Generation method and system for typical working condition of energy storage unit
By building a joint power plant model and feature evaluation system, a typical charge and discharge curve that meets the real working conditions is generated, which solves the problem of difficult to simulate the real operating status of energy storage power plant batteries in the existing technology, and achieves efficient and safe in the laboratory Performance testing and design optimization improve the reliability and stability of energy storage power plants.
Patent Information
- Application Number
- CN202410960241.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-08
AI Technical Summary
The existing technology is difficult to effectively simulate the complex operation of energy storage power station batteries in real working conditions, resulting in uncertainty and safety risks about the performance of retired batteries, hindering its standardized application and large-scale development in energy storage systems.
By building a joint power station model, obtaining the operating data and characteristic data of the energy storage unit, performing working status classification, dimensionality reduction processing, clustering analysis and optimization algorithm splicing, generating typical charge and discharge curves that meet the real working conditions, and establishing feature evaluation system, verify the characteristics consistency of the generated typical working conditions.
It realizes the real operation of the energy storage power station in the laboratory, reduces experimental costs, improves testing efficiency and safety, provides strong technical support for the design optimization and performance evaluation of the energy storage power station, and discovers potential problems in advance. , verify the rationality of the design plan and improve the reliability and stability of the system.
Smart Images

Figure CN118916794A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of novel energy storage technology, and in particular relates to a method and system for generating typical working conditions of an energy storage unit. Background Art
[0002] With the booming electric vehicle market, the number of retired batteries has surged, becoming an important issue facing the energy sector. Using them in energy storage power stations can not only effectively extend the battery life cycle, but also improve energy efficiency and promote green and sustainable development. However, there are deficiencies in the existing research on the performance of batteries in energy storage power stations, which directly leads to a lack of knowledge about the performance, life expectancy, safety and operating efficiency of retired batteries in energy storage systems. This situation not only exacerbates the uncertainty and potential risks in the process of technology application, but also hinders the standardized application and large-scale development of retired batteries in energy storage power stations due to the lack of unified technical specifications and performance evaluation standards. More seriously, the limited understanding of battery performance may mask safety hazards such as thermal runaway and short circuit, thereby increasing the operating risks of energy storage power stations and limiting their widespread acceptance and application in the market. Therefore, it is necessary to carry out experimental research on batteries for energy storage power stations.
[0003] At present, there are two ways to conduct experimental research on batteries in energy storage power stations. One is to conduct experiments by building a micro energy storage power station. Although this method can directly obtain rich actual operation data, its high cost and strict restrictions for safety considerations often make comprehensive and in-depth battery data collection difficult. The other is to use a charge and discharge tester in the laboratory to charge and discharge single cells at a fixed rate. This method has the advantages of controllable experimental conditions, accurate data collection, safe operation, low cost and high efficiency, but it cannot reflect the situation of retired batteries in the actual working state. Therefore, in view of the improvement of experimental efficiency and the guarantee of operational safety, researchers usually tend to adopt the second experimental method, that is, using a charge and discharge tester to test single cells. At the same time, in order to make up for its limitation that it cannot fully reflect the actual working state, it is particularly important to build a test environment that can simulate the actual operating conditions of energy storage power station batteries. Summary of the invention
[0004] In order to solve the above technical problems, the present invention proposes a method and system for generating typical working conditions of energy storage units. By constructing an energy storage system model in a real scenario and obtaining the charge and discharge curve of a typical working condition, a solution is provided for the planning and configuration of energy storage power stations and the laboratory simulation of power batteries. In the full life cycle experiment, it is of great significance to be able to find characteristic curves that meet the actual BESS working conditions and conduct full life cycle experiments in the laboratory. Therefore, it is necessary to construct a long series of load data and extract the corresponding features to splice into a typical working condition containing most of the working condition characteristics, so that the test of power batteries in energy storage power stations can be transferred to the laboratory with similar working conditions, and it is also more convenient to conduct various characterization tests of batteries, making the extraction of working conditions very necessary.
[0005] To achieve the above object, the present invention provides a method for generating a typical operating condition of an energy storage unit, comprising:
[0006] Establish a combined power plant model, and obtain the operation data and characteristic data of the energy storage unit based on the combined power plant model;
[0007] Classifying the operating data into working states to obtain a plurality of working state segments;
[0008] Using a dimensionality reduction method, the feature data is subjected to dimensionality reduction processing to extract the core feature data of each working state, which is the dimensionality reduction data;
[0009] Performing cluster analysis on the reduced-dimensional data, identifying the inherent patterns and associations in the reduced-dimensional data, obtaining cluster data, and reclassifying the cluster data into respective working states according to a number of working state fragments to form classified data divided by working state; using an optimization algorithm, finding the optimal data fragments in the classified data, splicing the optimal data fragments, and generating a typical working condition of the energy storage unit of the combined power station that meets the working condition time length;
[0010] A characteristic evaluation system is established, and the typical operating condition of the energy storage unit is verified based on the characteristic evaluation system to determine whether the generated typical operating condition of the energy storage unit of the combined power station has characteristics consistent with the operating data.
[0011] Preferably, the plurality of working state segments include: charging scene data, discharging scene data and stationary scene data.
[0012] Preferably, before the dimension reduction processing is performed on the feature data, the method further includes:
[0013] Based on the KMO test, the correlation between the characteristic data of several working states is evaluated, and based on the correlation, the data dimension reduction method is determined.
[0014] Preferably, the process of performing dimensionality reduction processing on the feature data includes:
[0015] The feature data is converted into a low-dimensional space by using a dimensionality reduction method, and the principal component information of the feature data is retained, wherein the principal component information reflects the change trend of the original data.
[0016] Preferably, the process of clustering the dimension-reduced data includes:
[0017] Based on a clustering algorithm, the dimension-reduced data is divided into different clusters to obtain data structure and data distribution.
[0018] Preferably, the fitness function of the optimization algorithm is:
[0019]
[0020] Among them, D(x i ,c i ) represents the representative point c j With data point x i The distance between them, N represents the number of points in the cluster, and M represents the number of points in the selected combination.
[0021] Preferably, the characteristic parameters in the characteristic evaluation system are used to reflect the state characteristics of each charging and discharging segment, wherein the characteristic parameters include: average charging power, average discharging power, maximum charging power, maximum discharging power, maximum average charging capacity, maximum average discharging capacity, minimum average charging capacity and minimum average discharging capacity.
[0022] In order to achieve the above technical objectives, the present invention also provides a system for generating typical working conditions of an energy storage unit, which is used to implement the method for generating typical working conditions of an energy storage unit, and the system includes:
[0023] A model building module is used to establish a combined power plant model, and obtain the operation data and characteristic data of the energy storage unit based on the combined power plant model;
[0024] The state classification module is used to classify the working state of the operation data to obtain a plurality of working state segments.
[0025] A data dimension reduction module is used to perform dimension reduction processing on the feature data respectively by using a dimension reduction method, and extract the core feature data of each working state, namely the dimension reduction data;
[0026] A data optimization module is used to perform cluster analysis on the reduced-dimensional data, identify the inherent patterns and associations in the reduced-dimensional data, obtain cluster data, and reclassify the cluster data into respective working states according to a number of working state fragments to form classified data divided by working state; use an optimization algorithm to find the optimal data fragments in the classified data, splice the optimal data fragments, and generate a typical working condition of the energy storage unit of the combined power station that meets the working condition time length;
[0027] The operating condition verification module is used to establish a characteristic evaluation system, verify the typical operating condition of the energy storage unit based on the characteristic evaluation system, and determine whether the generated typical operating condition of the energy storage unit of the combined power station has characteristics consistent with the operating data.
[0028] The present invention also provides a computer storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method for generating the typical working condition of the energy storage unit is implemented.
[0029] Compared with the prior art, the present invention has the following advantages and technical effects:
[0030] The present invention provides a method for generating a typical working condition of an energy storage unit, comprising: establishing a combined power station model, and obtaining operating data and characteristic data of the energy storage unit based on the combined power station model; classifying the operating data into working state categories to obtain a plurality of working state segments; using a dimensionality reduction method to perform dimensionality reduction processing on the characteristic data, and extracting core characteristic data of each working state, namely, dimensionality reduction data; performing cluster analysis on the dimensionality reduction data, identifying the inherent patterns and associations in the dimensionality reduction data, and obtaining cluster data; and reclassifying the cluster data into respective working states according to a plurality of working state segments to form classified data divided by working state; using an optimization algorithm to find the optimal data segments in the classified data, and splicing the optimal data segments to generate a typical working condition of the combined power station energy storage unit that meets the working condition time length; and establishing a characteristic evaluation system, and verifying the typical working condition of the energy storage unit based on the characteristic evaluation system to determine that the generated typical working condition of the combined power station energy storage unit has characteristics consistent with the operating data.
[0031] The present invention proposes a construction method for test conditions that can be used to simulate energy storage power stations based on an energy storage model constructed using real data. This method overcomes the limitation that it is difficult to simulate the actual complex operating state of an energy storage power station when using a charge and discharge tester to charge and discharge battery cells at a fixed rate in a laboratory environment, and provides an effective way to achieve performance tests that are closer to reality. This not only greatly reduces the need for a large number of experiments in actual energy storage power stations, saves costs, but also significantly improves test efficiency and safety. The application of this test condition construction method provides strong technical support for the design optimization, performance evaluation and fault prediction of energy storage power stations. By simulating the actual operation of energy storage power stations, researchers can discover potential problems in advance and verify the rationality of the design scheme, thereby optimizing system performance and improving the reliability and stability of energy storage power stations. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0033] Figure 1 A schematic diagram of a combined power plant model according to an embodiment of the present invention;
[0034] Figure 2 A flow chart of a method for generating a typical operating condition of an energy storage unit according to an embodiment of the present invention;
[0035] Figure 3 A schematic diagram of BIRCH clustering according to an embodiment of the present invention;
[0036] Figure 4 A genetic algorithm structure diagram of an embodiment of the present invention;
[0037] Figure 5 The error result diagram of energy storage conditions obtained by using different modeling and algorithms in the embodiment of the present invention;
[0038] Figure 6 The following are the comparison results of the typical working condition obtained by the modeling and algorithm with the minimum error in the embodiment of the present invention and the other five. DETAILED DESCRIPTION
[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0040] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0041] Embodiment 1
[0042] like Figure 1 As shown, this embodiment provides a convenient and portable combined power plant model, including the following:
[0043] Photovoltaic power generation module: The local solar energy resources are collected from the NASA ground meteorology and solar energy database. The model uses an 80kW / h photovoltaic array, and its parameters are shown in Table 1 below;
[0044] Table 1
[0045]
[0046] The formula for solar power generation per unit area is as follows:
[0047] P=η×I×cos(θ)×cos(φ) (1)
[0048] Among them, η is the efficiency of the photovoltaic cell, I is the irradiance, θ is the angle between the sun's rays and the ground, and φ is the angle between the sun's rays and the south direction.
[0049] After analysis, the average output power is 13.9kW / h, the power generation efficiency is 17.4%, and the annual power generation is 122160kW / h.
[0050] Wind power generation module: The local wind resource data is collected from local wind turbines. The module uses 18 10kw FD series wind turbines, and its specific parameters are shown in Table 2;
[0051] Table 2
[0052]
[0053] The total wind power generation formula is:
[0054] P w =P ws ×18 (2)
[0055] Where P ws The power generation corresponding to the wind speed at the current moment.
[0056] After analysis, the average wind power output throughout the year was 56.1kW / h, the power generation efficiency was 31.2%, and the annual power generation was 491,265kW / h.
[0057] Energy storage power station module: The energy storage power station uses lithium-ion batteries as energy storage equipment, and selects 60 groups of 1kw energy storage equipment to form 30 series and 2 parallel energy storage power station modules. The parameters of a single 1kw energy storage power station are shown in Table 3;
[0058] Table 3
[0059]
[0060]
[0061] The module does not consider the impact of battery life degradation, cycle times and other parameters. Assuming that the energy storage unit has good heat dissipation, insulation and ventilation measures, the external temperature effect is not considered during the battery charging and discharging process.
[0062] The annual charging capacity is 10,875kW, the discharging capacity is 9,806kW, and the internal resistance loss is 29.5kW.
[0063] Embodiment 2
[0064] like Figure 2 As shown, this embodiment provides a method for generating a typical operating condition of an energy storage unit, including:
[0065] Step 1: First, build a model for the combined power station to obtain long-term operation data and characteristic data of the energy storage unit;
[0066] Specifically, Homer Pro software is used to model the combined power station, and the long-term operation data of the energy storage unit is obtained by combining the real data of the configured area; the characteristic data include: solar energy, solar panel incident angle, wind speed, wind output energy, storage capacity, state of charge (SOC), etc.
[0067] Step 2: The operating data working state of the energy storage unit obtained in step 1 is divided into a charging scene, a discharging scene, and a stationary scene, and then the feature data in each scene is subjected to dimensionality reduction processing, and the feature data in different states is subjected to dimensionality reduction processing using the kernel principal component analysis method;
[0068] Specifically, the specific implementation steps of the working state division include: first, check the data to see if there is any abnormal data, such as: abnormal battery charge and discharge power, such as overcharge and over discharge, delete the abnormal data, and then interpolate and supplement the deleted abnormal data according to the characteristics of the charge and discharge power curve. Then, segment division is performed, with 0 as the boundary (because when the data is zero, it is in a static state), the operating data is divided into segments, and the segments are processed, and the time length is less than 5 minutes. Delete, because the short time segment does not contain too much feature information.
[0069] Specifically, KPCA principal component analysis performs dimensionality reduction processing on the characteristic data of several working states respectively, and extracts the core characteristic data of each working state, which includes KMO test and kernel principal component analysis;
[0070] The combined power plant model is used to obtain some characteristic data associated with the working fragment. Since these characteristic data have high dimensions and contain a lot of redundancy, direct analysis will lead to high computational complexity and poor subsequent clustering results. Therefore, it is necessary to perform kernel principal component analysis on these data to reduce the dimension and compress the data while retaining the main features and structure of the data. The KMO test and Bartlett sphericity test are used to evaluate the correlation between several characteristic data, and the data dimensionality reduction method is determined based on the correlation.
[0071] The KMO test is used to evaluate the correlation between observed variables. Its value range is 0 to 1. The higher the value, the better the correlation between the variables. Its calculation formula is expressed as:
[0072]
[0073] where r ij is the simple correlation coefficient between the variables, r ij…k is the partial correlation coefficient between variables. The closer the KMO value is to 1, the better the correlation between variables is, and the more suitable it is for kernel principal component analysis.
[0074] Kernel principal component analysis: Kernel principal component analysis converts high-dimensional data into low-dimensional space to retain the main information of the data. The data sets of energy storage systems are highly correlated. KPCA can effectively reduce the correlation between data, ensure that the extracted principal components better capture the changing trends of the original data, simplify the data structure, and reduce the dimensions. Provide a more reliable basis for subsequent analysis and clustering. KPCA not only inherits the advantages of traditional principal component analysis (PCA) in dimensionality reduction, but also effectively solves the problem of nonlinear data dimensionality reduction by introducing the kernel function method. According to the eigenvalue, contribution rate, and cumulative contribution rate of each principal component, it is confirmed that several principal component information is left to represent the information contained in the original feature data.
[0075] In the initial projection process, there are
[0076]
[0077] Derived
[0078]
[0079] Right now
[0080]
[0081] Can be transformed into
[0082]
[0083] For Φ(x i ) introduces the kernel function to represent, and simplifies it to
[0084]
[0085] The kernel function K is
[0086]
[0087] where x i , x j are the eigenvalues of the original matrix, ||x i || is the eigenvalue norm.
[0088] Step 3: Perform cluster analysis on the dimensionality reduction results in step 2, classify the working conditions in different scenarios according to the clustering results, and then use genetic algorithms to find the optimal data fragments in the clustering results, and finally splice them into typical working conditions of the energy storage unit of the combined power station that meet the working condition time length. The structure of clustering processing is as follows Figure 3 shown.
[0089] Specifically, clustering and genetic algorithm optimization are performed on the data after dimensionality reduction in step 2. The number of clusters is determined according to the clustering effect based on the clustering purpose. Auxiliary methods to help confirm the number of clusters include the elbow rule, which finds the point in the curve where the rate of decline slows down significantly, namely the "elbow point". Methods for evaluating clustering effects include silhouette coefficient, Calinski-Harabasz index, David-Bladeyn index, etc. The number of clusters can also be confirmed based on the clustering effect.
[0090] BIRCH uses a structure called CF-tree, which has the characteristics of efficiently and dynamically organizing data;
[0091] CF nodes satisfy the following linear relationship:
[0092] CF1+CF2=(N1+N2,L1+L2,S1+S2) (10)
[0093] That is, for each CF node in the parent node, the value of its (N, L, S) triple is equal to the sum of the triplets of all the child nodes it points to.
[0094] The central Manhattan distance was chosen as the distance metric:
[0095]
[0096] In order to find the typical representative value of each cluster after classification, a genetic algorithm is introduced to maximize the similarity between the data points in each cluster and its representative value to obtain more representative typical values. Since the points near the cluster center do not necessarily represent the global optimal clustering results, the genetic algorithm is used to generate 100 groups of data in the iterative process to ensure that each group of data is retained in its respective cluster and meets the constraints of the time combination. These data combinations are then ranked using the silhouette coefficient as an evaluation indicator. The top ten data combinations are selected as the parents of the next generation, and the iterations are repeated until the termination conditions are met. These fragments can better represent each cluster. By comparing these optimized fragments with the original data, the best data for typical working conditions can be found. The genetic algorithm flow chart is as follows Figure 4 As shown in the figure, first initialize the various parameters required for setting the genetic algorithm, determine the initial typical value and its encoding method, randomly generate the initial population t, set the maximum number of iterations, and if the maximum number of iterations is reached, jump to the "decoding" step, otherwise calculate the fitness of the individuals in the population; perform genetic operations from left to right to update the population, and check whether the population meets the predetermined termination condition. If not, proceed to the next iteration until the termination condition is met, otherwise jump to the "decoding" step and output the typical representative value of each cluster finally selected.
[0097] Its fitness function is
[0098]
[0099] Where D(x i ,c i ) represents the representative point c j With data point x i The distance between them, N represents the number of points in the cluster, and M represents the number of points in the selected combination.
[0100] Step 4: Establish a feature evaluation system and perform verification to ensure that the generated results can represent the original working conditions.
[0101] Specifically, some characteristic parameters are selected to reflect the state characteristics of each charge and discharge segment, and then evaluate whether the final curve is reasonable. Based on the following formula:
[0102]
[0103] The characteristic parameters of the characteristic evaluation system are, from top to bottom, average charging power, average discharging power, maximum charging power, maximum discharging power, maximum average charging capacity, maximum average discharging capacity, minimum average charging capacity, and minimum average discharging capacity.
[0104] Where n is the number of charging segments under the operating conditions, m is the number of discharging segments under the operating conditions, and T is the duration of the charging and discharging segments.charge,i ) is the fitting curve of the i-th charging segment. c ,t2 c Indicates the start and end time of maximum charging. t3 c ,t4 c Indicates the start and end time of minimum charging.
[0105] Figure 5 is the error result diagram calculated based on the defined features; Figure 6 This is a comparison chart of the typical operating condition obtained by the modeling and algorithm provided in this embodiment and the other five.
[0106] Embodiment 3
[0107] This embodiment provides a system for implementing the method for generating a typical operating condition of an energy storage unit, the system comprising:
[0108] A model building module is used to establish a combined power plant model, and obtain the operation data and characteristic data of the energy storage unit based on the combined power plant model;
[0109] The state classification module is used to classify the working state of the operation data to obtain a plurality of working state segments.
[0110] A data dimension reduction module is used to perform dimension reduction processing on the characteristic data respectively by using a kernel principal component analysis method, and extract the core characteristic data of each working state, namely, the dimension reduction data;
[0111] A data optimization module is used to perform cluster analysis on the reduced-dimensional data, identify the inherent patterns and associations in the reduced-dimensional data, obtain cluster data, and reclassify the cluster data into respective working states according to a number of working state fragments to form classified data divided by working state; use a genetic algorithm to find the optimal data fragments in the classified data, splice the optimal data fragments, and generate a typical working condition of the energy storage unit of the combined power station that meets the working condition time length;
[0112] The operating condition verification module is used to establish a characteristic evaluation system, verify the typical operating condition of the energy storage unit based on the characteristic evaluation system, and determine whether the generated typical operating condition of the energy storage unit of the combined power station has characteristics consistent with the operating data.
[0113] Embodiment 4
[0114] This embodiment further provides a computer storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method for generating a typical operating condition of an energy storage unit is implemented.
[0115] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for generating typical working conditions of an energy storage unit, It is characterized in that The following steps are involved: Establish a combined power plant model, and obtain the operation data and characteristic data of the energy storage unit based on the combined power plant model; Classifying the operating data into working states to obtain a plurality of working state segments; Using a dimensionality reduction method, the feature data is subjected to dimensionality reduction processing to extract the core feature data of each working state, which is the dimensionality reduction data; Performing cluster analysis on the reduced-dimensional data, identifying the inherent patterns and associations in the reduced-dimensional data, obtaining cluster data, and reclassifying the cluster data into respective working states according to a number of working state fragments to form classified data divided by working state; using an optimization algorithm, finding the optimal data fragments in the classified data, splicing the optimal data fragments, and generating a typical working condition of the energy storage unit of the combined power station that meets the working condition time length; A characteristic evaluation system is established, and the typical operating condition of the energy storage unit is verified based on the characteristic evaluation system to determine whether the generated typical operating condition of the energy storage unit of the combined power station has characteristics consistent with the operating data.
2. The method for generating typical operating conditions of an energy storage unit according to claim 1, It is characterized in that The several working state segments include: charging scene data, discharging scene data and stationary scene data.
3. The method for generating typical working conditions of an energy storage unit according to claim 1, It is characterized in that Before the dimension reduction process is performed on the feature data, the method further includes: Based on the KMO test, the correlation between the characteristic data of several working states is evaluated, and based on the correlation, the data dimension reduction method is determined.
4. The method for generating typical operating conditions of an energy storage unit according to claim 1, It is characterized in that The process of performing dimensionality reduction processing on the feature data includes: The feature data is converted into a low-dimensional space by using a dimensionality reduction method, and the principal component information of the feature data is retained, wherein the principal component information reflects the change trend of the original data.
5. The method for generating typical working conditions of an energy storage unit according to claim 1, It is characterized in that The process of clustering the dimension-reduced data includes: Based on a clustering algorithm, the dimension-reduced data is divided into different clusters to obtain a data structure and a data distribution, wherein the number of clusters is determined according to the clustering purpose and the clustering effect.
6. The method for generating typical working conditions of an energy storage unit according to claim 1, It is characterized in that The fitness function of the optimization algorithm is: Among them, D(x i ,c i ) represents the representative point c j With data point x i The distance between them, N represents the number of points in the cluster, and M represents the number of points in the selected combination.
7. The method for generating typical operating conditions of an energy storage unit according to claim 1, It is characterized in that The characteristic parameters in the characteristic evaluation system are used to reflect the state characteristics of each charging and discharging segment, wherein the characteristic parameters include: average charging power, average discharging power, maximum charging power, maximum discharging power, maximum average charging capacity, maximum average discharging capacity, minimum average charging capacity and minimum average discharging capacity.
8. A system for generating typical working conditions of energy storage units, It is characterized in that A method for generating a typical operating condition of an energy storage unit according to any one of claims 1 to 7, the system comprising: A model building module is used to establish a combined power plant model, and obtain the operation data and characteristic data of the energy storage unit based on the combined power plant model; The state classification module is used to classify the working state of the operation data to obtain a plurality of working state segments. A data dimension reduction module is used to perform dimension reduction processing on the feature data respectively by using a dimension reduction method, and extract the core feature data of each working state, namely the dimension reduction data; A data optimization module is used to perform cluster analysis on the reduced-dimensional data, identify the inherent patterns and associations in the reduced-dimensional data, obtain cluster data, and reclassify the cluster data into respective working states according to a number of working state fragments to form classified data divided by working state; use an optimization algorithm to find the optimal data fragments in the classified data, splice the optimal data fragments, and generate a typical working condition of the energy storage unit of the combined power station that meets the working condition time length; The operating condition verification module is used to establish a characteristic evaluation system, verify the typical operating condition of the energy storage unit based on the characteristic evaluation system, and determine whether the generated typical operating condition of the energy storage unit of the combined power station has characteristics consistent with the operating data.
9. A computer storage medium, It is characterized in that The computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method for generating a typical operating condition of an energy storage unit as described in any one of claims 1 to 7 is implemented.
