exceeds the conventional configuration in the optimization result is the backup capacity, which is used to cope with sudden drops in output under extreme weather conditions (such as sandstorms causing P pv When it approaches 0, additional capacity is required for power supply).

[0148] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention uses the second historical operating data of the photovoltaic power station under extreme weather scenarios to construct a second multidimensional time series vector, and calculates the energy storage backup capacity based on this second multidimensional time series vector. This enables the energy storage configuration solution to have active defense capabilities against abnormal climate conditions and enhances the system's operational robustness in real complex environments.

[0149] In some optional implementations, the above photovoltaic power station energy storage optimization configuration method further includes:

[0150] Step c1: obtaining the output capacity data, actual grid power data, and power resource data of the photovoltaic power station in the current period.

[0151] The photovoltaic power station's real-time monitoring system collects output capacity data, actual grid-connected power data, and power resource data for the current period. Output capacity data is provided by the photovoltaic array power prediction system and is the theoretical maximum power generation (unit: MW) calculated based on parameters such as light intensity, temperature, and component efficiency. Actual grid-connected power data is the actual amount of power uploaded to the grid (unit: MWh) collected in real time by smart meters at the grid connection point. Power resource data is the real-time node electricity price (unit: RMB / MWh) obtained from the power market trading platform, reflecting the current power supply and demand situation and market value.

[0152] In step c2, if the difference between the output capacity data and the online power data is greater than the first preset threshold, it is determined whether the energy storage capacity of the photovoltaic power station has reached the second preset threshold; if the energy storage capacity has not reached the second preset threshold, the energy storage charging operation is performed; if the energy storage capacity has reached the second preset threshold, the photovoltaic power curtailment operation is performed.

[0153] like Figure 4 As shown in the figure, when the output capacity data minus the actual grid power data exceeds the first preset threshold, it indicates that the photovoltaic power generation far exceeds the current grid demand. If the current energy storage power is less than the second preset threshold, the energy storage charging operation is initiated to charge the excess photovoltaic power into the energy storage system. If the energy storage power is greater than or equal to the second preset threshold, the photovoltaic power curtailment operation is executed, actively limiting the photovoltaic output to the actual grid demand value to prevent grid overload.

[0154] Step c3: If the difference between the output capacity data and the online power data is less than the first preset threshold, it is determined whether the remaining energy storage capacity of the photovoltaic power station reaches the third preset threshold; if the energy storage capacity does not reach the third preset threshold, the photovoltaic power curtailment operation is performed; if the energy storage capacity reaches the third preset threshold, it is determined whether the power resource data reaches the fourth preset threshold; if the power resource data reaches the fourth preset threshold, the energy storage discharge operation is performed; if the power resource data does not reach the fourth preset threshold, the energy storage static operation is performed.

[0155] like Figure 4 As shown in the figure, when the output capacity data minus the actual grid-connected power data is less than the first preset threshold, it indicates that the PV power generation cannot meet the grid-connected demand. If the energy storage power is less than or equal to the third preset threshold, the PV power curtailment operation is executed. Since the energy storage cannot fill the gap, the actual grid-connected power is directly limited to the PV output capacity value.

[0156] If the energy storage capacity exceeds the third preset threshold, the system checks the power resource data. If the real-time electricity price exceeds the fourth preset threshold, the system initiates the energy storage discharge operation. If the real-time electricity price is less than the fourth preset threshold, the system executes the energy storage standby operation, retaining the energy storage capacity until a higher electricity price period occurs, and the photovoltaic system is connected to the grid according to its actual output capacity.

[0157] The photovoltaic power station energy storage optimization configuration method provided by the embodiment of the present invention triggers energy storage operation by comparing the difference between the output capacity and the actual grid power in real time, and sets the energy storage capacity threshold as the basis for secondary decision-making, forming a dynamic response mechanism with dual threshold control. This not only effectively utilizes surplus photovoltaic power generation to reduce abandoned light, but also avoids the risk of energy storage overcharging, thereby improving the safety of system operation and resource utilization. Through a three-level conditional hierarchical decision-making mechanism, the output gap is first determined, then the remaining energy storage capacity is checked, and finally the discharge is triggered by coupling with power resource data, realizing refined energy management under abnormal operating conditions, maximizing economic benefits while ensuring energy storage safety, and significantly improving the system's intelligent response in low-output scenarios.

[0158] This embodiment also provides a photovoltaic power plant energy storage optimization configuration device, which is used to implement the above-mentioned embodiments and preferred implementations. Details already described are not repeated here. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0159] This embodiment provides a photovoltaic power station energy storage optimization configuration device, such as Figure 5 Shown, including:

[0160] A first acquisition module 501 is configured to acquire first historical operating data of a photovoltaic power station;

[0161] A first construction module 502 is configured to construct a first multi-dimensional time series vector corresponding to the photovoltaic power station using the first historical operation data;

[0162] A clustering module 503 is configured to perform cluster analysis on the first multi-dimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station;

[0163] The optimization module 504 is used to optimize the energy storage configuration parameters of the photovoltaic power station based on typical scenario data and preset energy storage charging and discharging strategies to obtain the best configuration solution.

[0164] In some optional implementations, the clustering module 503 includes:

[0165] a partitioning submodule, configured to partition the first multidimensional time series vector according to a time dimension to obtain a plurality of time series vector subsets;

[0166] A first determination submodule is configured to determine, for any subset of time series vectors, the optimal number of clusters corresponding to the subset of time series vectors using a time series clustering method;

[0167] The clustering submodule is used to cluster the time series vector subsets according to the optimal number of clusters to generate typical scenario data corresponding to the photovoltaic power station.

[0168] In some optional implementations, the clustering submodule includes:

[0169] The clustering unit is used to cluster the time series vector subsets according to the optimal number of clusters to obtain multiple clusters;

[0170] A first determining unit is configured to determine, for any cluster, a dynamic time warping distance between a first vector corresponding to a unit duration and a second vector corresponding to a remaining duration within the cluster;

[0171] an acquiring unit, configured to acquire a third vector corresponding to a minimum value among the plurality of dynamic time warping distances, and determine the third vector as a target vector of the cluster;

[0172] The second determining unit is configured to determine typical scene data corresponding to the photovoltaic power station using the target vector.

[0173] In some optional implementations, the second determining unit includes:

[0174] Determine the subunits, which are used to determine the ratio of the unit duration of each cluster to the corresponding target duration, and obtain the target weight corresponding to each cluster;

[0175] The weighting subunit is used to use the target weights corresponding to the clusters to perform weighted averaging on the target vectors of the clusters belonging to the target duration to obtain the duration scene data corresponding to the target duration;

[0176] The splicing subunit is used to splice the scene data of each duration in chronological order to obtain the typical scene data corresponding to the photovoltaic power station.

[0177] In some optional implementations, the optimization module 504 includes:

[0178] The acquisition submodule is used to obtain the energy storage system parameters and decision variable parameters of the photovoltaic power station;

[0179] The first tuning submodule is used to optimize the resource consumption parameters and resource replacement parameters of the photovoltaic power station under preset constraints by using energy storage system parameters, typical scenario data, and energy storage charging and discharging strategies to obtain optimal resource consumption parameters and optimal resource replacement parameters;

[0180] The second tuning submodule is used to tune the decision variable parameters using a preset genetic algorithm under the conditions of optimal resource consumption parameters and optimal resource replacement parameters to obtain target decision variable parameters;

[0181] The second determination submodule is used to determine the optimal configuration solution using the target decision variable parameters.

[0182] In some optional embodiments, the photovoltaic power station energy storage optimization configuration device further includes:

[0183] A second acquisition module is used to obtain second historical operation data of the photovoltaic power station under extreme weather scenarios;

[0184] A second construction module is used to construct a second multi-dimensional time series vector of the photovoltaic power station under an extreme weather scenario using the second historical operation data;

[0185] The determination module is used to determine the energy storage backup capacity of the photovoltaic power station using the second multi-dimensional time series vector.

[0186] In some optional embodiments, the photovoltaic power station energy storage optimization configuration device further includes:

[0187] The third acquisition module is used to obtain the output capacity data, actual grid power data and power resource data of the photovoltaic power station in the current period;

[0188] A first determination module is configured to determine whether the energy storage capacity of the photovoltaic power station reaches a second preset threshold if the difference between the output capacity data and the online power data is greater than a first preset threshold;

[0189] The first execution module is configured to execute an energy storage charging operation if the energy storage capacity does not reach a second preset threshold; and to execute a photovoltaic power curtailment operation if the energy storage capacity reaches the second preset threshold.

[0190] In some optional embodiments, the photovoltaic power station energy storage optimization configuration device further includes:

[0191] The second determination module is configured to determine whether the remaining energy storage capacity of the photovoltaic power station reaches a third preset threshold if the difference between the output capacity data and the online power data is less than the first preset threshold;

[0192] a second execution module, configured to execute a photovoltaic power curtailment operation if the energy storage capacity does not reach a third preset threshold;

[0193] The third execution module is used to determine whether the power resource data reaches a fourth preset threshold if the energy storage capacity reaches a third preset threshold; if the power resource data reaches the fourth preset threshold, perform an energy storage discharge operation; if the power resource data does not reach the fourth preset threshold, perform an energy storage static operation.

[0194] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0195] The photovoltaic power station energy storage optimization configuration device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0196] The photovoltaic power station energy storage optimization configuration device provided by the embodiment of the present invention constructs a first multidimensional time series vector reflecting the comprehensive operating characteristics of the photovoltaic power station and the characteristics of the electricity market through first historical operating data, eliminating the deviation of the theoretical model. Cluster analysis is used to extract representative typical scenario data, avoiding the configuration inaccuracy caused by missing scenarios. Parameter optimization is combined with preset charging and discharging strategies, forcing parameters to meet engineering constraints and market rules, thereby realizing a closed-loop optimization process that is data-driven, scenario-adaptive, and strategy-coordinated, and improving the practicality and feasibility of the configuration solution.

[0197] The embodiment of the present invention also provides a computer device having the above Figure 5 The photovoltaic power station energy storage optimization configuration device shown.

[0198] See also Figure 6 , Figure 6 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0199] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0200] The memory 20 stores instructions that can be executed by at least one processor 10, so as to enable at least one processor 10 to execute the method shown in the above embodiment.

[0201] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0202] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0203] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0204] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0205] A portion of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0206] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A photovoltaic power station energy storage optimization configuration method, characterized in that: The method comprises: Obtain the first historical operating data of the photovoltaic power station; constructing a first multi-dimensional time series vector corresponding to the photovoltaic power station using the first historical operation data; performing cluster analysis on the first multidimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station; Based on the typical scenario data and the preset energy storage charging and discharging strategy, the energy storage configuration parameters of the photovoltaic power station are optimized to obtain the optimal configuration solution.

2. The method according to claim 1, characterized in that The performing cluster analysis on the first multi-dimensional time series vector to generate typical scenario data corresponding to the photovoltaic power station includes: Dividing the first multidimensional time series vector according to the time dimension to obtain multiple time series vector subsets; For any of the time series vector subsets, determine the optimal number of clusters corresponding to the time series vector subset using a time series clustering method; The time series vector subsets are clustered according to the optimal number of clusters to generate typical scenario data corresponding to the photovoltaic power station.

3. The method according to claim 2, characterized in that Clustering the time series vector subsets according to the optimal number of clusters to generate typical scenario data corresponding to the photovoltaic power station includes: Clustering the time series vector subset according to the optimal number of clusters to obtain multiple clusters; For any of the clusters, determining a dynamic time warping distance between a first vector corresponding to a unit duration and a second vector corresponding to a remaining duration within the cluster; Obtaining a third vector corresponding to a minimum value among the plurality of dynamic time warping distances, and determining the third vector as a target vector of the cluster; The target vector is used to determine typical scene data corresponding to the photovoltaic power station.

4. The method according to claim 3, characterized in that The determining of typical scene data corresponding to the photovoltaic power station using the target vector includes: Determine the ratio of the unit duration of each cluster to the corresponding target duration, and obtain the target weight corresponding to each cluster; Using the target weights corresponding to the clusters, weighted averaging the target vectors of the clusters belonging to the target duration is performed to obtain duration scene data corresponding to the target duration; The respective duration scene data are spliced ​​in chronological order to obtain typical scene data corresponding to the photovoltaic power station.

5. The method according to claim 1, wherein The energy storage configuration parameters of the photovoltaic power station are optimized based on the typical scenario data and the preset energy storage charging and discharging strategy to obtain the optimal configuration solution, including: Obtaining energy storage system parameters and decision variable parameters of the photovoltaic power station; Under preset constraints, using the energy storage system parameters, the typical scenario data, and the energy storage charging and discharging strategy, the resource consumption parameters and resource replacement parameters of the photovoltaic power station are optimized to obtain optimal resource consumption parameters and optimal resource replacement parameters; Under the conditions of the optimal resource consumption parameters and the optimal resource replacement parameters, the decision variable parameters are tuned using a preset genetic algorithm to obtain target decision variable parameters; The optimal configuration solution is determined using the target decision variable parameters.

6. The method according to claim 1, characterized in that Also includes: Acquiring second historical operating data of the photovoltaic power station under an extreme weather scenario; constructing a second multidimensional time series vector of the photovoltaic power station under an extreme weather scenario using the second historical operation data; The energy storage backup capacity of the photovoltaic power station is determined using the second multi-dimensional time series vector.

7. The method according to claim 1, characterized in that Also includes: Obtaining output capacity data, actual grid power data, and power resource data of the photovoltaic power station in the current period; If the difference between the output capacity data and the online power data is greater than a first preset threshold, determining whether the energy storage capacity of the photovoltaic power station reaches a second preset threshold; If the energy storage capacity does not reach the second preset threshold, performing an energy storage charging operation; If the energy storage capacity reaches the second preset threshold, a photovoltaic power curtailment operation is performed.

8. The method according to claim 7, characterized in that Also includes: If the difference between the output capacity data and the online power data is less than the first preset threshold, determining whether the remaining energy storage capacity of the photovoltaic power station reaches a third preset threshold; If the energy storage capacity does not reach the third preset threshold, executing a photovoltaic power abandonment operation; If the energy storage capacity reaches the third preset threshold, determining whether the power resource data reaches a fourth preset threshold; If the power resource data reaches a fourth preset threshold, performing an energy storage discharge operation; If the power resource data does not reach the fourth preset threshold, an energy storage static operation is performed.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the photovoltaic power station energy storage optimization configuration method according to any one of claims 1 to 8 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the photovoltaic power station energy storage optimization configuration method according to any one of claims 1 to 8.

Images