Energy storage system energy management method based on demand response

Abstract

The invention discloses a demand response-based energy management method for an energy storage system, and relates to the field related to electric energy management, and the method comprises the steps: collecting and obtaining a characteristic data set and a power generation data set, and carrying out the correlation mapping, and obtaining an energy storage system characteristic-power generation data set; demand influence factor extraction is carried out on the energy storage system characteristic-power generation data set, a demand side influence factor set is determined, clustering analysis is carried out on the energy storage system characteristic-power generation data set, and a multi-energy storage system power generation data cluster is obtained; performing power generation control fitting on the power generation data cluster of the multi-energy storage system according to the demand side influence factor set, and establishing multiple energy control channels; and monitoring demand side response data of the target energy storage system in real time, performing matching analysis by adopting energy control multiple channels, determining power generation control parameters, and performing energy feedback control. The technical problems that an existing energy storage system is poor in control strategy adaptability and insufficient in response precision in energy management are solved, and the technical effect of improving the control strategy adaptability and the response precision is achieved.

Description

Technical Field

[0001] The present application relates to fields related to electric energy management, and in particular to an energy management method for an energy storage system based on demand response. Background Art

[0002] Energy management in energy storage systems plays a crucial role in improving grid stability, reducing electricity costs, and promoting renewable energy integration. Currently, this field primarily employs fixed-threshold control strategies or single-objective optimization algorithms for power generation scheduling. These approaches fail to fully account for the dynamic nature of demand-side response, resulting in insufficient control accuracy, delayed response, and limited energy efficiency optimization.

[0003] In the current related technologies, energy management of energy storage systems has technical problems such as poor adaptability of control strategies and insufficient response accuracy. Summary of the Invention

[0004] This application provides an energy management method for energy storage systems based on demand response. It adopts parallel collection of energy storage system characteristic data sets and power generation data sets, generates a fused data set through association mapping, and simultaneously performs demand influencing factor extraction (obtaining an influencing factor set) and cluster analysis (generating power generation data clusters). The two are input into the power generation control fitting module to construct a multi-channel energy control model. By real-time monitoring of demand-side response data, matching and parsing the optimal power generation control parameters, and performing energy feedback control and other technical means, the application solves the technical problems of poor adaptability of control strategies and insufficient response accuracy in the energy management of existing energy storage systems, and achieves the technical effect of improving the adaptability of control strategies and response accuracy.

[0005] The present application provides an energy management method for an energy storage system based on demand response, comprising: collecting and acquiring an energy storage system characteristic data set and an energy storage system power generation data set, and performing correlation mapping on the energy storage system characteristic data set and the energy storage system power generation data set to obtain an energy storage system characteristic-power generation data set; extracting demand influencing factors from the energy storage system characteristic-power generation data set to determine a demand-side influencing factor set, and performing cluster analysis on the energy storage system characteristic-power generation data set to obtain multiple energy storage system power generation data clusters; performing power generation control fitting on the multiple energy storage system power generation data clusters according to the demand-side influencing factor set to establish a multi-channel energy control for the energy storage system; monitoring the demand-side response data of the target energy storage system in real time, using the energy storage system energy control multi-channel to match and analyze the demand-side response data of the target energy storage system, determining energy storage system power generation control parameters, and performing energy feedback control based on the energy storage system power generation control parameters.

[0006] In a possible implementation, the energy storage system characteristic-power generation dataset is obtained by performing the following processing: initializing a noise filter based on noise characteristic information of the energy storage system characteristic dataset and the energy storage system power generation dataset; using the noise filter to filter and de-noise the energy storage system characteristic dataset and the energy storage system power generation dataset to obtain a usable energy storage system characteristic dataset and a usable energy storage system power generation dataset; downsampling the usable energy storage system characteristic dataset and the usable energy storage system power generation dataset according to data application requirements to obtain a reduced-dimensionality energy storage system characteristic dataset and a reduced-dimensionality energy storage system power generation dataset; and performing data alignment and association mapping on the reduced-dimensionality energy storage system characteristic dataset and the reduced-dimensionality energy storage system power generation dataset according to time series to obtain the energy storage system characteristic-power generation dataset.

[0007] In a possible implementation, the determination of the demand-side influencing factor set includes the following processing: constructing a demand-side influencing factor system, the demand-side influencing factor system including time factors, meteorological factors, electricity price factors, and load factors; extracting associated influencing factors from the energy storage system characteristics-power generation data set according to the demand-side influencing factor system to obtain a demand-side associated influencing factor set; performing principal component analysis and importance scoring on the demand-side associated influencing factor set based on the energy storage system characteristics-power generation data set to obtain a set of associated influencing factor importance coefficients; and screening the demand-side associated influencing factor set based on the associated influencing factor importance coefficient set to determine the demand-side influencing factor set.

[0008] In a possible implementation, the demand-side associated influencing factor set is screened based on the associated influencing factor importance coefficient set to determine the demand-side influencing factor set, and the following processing is performed: hierarchical analysis and weight assignment are performed on each factor type in the demand-side influencing factor system to generate an influencing factor system weight factor; based on the influencing factor system weight factor, the influencing factor type screening proportion is determined; based on the associated influencing factor importance coefficient set, the demand-side associated influencing factor set is proportionally screened according to the influencing factor type screening proportion to determine the demand-side influencing factor set.

[0009] In a possible implementation, the multi-energy storage system power generation data cluster is obtained by performing the following processing: performing feature extraction and dimensionality reduction processing on the energy storage system characteristic-power generation data set to obtain a key feature set of energy storage power generation; determining the energy storage power generation feature dimension based on the key feature set of energy storage power generation; performing initial K-means clustering on the energy storage system characteristic-power generation data set according to the energy storage power generation feature dimension to obtain an initial power generation data cluster; and performing silhouette coefficient evaluation and iterative strategy optimization on the initial power generation data cluster to obtain the multi-energy storage system power generation data cluster.

[0010] In a possible implementation, the energy storage system energy control multi-channel is constructed by performing the following processing: identifying the power generation demand of the multiple energy storage system power generation data clusters according to the demand-side influencing factor set to obtain the multiple energy storage system power generation sample clusters; setting the energy storage system energy control tasks according to the energy storage system energy management objectives; and performing power generation control fitting on the multiple energy storage system power generation sample clusters based on the energy storage system energy control tasks to construct the energy storage system energy control multi-channel.

[0011] In a possible implementation, the power generation control fitting is performed on the multiple energy storage system power generation sample clusters based on the energy storage system energy control task, and a multi-channel energy control of the energy storage system is established, and the following processing is performed: based on the energy storage system energy control task, the power generation data of the multiple energy storage system power generation sample clusters are sequentially associated to obtain a data set of energy storage task associated sample clusters; and the power generation control fitting is performed on each of the energy storage task associated sample cluster data sets to establish a multi-channel energy storage system energy control.

[0012] In a possible implementation, the power generation control fitting is performed on the sample cluster data sets associated with the energy storage task respectively, and a multi-channel energy control of the energy storage system is constructed, and the following processing is performed: according to the energy control task of the energy storage system, an energy storage task network structure is selected; the power generation control fitting is performed on the sample cluster data sets associated with the energy storage task respectively using the energy storage task network structure to generate a multi-energy storage cluster energy control branch channel set; the multi-energy storage cluster energy control branch channel set is labeled according to the multi-energy storage system power generation data cluster, and a multi-channel energy storage system energy control is constructed.

[0013] In a possible implementation, the multi-energy storage cluster energy control branch channel set is identified according to the multi-energy storage system power generation data cluster, and an energy storage system energy control multi-channel is constructed, and the following processing is performed: the multi-energy storage cluster energy control branch channel set is classified and identified according to the multi-energy storage system power generation data cluster to obtain a multi-energy storage cluster associated branch channel set; the multi-energy storage cluster associated branch channel sets are respectively integrated in parallel to construct an energy storage system energy control multi-channel, wherein the number of channels of the energy storage system energy control multi-channel corresponds one-to-one to the multi-energy storage system power generation data cluster.

[0014] In a possible implementation, the energy feedback control is performed based on the power generation control parameters of the energy storage system, and the following processing is performed: energy management monitoring of the target energy storage system is performed based on the power generation control parameters of the energy storage system to obtain energy storage control feedback parameters; a PID controller is used to adjust and correct the power generation control parameters of the energy storage system based on the energy storage control feedback parameters, and energy management of the energy storage system is performed using the corrected power generation control parameters of the energy storage system.

[0015] The energy management method for energy storage systems based on demand response proposed in this application first collects and acquires an energy storage system characteristic dataset and an energy storage system power generation dataset, associates and maps the energy storage system characteristic dataset and the energy storage system power generation dataset to obtain an energy storage system characteristic-power generation dataset, then extracts demand influencing factors from the energy storage system characteristic-power generation dataset to determine a set of demand-side influencing factors, and simultaneously performs cluster analysis on the energy storage system characteristic-power generation dataset to obtain multiple energy storage system power generation data clusters. Then, power generation control fitting is performed on the multiple energy storage system power generation data clusters according to the set of demand-side influencing factors, and a multi-channel energy control system is established. Finally, the demand-side response data of the target energy storage system is monitored in real time, and the energy storage system energy control multi-channel is used to match and analyze the demand-side response data of the target energy storage system, determine the energy storage system power generation control parameters, and perform energy feedback control based on the energy storage system power generation control parameters. This achieves the technical effect of improving the adaptability of the control strategy and the response accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the methods according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A flow chart of a demand response-based energy storage system energy management method provided in an embodiment of the present application.

[0018] Figure 2 A schematic diagram of a process for obtaining an energy storage system characteristic-power generation data set in a demand response-based energy storage system energy management method provided in an embodiment of the present application.

[0019] Figure 3 A schematic diagram of a process for determining a set of demand-side influencing factors in a demand-response-based energy storage system energy management method provided in an embodiment of the present application.

[0020] Figure 4 A schematic diagram of a process for screening a set of demand-side associated influencing factors in the demand response-based energy storage system energy management method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0022] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0023] In the following description, reference is made to “some embodiments” which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms “including” and “having” and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0024] The embodiment of the present application provides an energy management method for an energy storage system based on demand response, such as Figure 1 As shown, the method includes: Step S100 , acquiring an energy storage system characteristic data set and an energy storage system power generation data set, and performing correlation mapping on the energy storage system characteristic data set and the energy storage system power generation data set to obtain an energy storage system characteristic-power generation data set.

[0025] Specifically, the energy storage system characteristic dataset contains various attribute data of the energy storage device, which can reflect the performance characteristics and operating characteristics of the energy storage device. The energy storage system power generation dataset includes relevant data of the energy storage system during the power generation process, which is used to describe the power generation performance of the energy storage system under different conditions.

[0026] Various sensors are used to collect the energy storage system's characteristic data (such as the capacity, self-discharge rate, charge and discharge efficiency, and cycle life of the energy storage device) and power generation data (such as power generation power and power generation time). These data are collected and preliminarily organized using a data acquisition card. Then, through data association mapping, the energy storage system characteristic dataset and the energy storage system power generation dataset are associated to generate the energy storage system characteristic-power generation dataset.

[0027] For example, for an energy storage system containing multiple battery packs, current sensors, voltage sensors, temperature sensors, etc. installed on the battery packs collect real-time characteristic data such as the battery's charge and discharge current, voltage, and temperature. At the same time, power sensors are used to collect the battery pack's power generation data. After aggregating this data, association mapping is used to associate and bind the characteristic data of different battery packs at the same time with the power generation data, forming a complete energy storage system characteristic-power generation data set.

[0028] like Figure 2 As shown, in one possible implementation, the step S100 of obtaining the energy storage system characteristic-generation dataset further includes step S110 of initializing a noise filter based on the noise characteristic information of the energy storage system characteristic dataset and the energy storage system generation dataset. Specifically, the noise characteristic information of the energy storage system characteristic dataset and the generation dataset is analyzed, such as the noise type (Gaussian noise, impulse noise, etc.), noise intensity, and frequency range. Based on this information, an appropriate filter type (such as a Kalman filter or a wavelet transform filter) is selected and filter parameters are initialized, for example, setting the filter's initial state and noise covariance matrix.

[0029] For example, if Gaussian noise is present in the energy storage system's power generation data and the noise intensity is low, a Kalman filter can be used. Based on prior knowledge of the data, the Kalman filter's initial state estimate is set to match the initial value of the actual data, and the noise covariance matrix can be initialized based on empirical noise intensity.

[0030] Step S120: Filter and de-noise the energy storage system characteristic dataset and the energy storage system power generation dataset using the noise filter to obtain an available energy storage system characteristic dataset and an available energy storage system power generation dataset. Specifically, the initialized noise filter is applied to the energy storage system characteristic dataset and the power generation dataset. For each data sample, a recursive filter algorithm is used to filter and remove noise components from the data, thereby obtaining an available energy storage system characteristic dataset and an available energy storage system power generation dataset.

[0031] For example, a Kalman filter is used to filter the energy storage system's power generation data. At each time point, the Kalman filter uses the previous state estimate and the current measurement value to calculate the current power generation estimate through the Kalman filter's prediction update and measurement update steps. This removes Gaussian noise from the original data and produces more accurate power generation data.

[0032] Step S130: Downsample the available energy storage system characteristic dataset and the available energy storage system power generation dataset according to data application requirements to obtain a reduced-dimensionality energy storage system characteristic dataset and a reduced-dimensionality energy storage system power generation dataset. Specifically, downsample the available energy storage system characteristic dataset and the available energy storage system power generation dataset according to set data application requirements, such as data storage space limitations and computational complexity requirements. A simple uniform sampling interval method or a non-uniform sampling method based on data characteristics can be used to reduce the data sampling rate to an appropriate level to obtain the reduced-dimensionality energy storage system characteristic dataset and the reduced-dimensionality energy storage system power generation dataset.

[0033] For example, if the sampling frequency of the original energy storage system characteristic data is 100 Hz, and the data application requirements require a sampling frequency of 10 Hz, the uniform sampling interval method can be used to select one data point every 10 sampling points, thereby reducing the data sampling rate to 10 Hz and achieving data dimensionality reduction.

[0034] In step S140, the reduced-dimensionality energy storage system characteristic dataset and the reduced-dimensionality energy storage system power generation dataset are aligned and associated with each other in a time series to obtain the energy storage system characteristic-power generation dataset. Specifically, the reduced-dimensionality energy storage system characteristic dataset and the reduced-dimensionality energy storage system power generation dataset are aligned in a time series to ensure that each data sample in the two datasets corresponds to the same time point. Then, an association mapping algorithm, such as principal component analysis (PCA) or linear regression, is used to establish an association between the two datasets, ultimately obtaining the energy storage system characteristic-power generation dataset.

[0035] For example, after dimensionality reduction, energy storage system characteristic data (such as battery capacity and self-discharge rate) and power generation data (such as power generation and generation time) are arranged in time series, aligning data samples at the same time point in time and space. Then, a linear regression algorithm is used, using the energy storage system characteristic data as the independent variable and the power generation data as the dependent variable, to establish a linear relationship model between the two, achieving correlation mapping and obtaining the energy storage system characteristic-power generation dataset.

[0036] This implementation method effectively removes the noise components in the data by filtering and denoising the original data. By lowering the data sampling rate, the data volume is reduced, the data storage space requirements and computational complexity are reduced, and the efficiency of subsequent data processing and analysis is improved.

[0037] Step S200 , extracting demand influencing factors from the energy storage system characteristic-power generation data set to determine a demand-side influencing factor set, and performing cluster analysis on the energy storage system characteristic-power generation data set to obtain a multi-energy storage system power generation data cluster.

[0038] Specifically, natural language processing and feature extraction algorithms are used to analyze and process the data in the energy storage system characteristics-power generation dataset, extracting demand-side influencing factors (factors that influence demand-side electricity consumption behavior and demand). These factors directly or indirectly affect the magnitude and changing trends of electricity load, such as time (electricity demand characteristics in different time periods), weather (the impact of temperature, light, humidity, etc. on electricity demand), and user behavior (such as industrial users' production hours and residential users' electricity usage habits), thereby forming a set of demand-side influencing factors. Clustering algorithms (such as K-means clustering and DBSCAN clustering) are also used to perform cluster analysis on the energy storage system characteristics-power generation dataset. Based on the similarities and differences in the data, the data is divided into multiple energy storage system power generation data clusters, each representing a category of cases with similar power generation and energy storage characteristics.

[0039] For example, given energy storage system power generation data under different weather conditions and time periods, a natural language processing algorithm analyzes and extracts the timestamps and weather information in the data, identifying time (such as daytime, nighttime, peak hours) and weather (sunny, cloudy, rainy, etc.) as demand-side influencing factors. Then, using the K-means clustering algorithm, this data is clustered based on characteristics such as the fluctuation range of generated power and the changes in the charge and discharge status of the energy storage device. This results in data clusters with high generated power and frequent energy storage device discharge during sunny daytime periods of peak electricity demand, and low generated power and energy storage device charging during rainy nights and off-peak periods of electricity demand.

[0040] like Figure 3As shown, in one possible implementation, the step S200 of determining a set of demand-side influencing factors further includes step S210, which constructs a demand-side influencing factor system. The demand-side influencing factor system includes time factors, meteorological factors, electricity price factors, and load factors. Specifically, the demand-side influencing factor system is clearly defined, where time factors include different time periods of the day (such as peak and off-peak periods), weekdays and non-workdays, and seasonal changes. For example, a day is divided into three periods: peak electricity consumption (8:00-12:00, 18:00-22:00), flat periods (6:00-8:00, 12:00-18:00, 22:00-24:00), and off-peak period (0:00-6:00), while distinguishing between weekdays and weekends, as well as the four seasons of spring, summer, autumn, and winter. Meteorological factors include meteorological conditions such as temperature, humidity, light intensity, and wind speed. For example, hourly data such as temperature, humidity, and solar radiation are recorded to reflect the impact of weather conditions on the energy storage system and electricity demand. Electricity pricing factors include time-of-use pricing, real-time pricing, and tiered pricing. For example, information on time-of-use pricing published by power grid companies is available, including rates for different time periods and pricing data adjusted in real time based on market supply and demand. Load factors include different types of loads (such as industrial, commercial, and residential loads) and their varying characteristics. For example, load fluctuations caused by industrial users' production shifts, peak loads during business hours for commercial users, and load variations caused by residential users' lifestyles can be analyzed.

[0041] In step S220, the energy storage system characteristic-power generation data set is subjected to associated influencing factors extraction according to the demand-side influencing factor system to obtain a demand-side associated influencing factor set. Specifically, data mining and association rule extraction algorithms, such as the Apriori algorithm or FP-Growth algorithm in association rule mining, are used to analyze the energy storage system characteristic-power generation data set according to the constructed demand-side influencing factor system to identify various influencing factors related to the demand side and form a demand-side associated influencing factor set. For example, through analysis, it is found that in hot weather (meteorological factors) and during peak electricity consumption periods (time factors), the load (load factor) of a specific industrial user will increase significantly. At the same time, the electricity price (electricity price factor) during this period is also high. These factors jointly affect the power generation and operation mode of the energy storage system, and are thus extracted into the demand-side associated influencing factor set.

[0042] Step S230 , based on the energy storage system characteristic-power generation data set, performs principal component analysis and importance scoring on the set of demand-side related influencing factors to obtain a set of importance coefficients of related influencing factors. Specifically, the principal component analysis (PCA) algorithm is used to perform dimensionality reduction processing on the set of demand-side related influencing factors, extract the main components, and simultaneously calculate the importance score of each influencing factor. Specifically, each factor in the set of demand-side related influencing factors is regarded as a variable, and the variance contribution rate and cumulative variance contribution rate of each variable are calculated through PCA to determine the original influencing factors corresponding to the main components. The importance of each influencing factor is scored based on indicators such as the variance contribution rate to obtain a set of importance coefficients of related influencing factors. For example, the analysis found that the variance contribution rate of the time period in the time factor and the temperature in the meteorological factor are two factors with high variance contribution rates, which are key factors affecting the power generation of the energy storage system, and their importance coefficients are relatively large.

[0043] Step S240, based on the set of importance coefficients of the associated influencing factors, the set of associated influencing factors on the demand side is screened to determine the set of influencing factors on the demand side. Specifically, based on the set of importance coefficients of the associated influencing factors, an importance threshold is set, for example, factors with an importance coefficient greater than 0.7 are selected, the set of associated influencing factors on the demand side is screened, and the set of influencing factors on the demand side is finally determined. In this way, the key factors that have a greater impact on the characteristics of the energy storage system - power generation data - can be retained, and the secondary factors with less influence can be removed, making subsequent analysis and control more targeted and efficient. For example, after screening, the time factor (time period), meteorological factor (temperature), electricity price factor (time-of-use electricity price) and load factor (industrial load) are determined as the final set of demand-side influencing factors.

[0044] This implementation method constructs a comprehensive demand-side influencing factor system and performs correlated influencing factor extraction, principal component analysis, and importance scoring. It can accurately screen out key factors that have a significant impact on energy storage system characteristics - power generation data from a large number of possible influencing factors, avoiding the waste of computing resources and increased complexity caused by blindly analyzing and processing all factors.

[0045] like Figure 4As shown, in one possible implementation, the set of demand-side related influencing factors is screened based on the set of related influencing factor importance coefficients to determine the demand-side influencing factor set. Step S240 further includes step S241, performing hierarchical analysis and weighting on each factor type in the demand-side influencing factor system to generate weight factors for the influencing factor system. Specifically, the analytic hierarchy process (AHP) is used to hierarchically divide each factor type (time factors, meteorological factors, electricity price factors, and load factors) in the demand-side influencing factor system, constructing a hierarchical structure model comprising a target layer (determining the demand-side influencing factor set), a criterion layer (each factor type), and an indicator layer (specific factors within each factor type, such as time periods and weekdays / non-workdays within the time factor). A relative importance judgment matrix is ​​obtained through expert scoring, questionnaires, and other methods. AHP calculation methods, such as normalization and consistency testing, are then used to calculate the weights of each factor type to generate the weight factors for the influencing factor system. For example, after calculation, it is found that the weight of the time factor is 0.4, the weight of the meteorological factor is 0.3, the weight of the electricity price factor is 0.2, and the weight of the load factor is 0.1.

[0046] For example, to determine the set of demand-side influencing factors, time, weather, electricity prices, and load factors are used as criterion layers. Experts compare and score each factor in each criterion layer to construct a judgment matrix. For example, time is slightly more important than weather (score 3), time is equally important than electricity prices (score 1), time is very important compared to load (score 5), and so on. The judgment matrix is ​​then normalized and tested for consistency to calculate the weighting factors for each factor. For example, the weight of the time factor is 0.4.

[0047] Step S242: Determine the screening weight of the influencing factor type based on the influencing factor system weight factor. Specifically, determine the screening weight of each factor type in the screening process based on the influencing factor system weight factor. The screening weight refers to the share of each factor type in the demand-side associated influencing factor set and is used to guide the screening operation. For example, based on the size of the weight factor, the time factor has the largest weight, and its screening weight can be determined to be 40% (corresponding to a weight of 0.4), the meteorological factor screening weight is 30% (corresponding to a weight of 0.3), the electricity price factor screening weight is 20% (corresponding to a weight of 0.2), and the load factor screening weight is 10% (corresponding to a weight of 0.1).

[0048] For example, based on the weight factors of the influencing factor system calculated in step S241, the weight factors are directly converted into screening proportions, that is, the screening proportions of time factors are 40%, meteorological factors are 30%, electricity price factors are 20%, and load factors are 10%.

[0049] Step S243 : The set of demand-side related influencing factors is proportionally screened based on the set of related influencing factor importance coefficients according to the screening weights of the influencing factor types to determine the set of demand-side influencing factors. Specifically, the set of demand-side related influencing factors is proportionally screened based on the screening weights of each influencing factor type and the set of related influencing factor importance coefficients. Specifically, for each influencing factor type, the number or proportion of influencing factors to be retained is determined based on its screening weight. For example, in the set of demand-side related influencing factors, if the time factor has 10 related influencing factors, and based on a screening weight of 40% and a ranking of importance coefficients from high to low, the top four (10 × 40%) time factors with high importance coefficients are retained. If the meteorological factor has 8 related influencing factors, the top 2.4 (8 × 30%) are retained, and after rounding, the top three meteorological factors with high importance coefficients are retained. Similarly, the electricity price factor and load factor are screened. Finally, the screened factors of each type are aggregated to determine the final set of demand-side influencing factors.

[0050] For example, in the demand-side influencing factors set, there are 10 specific time factors. After sorting the related influencing factors from high to low importance, the top four factors are retained based on a 40% screening weight. There are 8 meteorological factors, the top three are retained; there are 5 electricity price factors, the top one is retained (5 × 20% = 1); there are 6 load factors, the top 0.6 are retained (6 × 10% = 0.6), and the top one is rounded up. After aggregation, the final set of demand-side influencing factors contains 4 (time) + 3 (meteorological) + 1 (electricity price) + 1 (load) = 9 factors.

[0051] This implementation method assigns weights to various factor types through the hierarchical analysis method, fully considering the differences in the importance of different factor types on the demand side, avoiding the situation where certain relatively important but low-weight factor types may be ignored when screening based solely on the importance coefficient, making the screened set of demand-side influencing factors more comprehensive and reasonable, and more in line with the complex influencing relationships on the actual demand side.

[0052] In one possible implementation, the step S200 of obtaining a multi-energy storage system power generation data cluster further includes step S250, performing feature extraction and dimensionality reduction on the energy storage system characteristic-power generation data set to obtain a key feature set for energy storage power generation. Specifically, feature extraction and dimensionality reduction techniques, such as principal component analysis (PCA) and linear discriminant analysis (LDA), are used to process the energy storage system characteristic-power generation data set to extract key features that represent the primary characteristics of the data, thereby forming a key feature set for energy storage power generation. For example, PCA can be used to convert the original multidimensional data, such as time series data and meteorological data, from the energy storage system power generation data, into a small number of principal components that retain most of the important information in the data, thereby achieving feature extraction and dimensionality reduction.

[0053] For example, suppose the original energy storage system characteristic - power generation dataset contains 10 feature dimensions, such as battery capacity, self-discharge rate, charge and discharge efficiency, cycle life, power generation, power generation time, ambient temperature, light intensity, humidity, and wind speed. PCA analysis shows that the first three principal components can explain more than 90% of the data variance. These three principal components then constitute the key feature set for energy storage power generation.

[0054] Step S260 determines the energy storage power generation feature dimensions based on the energy storage power generation key feature set. Specifically, based on the energy storage power generation key feature set, appropriate energy storage power generation feature dimensions are determined by analyzing indicators such as the variance contribution rate and cumulative variance contribution rate of each principal component. This determines the number of key features used for subsequent cluster analysis. For example, the number of principal components whose cumulative variance contribution rate reaches a certain threshold (e.g., 85%-95%) is selected as the feature dimension.

[0055] For example, in the above PCA analysis, if the cumulative variance contribution rate of the first two principal components has reached 88%, while the variance contribution rate of the third principal component is relatively low, and increasing the feature dimension has a greater impact on the computational complexity of the subsequent cluster analysis, then it can be determined that the energy storage power generation feature dimension is 2.

[0056] Step S270, performing initial K-means clustering on the energy storage system characteristics-power generation data set according to the energy storage power generation feature dimension to obtain an initial power generation data cluster. Specifically, performing initial K-means clustering on the energy storage system characteristics-power generation data set according to the determined energy storage power generation feature dimension. First, randomly initialize K cluster centers, then calculate the distance from each data point to each cluster center, assign each data point to the cluster to which the nearest cluster center belongs, then recalculate the cluster center of each cluster, and repeat the above process until the cluster center no longer changes or the set number of iterations is reached, to obtain the initial power generation data cluster.

[0057] For example, assuming the energy storage power generation feature dimension is determined to be 2, the energy storage system characteristic-power generation dataset is mapped into a two-dimensional space, and three cluster centers are randomly initialized. The Euclidean distance from each data point to the three cluster centers is calculated, and the data point is assigned to the cluster to which the closest cluster center belongs. The coordinates of each cluster center are recalculated, and the data points are assigned again. After several iterations, the cluster centers remain unchanged, resulting in three initial power generation data clusters.

[0058] Step S280, performing a silhouette coefficient evaluation and iterative strategy optimization on the initial power generation data cluster to obtain the multi-energy storage system power generation data cluster. Specifically, the silhouette coefficient is used to evaluate the quality of the initial power generation data cluster. The value range of the silhouette coefficient is [-1, 1]. The larger the value, the better the clustering effect. Based on the silhouette coefficient evaluation results, the initial power generation data cluster is optimized using an iterative strategy, such as adjusting the cluster center, changing the parameters of the clustering algorithm, etc., until the optimal multi-energy storage system power generation data cluster is obtained. For example, if the silhouette coefficient of the initial cluster is low, it can be optimized by reselecting the initial cluster center, increasing the number of clustering iterations, etc., to increase the silhouette coefficient and improve the clustering effect.

[0059] For example, the silhouette coefficient of the initial power generation data cluster is calculated to be 0.5. In order to optimize the clustering effect, multiple K-means clustering is performed using different initial cluster centers. The silhouette coefficient is calculated each time, and the clustering result with the highest silhouette coefficient is finally selected as the final multi-energy storage system power generation data cluster. Assume that the silhouette coefficient after optimization is increased to 0.7.

[0060] This implementation method determines the characteristic dimensions of energy storage power generation based on the key feature set of energy storage power generation, ensuring that the main features and information of the data are retained, while avoiding the introduction of noise and complexity by too many feature dimensions, so that cluster analysis can more accurately reflect the true structure and laws of the data, and improve the reliability and interpretability of the clustering results.

[0061] Step S300 : performing power generation control fitting on the power generation data cluster of the multiple energy storage systems according to the set of demand-side influencing factors, and establishing a multi-channel energy control system for the energy storage system.

[0062] Specifically, power generation control fitting involves building a power generation control model based on existing data and demand-side influencing factors, and then training the model through algorithms to ensure that the model can better adapt to the power generation control requirements of the energy storage system under different demand-side influencing factors. The energy storage system's multi-channel energy control system is a system of multiple independent power generation control channels established for different combinations of demand-side influencing factors. Each channel can achieve energy control of the energy storage system under specific circumstances.

[0063] Based on a set of demand-side influencing factors, different power generation control models are established. These models are trained using machine learning algorithms (such as neural networks and support vector machines), enabling them to fit clusters of energy storage system power generation data based on different demand-side influencing factors. This creates multiple energy storage system energy control channels. Each channel corresponds to a specific power generation control method for the energy storage system under a specific combination of demand-side influencing factors. These channels enable targeted power generation control of the energy storage system.

[0064] For example, multiple power generation control models are established for demand-side factors under different weather and time combinations. For the combination of sunny daytime and peak electricity demand, the model is trained through machine learning algorithms to fit the optimal power generation control method for the energy storage system in this situation. For example, when the power generation is low, the energy storage device will prioritize the release of stored electricity to meet peak electricity demand. At the same time, when the power generation is high and can meet electricity demand, the charging state of the energy storage device is appropriately controlled to avoid overcharging. The channel corresponding to such a model is the power generation control channel of the energy storage system for sunny daytime and peak electricity demand. Similarly, channels corresponding to various other combinations can be established to form a multi-channel energy storage system energy control.

[0065] In one possible implementation, the energy storage system energy control multi-channel is constructed, and step S300 further includes step S310, in which the power generation demand of the multi-energy storage system power generation data cluster is identified according to the demand-side influencing factor set to obtain a multi-energy storage system power generation sample cluster. Specifically, according to the determined demand-side influencing factor set, each data cluster in the multi-energy storage system power generation data cluster is analyzed to determine its corresponding power generation demand characteristics, and the corresponding power generation demand is identified. For example, based on the time factors, meteorological factors, electricity price factors, and load factors in the demand-side influencing factor set, a certain power generation data cluster is identified as a power generation data cluster under the scenario of "peak period, high temperature weather, high electricity price, and high industrial load demand".

[0066] For example, for a cluster containing power generation data from multiple energy storage systems, analysis found that the data in this cluster is mostly concentrated between 2:00 PM and 4:00 PM in the summer (time factor), with temperatures above 35°C (meteorological factor), electricity prices at peak times (price factor), and industrial user loads at peak times (load factor). Therefore, it is identified as the "Summer High Temperature Industrial Peak Load High Electricity Price Power Generation Demand" data cluster.

[0067] Step S320: Set energy control tasks for the energy storage system based on the energy management objectives of the energy storage system. Specifically, energy control tasks for the energy storage system are set based on the energy management objectives of the energy storage system, such as improving grid stability, reducing electricity costs, extending the life of energy storage equipment, and maximizing renewable energy utilization. For example, for the "summer high-temperature industrial peak load and high-price power generation demand" scenario, the energy control task is set as follows: while meeting industrial load demand, prioritize the use of low-priced electricity stored in the energy storage system for discharge during high-price periods, while reasonably controlling the charge and discharge depth of the energy storage equipment to reduce electricity costs and extend equipment life.

[0068] For example, for the data cluster identified above, "Summer High Temperature Industrial Peak Load High Electricity Price Power Generation Demand," the energy control tasks include: fully charging the energy storage equipment during periods of low electricity prices (such as 11:00 PM to 5:00 AM the next day); controlling the energy storage equipment to discharge at maximum power during peak hours of 2:00 PM to 4:00 PM to meet industrial load demand; and ensuring that the charge and discharge depth of the energy storage equipment does not exceed its safety threshold (such as charging not exceeding 90% and discharging not less than 10%) to protect the equipment.

[0069] Step S330: Based on the energy storage system energy control task, power generation control fitting is performed on the multiple energy storage system power generation sample clusters to establish a multi-channel energy storage system energy control. Specifically, based on the set energy storage system energy control task, power generation control fitting is performed for each data cluster in the multiple energy storage system power generation sample clusters using a machine learning algorithm (such as a neural network, support vector machine, etc.) or a control strategy optimization algorithm (such as dynamic programming, genetic algorithm, etc.). By training the model or optimizing the control strategy, the energy storage system can perform precise energy management in accordance with the preset energy control tasks under different power generation demand scenarios, thereby establishing a multi-channel energy storage system energy control. Each channel corresponds to a specific power generation demand scenario and a corresponding control task.

[0070] For example, a neural network algorithm was used to fit power generation control for the data cluster "Summer High Temperature Industrial Peak Load High Electricity Price Power Generation Demand." The neural network model was trained using input variables such as the charge and discharge power and time of the energy storage device, with output objectives such as meeting industrial load demand, reducing electricity costs, and extending equipment life. After training on a large amount of data, the model was able to output the optimal charge and discharge control strategy based on real-time power generation demand and energy storage status, thereby establishing an energy storage system energy control channel tailored to this scenario.

[0071] This approach, by identifying power generation demand across multiple energy storage system power generation data clusters, can clearly identify the characteristics of power generation demand in different scenarios, making the configuration of energy storage system energy control tasks more targeted. Tightly linking control tasks to specific power generation demand scenarios enables the energy storage system to accurately respond to different demand scenarios, improving the sophistication of energy management.

[0072] In one possible implementation, the power generation control fitting is performed on the multiple energy storage system power generation sample clusters based on the energy storage system energy control task, and a multi-channel energy control of the energy storage system is established. Step S330 further includes step S331, in which the power generation data of the multiple energy storage system power generation sample clusters are associated in sequence based on the energy storage system energy control task to obtain a data set of energy storage task associated sample clusters. Specifically, based on the set energy storage system energy control task, the multiple energy storage system power generation sample clusters are analyzed in sequence, and the power generation data in each power generation sample cluster is associated and bound with the corresponding energy control task to form a data set of energy storage task associated sample clusters. This association can be achieved by adding tags, establishing mapping relationships, etc., to ensure that each power generation data sample clearly corresponds to a specific energy control task.

[0073] For example, for the data cluster "Summer high-temperature industrial peak load and high electricity price power generation demand," the corresponding energy control task is "fully charge the energy storage equipment during low electricity price periods, control the energy storage equipment to discharge at maximum power during the peak period of 2:00 PM to 4:00 PM, and ensure that the charge and discharge depth is within the safe threshold." Each power generation data sample in this data cluster is labeled with this energy control task, thus establishing a data set of sample clusters associated with energy storage tasks.

[0074] Step S332 performs power generation control fitting on each of the energy storage task-associated sample cluster datasets to establish multiple energy storage system energy control channels. Specifically, power generation control fitting is performed using a power generation control fitting algorithm (such as linear regression or decision tree algorithms used in machine learning algorithms for fitting control strategies) for each energy storage task-associated sample cluster dataset. By training the model, the energy storage system can output appropriate charge and discharge control parameters (such as charge and discharge power and time) according to the corresponding energy control task based on the characteristic information in the power generation data samples. This allows the energy storage system to establish multiple independent energy storage system energy control channels, each corresponding to a specific energy control task and power generation scenario.

[0075] For example, a linear regression algorithm was used to fit power generation control for the energy storage task-associated sample cluster dataset labeled "Summer High Temperature Industrial Peak Load High Electricity Price Power Generation Demand." The linear regression model was trained using features from the power generation data, such as ambient temperature, industrial load size, and current electricity price, as input variables, and the energy storage device's charge and discharge power, charge and discharge start time, and other output variables. After training on a large number of samples, the model was able to output charge and discharge control parameters that meet the energy control task requirements based on the real-time input power generation data features, thereby establishing the corresponding energy storage system energy control channel.

[0076] This approach correlates power generation data across multiple energy storage system sample clusters, explicitly integrating the energy storage system's energy control tasks with the power generation data. This allows the energy storage system's control strategy to be tightly centered around specific energy control tasks. This enables precise control of the energy storage system in various power generation scenarios, ensuring that the energy storage system operates according to established energy management objectives under various operating conditions.

[0077] In one possible implementation, performing power generation control fitting on each of the energy storage task-associated sample cluster datasets to establish a multi-channel energy storage system energy control system, step S332 further includes step S3321, selecting an energy storage task network structure based on the energy storage system energy control task. Specifically, a suitable energy storage task network structure is selected based on the characteristics and requirements of the energy storage system energy control task, including neural networks, deep learning networks, support vector machines, and the like. For example, a multilayer perceptron (MLP) neural network can be selected for complex nonlinear control tasks; a recurrent neural network (RNN) or a long short-term memory network (LSTM) can be selected for power generation control tasks with time series characteristics.

[0078] For example, if the energy storage system's energy control task is to control charging and discharging based on real-time electricity prices and load demand, and the power generation data has obvious time series characteristics, then the LSTM network is selected as the energy storage task network structure because LSTM can effectively process and predict long-term dependencies in time series data.

[0079] Step S3322: Utilizing the energy storage task network structure, the energy storage task-associated sample cluster datasets are individually trained and fitted for power generation control, generating a multi-energy storage cluster energy control branch channel set. Specifically, utilizing the selected energy storage task network structure, the energy storage task-associated sample cluster datasets are individually trained and fitted. Through iterative training, the network learns the mapping relationship between power generation data and energy control tasks, thereby generating a multi-energy storage cluster energy control branch channel set. Each branch channel corresponds to a specific energy storage task-associated sample cluster dataset and can independently perform power generation control.

[0080] For example, a selected LSTM network is trained on a dataset of sample clusters associated with energy storage tasks labeled "Summer High Temperature Industrial Peak Load High Electricity Price Power Generation Demand." During training, the network uses time series features from the power generation data (such as electricity prices, loads, and ambient temperatures over the past few hours) as input and the charge and discharge power and time of the energy storage devices as output. The network's weights and parameters are continuously adjusted to ensure that the network's predicted output closely matches the actual control task requirements. After sufficient training, the corresponding energy control branch channel is generated. This channel automatically outputs charge and discharge control parameters that meet the energy control task requirements based on the real-time input power generation data features.

[0081] Step S3323: Identify the multi-storage cluster energy control branch channel set according to the multi-storage system power generation data cluster, and establish a multi-channel energy storage system energy control system. Specifically, the generated multi-storage cluster energy control branch channel set is identified according to the multi-storage system power generation data cluster. That is, each branch channel is clearly associated with the corresponding power generation data cluster. This allows for rapid call-up of the corresponding control channel based on the cluster to which the power generation data belongs during actual operation. Ultimately, a multi-channel energy storage system energy control system is established, enabling precise energy management of the energy storage system in different power generation scenarios.

[0082] For example, a trained energy control branch channel corresponding to the "Summer High Temperature Industrial Peak Load High Price Power Generation Demand" data cluster is identified as Channel A. In actual operation, when the monitored energy storage system's power generation data matches the characteristics of this data cluster, the system can quickly identify and invoke Channel A, controlling the energy storage system's charge and discharge according to the control strategy in Channel A to meet the energy management objectives for this scenario.

[0083] This implementation method selects an appropriate network structure based on the energy control tasks of different energy storage systems and performs specialized power generation control fitting on the sample cluster data set associated with each energy storage task. This allows for customized, highly matched control strategies for each power generation scenario, more accurately meeting the energy management needs of different scenarios and improving the operating efficiency and performance of the energy storage system. The generation and identification of multi-storage cluster energy control branch channel sets enables the energy storage system to quickly switch and call corresponding control channels for different power generation data clusters, achieving efficient energy management. When faced with complex and changing power generation environments and demand-side conditions, the system can respond promptly, optimize the charging and discharging process of the energy storage system, reduce energy loss, and improve energy utilization efficiency.

[0084] In one possible implementation, the multi-storage cluster energy control branch channel set is identified according to the multi-storage system power generation data cluster to establish the energy storage system energy control multi-channel. Step S3323 further includes step S33231, in which the multi-storage cluster energy control branch channel set is classified and identified according to the multi-storage system power generation data cluster to obtain a multi-storage cluster associated branch channel set. Specifically, the multi-storage cluster energy control branch channel set is classified and identified according to the multi-storage system power generation data cluster, that is, each branch channel is clearly labeled with its corresponding power generation data cluster category to form a multi-storage cluster associated branch channel set. For example, the energy control branch channel corresponding to the "summer high temperature industrial peak load high electricity price power generation demand" data cluster is identified as category A, and the energy control branch channel corresponding to the "nighttime off-peak load low electricity price" data cluster is identified as category B. Through classification identification, the system can clearly identify the specific power generation scenario targeted by each branch channel.

[0085] For example, after training multiple energy control branch channels, each branch channel is assigned a corresponding category label based on the source power generation data cluster of the sample cluster dataset associated with the energy storage task corresponding to the branch channel. For example, if a branch channel is trained based on the "Spring Weekday Commercial Load Medium Electricity Price Power Generation Demand" data cluster, it is labeled as category C.

[0086] Step S33232, respectively, integrate the branch channel sets associated with the multiple energy storage clusters in parallel to build a multi-channel energy control system for the energy storage system, wherein the number of channels of the multi-channel energy control system for the energy storage system corresponds one-to-one to the power generation data clusters of the multiple energy storage systems. Specifically, respectively integrate the branch channel sets associated with the multiple energy storage clusters in parallel, that is, connect the multiple branch channels in parallel to form a multi-channel energy control system for the energy storage system. Each channel operates independently without interfering with each other, and the input end of each channel receives the corresponding power generation data, and the output end outputs the corresponding energy control instruction. The number of channels of the energy storage system energy control multi-channel finally formed corresponds one-to-one to the power generation data clusters of the multiple energy storage systems, ensuring that the system can quickly and accurately perform energy control through the corresponding channels when facing different power generation scenarios.

[0087] This implementation utilizes a multi-channel architecture in parallel, enabling each energy control channel to operate independently, preventing a single channel failure from impacting the entire system. Even if a channel experiences an anomaly, the others remain operational, ensuring the reliability and stability of the energy storage system and reducing the risk of energy management failures caused by system failures.

[0088] Step S400 , real-time monitoring of the demand-side response data of the target energy storage system, matching and parsing the demand-side response data of the target energy storage system using the energy storage system energy control multi-channel, determining the energy storage system power generation control parameters, and performing energy feedback control based on the energy storage system power generation control parameters.

[0089] Specifically, demand-side response data includes real-time electricity load data, electricity price signals, changes in user electricity demand, and other information. These data reflect the demand-side demand for electricity resources and its dynamic changes. IoT technology is used to monitor the target energy storage system's demand-side response data in real time, such as real-time electricity load changes and electricity price fluctuation signals. This data is then sent to the energy storage system's energy control multi-channel via a data transmission module. In the multi-channel, a matching algorithm (such as a pattern matching algorithm) is used to parse the real-time demand-side response data. Based on the parsing results, the energy storage system's power generation control parameters, such as the energy storage device's charge and discharge power and charge and discharge time, are determined. Then, actuators (such as power regulators and switch control devices) are used to perform energy feedback control on the energy storage system based on these parameters, enabling the energy storage system to respond dynamically to the demand side.

[0090] For example, in a distributed energy system containing multiple energy storage devices, smart meters and monitoring devices installed at the user end capture demand-side response data, such as load changes and electricity price signals, in real time. This data is then transmitted to the energy storage system's multi-channel energy control system in real time. When the load suddenly increases and electricity prices are at peak times, the matching algorithm in the multi-channel will match the corresponding energy storage system power generation control parameters based on the set rules. For example, it will increase the discharge power of the energy storage device to allow it to quickly release energy to meet the increase in electricity demand. At the same time, during periods of lower electricity prices, the energy storage device will be controlled to charge at an appropriate charging power, enabling the energy storage system to implement real-time energy feedback control of demand-side changes.

[0091] In one possible implementation, the energy feedback control is performed based on the energy storage system power generation control parameters, and step S400 further includes step S410, performing energy management monitoring on the target energy storage system based on the energy storage system power generation control parameters to obtain energy storage control feedback parameters. Specifically, based on the energy storage system power generation control parameters, real-time energy management monitoring of the target energy storage system is performed through various sensors and monitoring equipment. The monitoring content includes key indicators such as the energy storage device's power state, charge and discharge power, charge and discharge efficiency, system frequency, voltage, and other parameters related to power generation control. After these monitoring data are collected and organized, they form energy storage control feedback parameters for subsequent control and adjustment.

[0092] For example, in actual operation, a lithium-ion battery energy storage system uses voltage sensors, current sensors, temperature sensors, and other devices installed on the battery pack to monitor battery voltage, current, temperature, and other parameters in real time. Simultaneously, combined with the energy storage system's charge and discharge control parameters (such as the set charge and discharge power and time), key indicators such as the battery's real-time state of charge (SOC) and charge and discharge efficiency are calculated and used as energy storage control feedback parameters.

[0093] In step S420, a PID controller is used to adjust and correct the energy storage system power generation control parameters based on the energy storage control feedback parameters, and the energy storage system energy management is performed using the corrected energy storage system power generation control parameters. Specifically, a PID controller (proportional-integral-differential controller) is used to adjust and correct the energy storage system power generation control parameters. The energy storage control feedback parameters are used as inputs to the PID controller, and the deviation between the set target value (such as the desired state of charge, charge and discharge power, etc.) and the actual feedback value is used as the error input of the PID controller. By adjusting the proportional coefficient (Kp), integral coefficient (Ki), and differential coefficient (Kd) of the PID controller, the corrected energy storage system power generation control parameters are calculated, and the energy storage system is energy managed based on these corrected parameters to achieve precise control of the energy storage system power generation.

[0094] For example, suppose the goal is to maintain the energy storage battery's state of charge (SOC) at around 50%. At a certain moment, the actual SOC is monitored to be 45%, a -5% deviation from the target. This deviation is input into the PID controller, which calculates the corrected charge and discharge power based on the pre-tuned Kp, Ki, and Kd parameters. For example, the PID controller might output a lower charging power to charge the battery at an appropriate rate, gradually restoring the SOC to near the target value. In this way, the energy storage system's charge and discharge processes are adjusted in real time, achieving precise energy management.

[0095] This implementation method can effectively reduce the fluctuations and oscillations of the energy storage system during the charging and discharging process by introducing a PID controller, enhance the stability of the system, and enable the energy storage system to maintain a relatively stable operating state when facing interference factors such as load changes and power generation power fluctuations, thereby avoiding system performance degradation or failure due to excessive parameter fluctuations.

[0096] The embodiment of the present application uses parallel collection of energy storage system characteristic data sets and power generation data sets, generates a fused data set through association mapping, and simultaneously performs demand influencing factor extraction (obtaining an influencing factor set) and cluster analysis (generating power generation data clusters). Both are input into the power generation control fitting module to construct a multi-channel energy control model. By real-time monitoring of demand-side response data, matching and parsing the optimal power generation control parameters, and performing energy feedback control and other technical means, the technical problems of poor control strategy adaptability and insufficient response accuracy in existing energy storage system energy management are solved, achieving the technical effect of improving the adaptability of the control strategy and response accuracy.

[0097] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An energy management method for an energy storage system based on demand response, characterized in that: The method comprises: Acquire an energy storage system characteristic data set and an energy storage system power generation data set, and associate and map the energy storage system characteristic data set and the energy storage system power generation data set to obtain an energy storage system characteristic-power generation data set; Extracting demand-side influencing factors from the energy storage system characteristic-power generation data set to determine a demand-side influencing factor set, and performing cluster analysis on the energy storage system characteristic-power generation data set to obtain a multi-energy storage system power generation data cluster; Performing power generation control fitting on the power generation data cluster of the multiple energy storage systems according to the set of demand-side influencing factors, and building a multi-channel energy control for the energy storage system; The target energy storage system's demand-side response data is monitored in real time, and the energy storage system's energy control multi-channel is used to match and analyze the target energy storage system's demand-side response data, determine the energy storage system's power generation control parameters, and perform energy feedback control based on the energy storage system's power generation control parameters.

2. The energy management method for an energy storage system based on demand response according to claim 1, characterized in that: The energy storage system characteristic-power generation data set is obtained, including: Initializing a noise filter according to noise characteristic information of the energy storage system characteristic dataset and the energy storage system power generation dataset; Using the noise filter to filter and denoise the energy storage system characteristic data set and the energy storage system power generation data set to obtain an available energy storage system characteristic data set and an available energy storage system power generation data set; Downsampling the available energy storage system characteristic dataset and the available energy storage system power generation dataset according to data application requirements to obtain a reduced-dimensionality energy storage system characteristic dataset and a reduced-dimensionality energy storage system power generation dataset; The reduced-dimensionality energy storage system characteristic dataset and the reduced-dimensionality energy storage system power generation dataset are aligned and associated mapped according to time series to obtain the energy storage system characteristic-power generation dataset.

3. The energy management method for an energy storage system based on demand response according to claim 1, characterized in that: Determining the set of demand-side influencing factors includes: Constructing a demand-side influencing factor system, wherein the demand-side influencing factor system includes time factors, meteorological factors, electricity price factors, and load factors; Extracting related influencing factors from the energy storage system characteristic-power generation data set according to the demand-side influencing factor system to obtain a demand-side related influencing factor set; Based on the energy storage system characteristic-power generation data set, principal component analysis and importance scoring are performed on the demand-side related influencing factor set to obtain a set of related influencing factor importance coefficients; The demand-side associated influencing factor set is screened based on the associated influencing factor importance coefficient set to determine the demand-side influencing factor set.

4. The energy management method for an energy storage system based on demand response according to claim 3, characterized in that: The screening of the demand-side associated influencing factor set based on the associated influencing factor importance coefficient set to determine the demand-side influencing factor set includes: Performing hierarchical analysis and weight assignment on each factor type in the demand-side influencing factor system to generate a weight factor of the influencing factor system; Determining the weight of the influencing factor type screening according to the influencing factor system weight factor; The demand-side associated influencing factor set is proportionally screened based on the associated influencing factor importance coefficient set according to the screening proportion of the influencing factor type to determine the demand-side influencing factor set.

5. The energy management method for energy storage system based on demand response according to claim 1, characterized in that: The obtaining of the multi-energy storage system power generation data cluster includes: Performing feature extraction and dimensionality reduction on the energy storage system characteristic-power generation data set to obtain a key feature set of energy storage and power generation; Determining energy storage power generation feature dimensions based on the energy storage power generation key feature set; Performing initial K-means clustering on the energy storage system characteristic-power generation data set according to the energy storage power generation feature dimension to obtain an initial power generation data cluster; The initial power generation data cluster is subjected to silhouette coefficient evaluation and iterative strategy optimization to obtain the multi-energy storage system power generation data cluster.

6. The energy management method for energy storage system based on demand response according to claim 1, characterized in that: The energy storage system energy control multi-channel construction includes: Performing power generation demand identification on the multi-energy storage system power generation data cluster according to the demand-side influencing factor set to obtain a multi-energy storage system power generation sample cluster; Set energy storage system energy control tasks according to the energy storage system energy management goals; Based on the energy storage system energy control task, power generation control fitting is performed on the multiple energy storage system power generation sample clusters to build a multi-channel energy storage system energy control.

7. The energy management method for an energy storage system based on demand response according to claim 6, characterized in that: The performing power generation control fitting on the multiple energy storage system power generation sample clusters based on the energy storage system energy control task to build a multi-channel energy storage system energy control includes: Based on the energy control task of the energy storage system, the power generation data of the multiple energy storage system power generation sample clusters are sequentially associated to obtain an energy storage task associated sample cluster data set; The power generation control fitting is performed on the energy storage task associated sample cluster data sets respectively, and a multi-channel energy control of the energy storage system is established.

8. The energy management method for an energy storage system based on demand response according to claim 7, characterized in that: The power generation control fitting is performed on the energy storage task associated sample cluster data sets respectively to build a multi-channel energy control for the energy storage system, including: Selecting an energy storage task network structure according to the energy storage system energy control task; Using the energy storage task network structure, power generation control fitting is performed on the energy storage task associated sample cluster data sets to generate a multi-energy storage cluster energy control branch channel set; The multi-energy storage cluster energy control branch channel set is identified according to the multi-energy storage system power generation data cluster, and the energy storage system energy control multi-channel is constructed.

9. The energy management method for energy storage system based on demand response according to claim 8, characterized in that: The step of labeling the multi-energy storage cluster energy control branch channel set according to the multi-energy storage system power generation data cluster to establish the energy storage system energy control multi-channel includes: Classify and identify the multi-energy storage cluster energy control branch channel set according to the multi-energy storage system power generation data cluster to obtain a multi-energy storage cluster associated branch channel set; The branch channel sets associated with the multiple energy storage clusters are respectively integrated in parallel to build a multi-channel energy control system, wherein the number of channels of the multi-channel energy control system corresponds one-to-one to the power generation data clusters of the multiple energy storage systems.

10. The energy management method for energy storage system based on demand response according to claim 1, characterized in that: The energy feedback control based on the energy storage system power generation control parameters includes: Performing energy management monitoring on the target energy storage system based on the energy storage system power generation control parameters to obtain energy storage control feedback parameters; A PID controller is used to adjust and correct the power generation control parameters of the energy storage system based on the energy storage control feedback parameters, and the energy storage system energy management is performed through the corrected power generation control parameters of the energy storage system.

Images