Energy storage planning method and system for power system

Abstract

The invention relates to the technical field of electric power energy storage planning, and discloses an energy storage planning method and system for an electric power system. The method comprises the steps of extracting a power system operation feature vector according to power grid data; carrying out time sequence analysis and prediction on the load and power generation data; performing probability statistical analysis on multiple scenes based on the prediction data to obtain an energy storage demand; determining capacity configuration parameters through multi-scene optimization calculation to form an energy storage scheme; and a target energy storage scheme including benefit indexes, capacity adjustment and operation optimization is obtained through comprehensive evaluation and dynamic correction. The technical problem of multi-scene collaborative configuration of the energy storage system is solved. Reasonable allocation and dynamic optimization of energy storage resources are realized through feature extraction, time sequence analysis, probability statistics and multi-scene optimization calculation.

Description

Technical Field

[0001] This application relates to the technical field of power energy storage planning, and particularly to an energy storage planning method and system for a power system. Background Art

[0002] With the large-scale access of renewable energy in the power system and the increasing complexity of the electricity load characteristics, energy storage systems play an increasingly important role in the operation of the power grid. Currently, energy storage systems are mainly applied in scenarios such as peak-valley regulation, new energy consumption, frequency regulation, and voltage support. In the prior art, energy storage planning methods mainly include capacity configuration based on historical data, scenario allocation based on experience, and operation optimization based on fixed models. These methods meet the basic requirements of the power system for energy storage to a certain extent.

[0003] However, there are obvious deficiencies in the prior art: energy storage systems are often pre-fixed in a specific application scenario and lack the ability to flexibly switch between different scenarios; the planning of energy storage capacity relies too much on empirical data and fails to fully consider the dynamic changes in the operation state of the power grid; the scenario allocation scheme lacks a scientific evaluation and optimization mechanism, resulting in low utilization rate of energy storage resources; the formulation of operation strategies lacks consideration of the multi-scenario synergy effect and it is difficult to maximize the benefits of energy storage. In addition, the existing energy storage planning methods have limited data processing capabilities and cannot effectively handle large-scale and multi-dimensional power system operation data. Summary of the Invention

[0004] This application provides an energy storage planning method and system for a power system, which solves the technical problem of multi-scenario collaborative configuration of energy storage systems. Through feature extraction, time series analysis, probability statistics, and multi-scenario optimization calculation, reasonable allocation and dynamic optimization of energy storage resources are realized.

[0005] In a first aspect, this application provides an energy storage planning method for a power system. The energy storage planning method for the power system includes: based on the real-time operation data of the power grid and the operation data of the energy storage system, through feature extraction processing, obtaining a power system operation feature vector including load change features, power generation output features, and power grid operation features; performing time series analysis processing on the load data and new energy power generation data in the power system operation feature vector to obtain a prediction data set including future load demands and power generation amounts; based on the prediction data set, performing probability statistical analysis on the operation data of the energy storage system in the peak-valley regulation scenario, new energy consumption scenario, frequency regulation scenario, and voltage support scenario to obtain energy storage demand data; according to the energy storage demand data, determining energy storage capacity configuration parameters through multi-scenario optimization calculation to obtain an energy storage configuration scheme including the total energy storage capacity, power configuration, and capacity allocation ratios in different scenarios; Based on the energy storage configuration plan, the energy storage planning plan is dynamically corrected through the comprehensive evaluation of the power system to obtain the target energy storage plan including the benefit evaluation index, the capacity adjustment parameter, and the operation optimization parameter.

[0006] In a second aspect, the present application provides an energy storage planning system for a power system. The energy storage planning system for a power system includes: An extraction module, configured to obtain a power system operation feature vector including a load change feature, a power generation output feature, and a power grid operation feature through feature extraction processing based on the real-time operation data of the power grid and the operation data of the energy storage system; An analysis module, configured to perform time series analysis processing on the load data and new energy power generation data in the power system operation feature vector to obtain a prediction data set including future load demand and power generation; A statistics module, configured to perform probability statistical analysis on the operation data of the energy storage system in the peak-valley regulation scenario, the new energy consumption scenario, the frequency regulation scenario, and the voltage support scenario based on the prediction data set to obtain energy storage demand data; A configuration module, configured to determine the energy storage capacity configuration parameter through multi-scenario optimization calculation according to the energy storage demand data to obtain an energy storage configuration plan including the total energy storage capacity, the power configuration, and the capacity allocation ratio in different scenarios; A correction module, configured to dynamically correct the energy storage planning plan through the comprehensive evaluation of the power system based on the energy storage configuration plan to obtain a target energy storage plan including the benefit evaluation index, the capacity adjustment parameter, and the operation optimization parameter.

[0007] In a third aspect of the present invention, there is provided a computer device, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned energy storage planning method for a power system.

[0008] In a fourth aspect of the present invention, there is provided a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer is made to execute the above-mentioned energy storage planning method for a power system.

[0009] In the technical solution provided by this application, through the feature extraction and processing based on the grid operation data and the energy storage system operation data, the effective integration of the load change characteristics, the power generation output characteristics, and the grid operation characteristics is achieved, providing a comprehensive data basis for energy storage planning. Through the time series analysis and processing of the load data and the new energy power generation data in the power system operation feature vector, more accurate future load demand and power generation prediction are obtained. In multi-scenario applications, the probability statistical analysis of the operation data of the energy storage system in the peak-valley regulation scenario, the new energy consumption scenario, the frequency regulation scenario, and the voltage support scenario effectively improves the accuracy of the energy storage demand assessment. Based on the multi-scenario optimization calculation of the energy storage demand data, reasonable energy storage capacity configuration parameters are determined, making the total energy storage capacity, the power configuration, and the capacity allocation ratio in different scenarios more scientific. Through the comprehensive evaluation of the power system, the energy storage planning scheme is dynamically corrected, and an objective energy storage scheme including benefit evaluation indicators, capacity adjustment parameters, and operation optimization parameters is obtained, realizing the unity of the economy and technology of energy storage planning. Especially in data processing, a temporal convolutional network is used for data analysis, deep reinforcement learning is used for multi-dimensional feature analysis, and an adaptive weight allocation algorithm is used for operation state adjustment. The application of these artificial intelligence algorithms significantly improves the scientificity and accuracy of energy storage planning. The algorithm features play an important role in aspects such as energy storage capacity optimization configuration, multi-scenario collaborative scheduling, and dynamic correction, enabling the energy storage system to better adapt to the complex operation environment of the power system and improving the utilization efficiency of energy storage resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0011] Figure 1 It is a schematic diagram of an embodiment of the energy storage planning method for a power system in an embodiment of this application; Figure 2 It is a schematic flowchart of the feature extraction and processing according to the real-time grid operation data and the operation data of the energy storage system in an embodiment of this application; Figure 3 It is a schematic diagram of an embodiment of the energy storage planning system for a power system in an embodiment of this application; Figure 4 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] The embodiments of the present application provide an energy storage planning method and system for a power system. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 , an embodiment of the energy storage planning method for a power system in the embodiments of the present application includes: Step S101: According to the real-time operation data of the power grid and the operation data of the energy storage system, through feature extraction processing, obtain a power system operation feature vector including load change features, power generation output features, and power grid operation features; Step S102: Perform time series analysis processing on the load data and new energy power generation data in the power system operation feature vector to obtain a prediction data set including future load demands and power generation amounts; Step S103: Based on the prediction data set, perform probability statistical analysis on the operation data of the energy storage system in peak-valley regulation scenarios, new energy consumption scenarios, frequency regulation scenarios, and voltage support scenarios to obtain energy storage demand data; Step S104: According to the energy storage demand data, determine the energy storage capacity configuration parameters through multi-scenario optimization calculation to obtain an energy storage configuration plan including the total energy storage capacity, power configuration, and capacity allocation ratios in different scenarios; Step S105: Based on the energy storage configuration plan, dynamically correct the energy storage planning plan through comprehensive evaluation of the power system to obtain a target energy storage plan including benefit evaluation indicators, capacity adjustment parameters, and operation optimization parameters.

[0014] It can be understood that the execution subject of the present application can be an energy storage planning system for a power system, or a terminal or a server. Specifically, it is not limited here. The embodiments of the present application are described by taking the server as the execution subject as an example.

[0015] Specifically, the operation data of the power grid and the energy storage system are processed through feature extraction. The operation data of the power grid includes information such as load curves, voltage fluctuations, and frequency deviations, while the operation data of the energy storage system includes information such as charge and discharge power and state of charge. Through feature extraction, the peak-valley difference and load factor of the load curve are calculated, the volatility of power generation and the capacity factor of power generation are calculated, and the voltage qualification rate and frequency deviation are statistically analyzed to obtain the operation feature vector of the power system. This feature vector contains information in three dimensions: load change characteristics, power generation output characteristics, and power grid operation characteristics. Time series analysis processing is performed on the load data and new energy power generation data in the operation feature vector of the power system. In the specific process, time series sample data is generated, the load data is marked with time, and load prediction data containing period characteristics, date characteristics, and holiday characteristics is obtained; the power generation data is correlated with meteorological factors to obtain power generation prediction data containing temperature characteristics, humidity characteristics, illumination characteristics, and wind speed characteristics. Through time series fusion processing, a prediction data set containing future load demand and power generation is formed.

[0016] In the probability statistics analysis stage, the operation data of the energy storage system under different scenarios is processed. In the peak-valley regulation scenario, the peak shaving and valley filling capacity demand of load fluctuations is calculated; in the new energy accommodation scenario, the power capacity required for suppressing power fluctuations is analyzed; in the frequency regulation scenario, the frequency response characteristic parameters are extracted; in the voltage support scenario, the reactive power compensation demand is statistically analyzed. These data are processed through data aggregation to form energy storage demand data. In the multi-scenario optimization calculation stage, the energy storage demand data is decomposed into investment cost, maintenance cost, and equipment replacement cost data for full-cycle cost calculation. At the same time, time series analysis of the peak-valley electricity price difference is carried out, and the revenue is accounted for in combination with the frequency modulation market compensation mechanism. Through the analysis of the equipment life curve and the calculation of the failure probability, the equipment reliability data is obtained. These data are processed through multi-dimensional constraints to form energy storage capacity calculation parameters, and the total energy storage capacity, power configuration, and capacity allocation ratio under different scenarios are determined.

[0017] The dynamic correction stage comprehensively evaluates and optimizes the energy storage configuration plan. Data such as peak-valley regulation benefits, frequency modulation compensation, and new energy consumption benefits are extracted and accumulated to obtain benefit evaluation data. The total energy storage capacity generation cost data is calculated over the entire life cycle to obtain cost evaluation data. The benefits and cost data are compared and analyzed and calculated according to the net present value to form a benefit evaluation index. According to the benefit evaluation index, the total energy storage capacity and power configuration are corrected to obtain capacity adjustment parameters. The capacity adjustment parameters are calibrated with the capacity allocation ratios of each scenario to obtain operation optimization parameters. Finally, boundary constraint processing is performed on the benefit evaluation index and the capacity adjustment parameters, and they are integrated with the operation optimization parameters to form the target energy storage plan. For example: The grid operation data of a certain energy storage power station shows that the peak-valley difference of the daily load curve is 8 MW, the load factor is 0.75, the volatility of new energy power generation is 15%, and the voltage qualification rate is 95%. After obtaining the operation feature vector through feature extraction, time series analysis shows that the weekday load demand is 20% higher than that on weekends, and for every 100 W / increase in light intensity, the photovoltaic power generation increases by 0.5 MW. Probability statistical analysis shows that the peak-valley regulation demand is 5 MW, the power fluctuation suppression demand is 2 MW, the frequency modulation response demand is 1 MW, and the reactive power compensation demand is 1 Mvar. Cost-benefit analysis shows that the investment cost is 5 million yuan / MW, the annual operation and maintenance cost is 2% of the investment amount, and the equipment life is 10 years. After dynamic correction, the total energy storage capacity is determined to be 6 MW, of which 4 MW is allocated for peak-valley regulation, 1 MW is allocated for power fluctuation suppression, and 0.5 MW is allocated for each of frequency modulation and voltage support.

[0018] In the embodiments of the present application, through the feature extraction and processing based on the grid operation data and the energy storage system operation data, the effective fusion of the load change characteristics, the power generation output characteristics, and the grid operation characteristics is realized, providing a comprehensive data basis for energy storage planning. Through the time series analysis and processing of the load data and the new energy power generation data in the power system operation feature vector, more accurate future load demand and power generation prediction are obtained. In multi-scenario applications, the probability statistical analysis of the operation data of the energy storage system in the peak-valley regulation scenario, the new energy consumption scenario, the frequency regulation scenario, and the voltage support scenario effectively improves the accuracy of the energy storage demand assessment. Based on the multi-scenario optimization calculation of the energy storage demand data, reasonable energy storage capacity configuration parameters are determined, making the total energy storage capacity, power configuration, and the capacity allocation ratio in different scenarios more scientific. Through the comprehensive evaluation of the power system, the energy storage planning scheme is dynamically corrected, and the target energy storage scheme including the benefit evaluation index, the capacity adjustment parameter, and the operation optimization parameter is obtained, realizing the unity of the economy and technology of the energy storage planning. Especially in data processing, the time series convolutional network is used for data analysis, the deep reinforcement learning is used for multi-dimensional feature analysis, and the adaptive weight allocation algorithm is used for operation state adjustment. The application of these artificial intelligence algorithms significantly improves the scientificity and accuracy of the energy storage planning. The algorithm features play an important role in aspects such as the optimal configuration of the energy storage capacity, the multi-scenario collaborative scheduling, and the dynamic correction, enabling the energy storage system to better adapt to the complex operation environment of the power system and improving the utilization efficiency of the energy storage resources.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Collect the load curve data, the renewable energy power generation output data, and the grid frequency fluctuation data in the grid real-time operation data; (2) Calculate the peak-valley difference of the load curve data and count the load factor to obtain the load change characteristics; (3) Calculate the power generation power volatility and the power generation capacity factor of the renewable energy power generation output data to obtain the power generation output characteristics; (4) Calculate the voltage qualification rate and count the frequency deviation of the grid frequency fluctuation data to obtain the grid operation characteristics; (5) Perform data fusion processing on the load change characteristics, the power generation output characteristics, and the grid operation characteristics to obtain the power system operation feature vector.

[0020] Specifically, as Figure 2As shown, it is a schematic flowchart of feature extraction processing based on real-time grid operation data and the operation data of the energy storage system in the embodiment of the present application; the load curve data, renewable energy power generation output data, and grid frequency fluctuation data are collected. Among them, the load curve data represents the curve of the electric load changing with time, recording the power consumption at each time point; the renewable energy power generation output data includes the real-time power generation of new energy sources such as wind power and photovoltaic power; the grid frequency fluctuation data records the real-time deviation value of the grid frequency. These data are collected at 15-minute intervals through smart meters, energy management systems, and power monitoring devices. The peak-valley difference of the load curve data is calculated, that is, the difference between the daily maximum load and the minimum load is calculated, reflecting the fluctuation degree of the power consumption load. The load factor statistics is calculated by calculating the ratio of the average load to the maximum load, characterizing the average utilization level of the load. The peak-valley difference calculation sorts the load data within 24 hours by time, finds the maximum and minimum values and then subtracts them; the load factor statistics calculates the average of the load values within 24 hours and then divides by the maximum load value. These two indicators together constitute the load change characteristics, describing the dynamic change law of the power consumption load.

[0021] The processing of the renewable energy power generation output data includes the calculation of the power generation volatility rate and the calculation of the power generation capacity factor. The power generation volatility rate is the ratio of the change in power generation between adjacent time points to the rated power, reflecting the stability of the power generation output; the power generation capacity factor is the ratio of the actual power generation to the theoretical maximum power generation, characterizing the utilization degree of the power generation equipment. The volatility rate calculation requires statistical analysis of the power difference between adjacent time points, and the capacity factor calculation requires comparing the actual power generation with the product of the installed capacity and the operation time. These two indicators constitute the power generation output characteristics, reflecting the instability characteristics of new energy power generation. The processing of the grid frequency fluctuation data includes the calculation of the voltage qualification rate and the statistics of the frequency deviation. The voltage qualification rate is the proportion of the time when the voltage is within the standard range, calculated by counting the number of time points when the voltage value is within ±5% of the rated voltage; the frequency deviation statistics is to statistically analyze the degree of deviation of the grid frequency from 50Hz, calculating the mean and standard deviation of the frequency deviation. These indicators constitute the grid operation characteristics, reflecting the stability status of the grid operation.

[0022] The data fusion processing integrates the load change characteristics, power generation output characteristics, and grid operation characteristics into the operation characteristic vector of the power system. In the fusion process, normalization processing is performed on each characteristic to make the indicators with different dimensions comparable, and then each characteristic is synthesized into a multi-dimensional vector through weighted summation. The determination of the weight is based on the importance of each characteristic to the energy storage planning. Generally, the weight of the load change characteristics is relatively large because it is directly related to the configuration of the energy storage capacity.

[0023] For example, the load curve of a certain power system on a working day shows that the maximum load is 100 MW, the minimum load is 60 MW, and the average load is 80 MW. By calculation, the peak-valley difference is 40 MW, and the load factor is 0.8. At the same time, the maximum power fluctuation of the wind farm in this system within a day is 30% of the installed capacity, and the generation capacity factor is 0.4, indicating that the power output fluctuation is large and the equipment utilization rate needs to be improved. The power grid operation data shows that the voltage qualification rate is 98% and the standard deviation of frequency deviation is 0.1 Hz. These data are normalized and weighted to form a feature vector.

[0024] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Generate load time series sample data by using the load data in the power system operation feature vector; (2) Extract new energy generation data from the power system operation feature vector to generate power generation time series sample data; (3) Perform time marking processing on the load time series sample data to obtain load prediction data including period characteristics, date characteristics, and holiday characteristics; (4) Perform meteorological factor correlation processing on the power generation time series sample data to obtain power generation prediction data including temperature characteristics, humidity characteristics, light characteristics, and wind speed characteristics; (5) Perform time series fusion processing on the load prediction data and the power generation prediction data to obtain a prediction data set including future load demand and power generation.

[0025] Specifically, the data in the power system operation feature vector is converted into a prediction data set of future load demand and power generation. Load time series sample data is generated by using the load data in the power system operation feature vector. This process uses data extraction and recombination techniques to separate the load curve data from the feature vector and organize it into a continuous time series in chronological order. The load time series sample data contains information such as historical load change trends, load peak-valley distributions, and load fluctuation rules. For example, the power grid dispatching center has collected 15-minute load sampling data for the past three years. Through unified timestamp processing and outlier cleaning, a complete load time series sample data is formed, which records the load change characteristics at different times on working days and rest days.

[0026] New energy power generation data mainly includes the power output data of photovoltaic power generation and wind power generation, which need to be screened from the feature vectors and reconstructed in chronological order. The power generation time series sample data reflects the volatility and intermittency characteristics of new energy power generation and records the variation law of power generation with time. In the actual application of a certain power dispatching center, technicians extracted the historical power generation data of 20 wind farms and 15 photovoltaic power stations in the region. After data standardization and missing value filling, high-quality power generation time series sample data were generated, which demonstrated the seasonal variation and intra-day fluctuation characteristics of new energy power generation. The time marking process enhanced the interpretability and prediction accuracy of the data by adding feature tags in the time dimension. The time period features divide a day into multiple time periods, such as early morning, morning, noon, afternoon, evening, and late night, reflecting the load characteristics of different periods. The date features include marks such as weekdays, weekends, the beginning of the month, the middle of the month, and the end of the month, reflecting the periodic variation of the load. The holiday features mark legal holidays, traditional festivals, and special event days, capturing the abnormal load changes on these special dates. In a certain energy storage planning project, technicians divided the load time series data into 96 15-minute time periods according to 24 hours and added weekday / weekend marks and holiday marks to each data point. It was found that the load peaks from Monday to Friday are concentrated in 10:00-12:00 and 15:00-17:00, while the load peak on weekends is concentrated in 19:00-21:00. These time period features significantly affect the demand allocation of the energy storage system. The meteorological factor correlation process improves the accuracy of power generation prediction by establishing a data mapping relationship between new energy power generation and meteorological conditions. The temperature feature reflects the impact of ambient temperature on the power generation efficiency of photovoltaic modules. The humidity feature captures the impact of air humidity on the performance of power generation equipment. The light feature includes key parameters such as light intensity, light duration, and light angle, which directly determine the output power of photovoltaic power generation. The wind speed feature records information such as wind speed magnitude, wind direction change, and wind speed duration, which is the core basis for wind power generation prediction. In the new energy planning of a certain region, technicians collected historical meteorological data and aligned them with the power generation data in time. It was found that there is a significant positive correlation between light intensity and photovoltaic power generation, and the power generation power of wind farms is proportional to the cube of the wind speed. For every 10°C increase in temperature, the photovoltaic power generation efficiency decreases by about 4%. These correlation features greatly improve the prediction accuracy of new energy power generation.

[0027] Performing time-series fusion processing on the load prediction data and power generation prediction data to obtain a prediction dataset containing future load demand and power generation is the last important step. Time-series fusion processing analyzes the load-side demand and power generation-side supply in the same time frame through time-axis alignment, feature integration, and prediction model training. The fused prediction dataset contains both load prediction values and new energy power generation prediction values, and can intuitively reflect the supply-demand balance state of the system. In a certain energy storage planning project, technicians used the sliding window method to align the load data and power generation data in time series, and applied a time series prediction algorithm to generate hourly load demand and power generation prediction values for the next 7 days. The prediction results show that during the day in winter, from 9:00 to 15:00, the peak photovoltaic power generation overlaps with the low valley of residential electricity consumption, and in the evening, from 18:00 to 22:00, the peak residential electricity consumption coincides with the increased fluctuation of wind power generation. These time-series characteristics provide a direct basis for the capacity configuration and operation strategy formulation of the energy storage system. The prediction dataset formed through time-series fusion processing directly supports the multi-scenario demand analysis and optimal configuration calculation of subsequent energy storage planning.

[0028] In a specific embodiment, the process of performing step S103 may specifically include the following steps: (1) Using the prediction dataset to perform peak shaving and valley filling analysis on the load fluctuation data in the peak-valley regulation scenario to obtain peak-valley regulation capacity data; (2) According to the prediction dataset, performing smoothing processing on the power fluctuation data in the new energy accommodation scenario to obtain power capacity data; (3) Based on the prediction dataset, performing feature extraction on the frequency response data in the frequency regulation scenario to obtain frequency regulation capacity data; (4) Performing statistical processing on the reactive power compensation data in the voltage support scenario in the prediction dataset to obtain voltage support capacity data; (5) Aggregating the peak-valley regulation capacity data, power capacity data, frequency regulation capacity data, and voltage support capacity data to obtain energy storage demand data.

[0029] Specifically, the process of using the prediction dataset to perform peak shaving and valley filling analysis on the load fluctuation data in the peak-valley regulation scenario and obtaining the peak-valley regulation capacity data involves the time series processing of the load curve. Peak shaving and valley filling analysis refers to a technical method of cutting off the peak part exceeding the set threshold in the power load curve and filling it into the valley period to smooth the load curve. Specifically, first extract the daily load curve from the prediction dataset, determine the peak load and valley load through fluctuation analysis, calculate the peak-valley difference, and then determine the amount of electricity to be transferred according to the peak shaving and valley filling balance principle, and calculate the required energy storage capacity. In practical applications, the load prediction data of a certain power dispatching area shows that the peak electricity consumption period is from 18:00 to 22:00 on weekdays, and the valley electricity consumption period is from 1:00 to 5:00 in the early morning. Through peak shaving and valley filling analysis, the amount of peak load electricity to be transferred is determined, and the energy storage capacity data suitable for peak-valley regulation in this area is calculated.

[0030] The suppression of power fluctuation is a technical method that compensates for the random fluctuations of new energy power generation through an energy storage system to make the output power smooth and stable. Among them, extract the time series of new energy power generation power in the prediction dataset, calculate the power change rate, determine the fluctuating part exceeding the allowable range, and then calculate the energy that the energy storage system needs to absorb or release according to the power balance principle, and determine the energy storage capacity required for power fluctuation suppression. In practical applications, the maximum volatility of the power generation power of a certain wind farm reaches 30% within 10 minutes. Through the suppression of power fluctuation, the energy storage capacity required to suppress these fluctuations is calculated to ensure the stable output of new energy power generation. Extracting the frequency response data in the frequency regulation scenario based on the prediction dataset and obtaining the frequency regulation capacity data is the third key link in energy storage planning. Frequency response feature extraction refers to a technical method of analyzing the relationship between grid frequency fluctuations and load changes and determining the capacity required for the energy storage system to participate in primary frequency regulation and secondary frequency regulation. When implementing, extract the historical frequency fluctuation data from the prediction dataset, calculate the statistical distribution of frequency deviation, analyze the frequency change rate, and determine the power and energy capacity required for the energy storage system to participate in frequency regulation according to the frequency regulation response speed requirement and the duration requirement. In practical applications, the frequency fluctuation records of a certain regional power grid show that after a large power generation unit trips off the grid, the maximum frequency deviation reaches 0.2 Hz and the duration is 15 minutes. Through frequency response feature extraction analysis, the energy storage capacity required for frequency regulation in this area is determined.

[0031] Reactive power compensation statistical processing refers to a technical method of analyzing the law of grid voltage fluctuations and determining the capacity required for the energy storage system to provide reactive power support. Specifically, extract the voltage fluctuation data and reactive power demand data from the prediction dataset, calculate the voltage deviation distribution, analyze the node voltage sensitivity, determine the reactive power compensation demand, and calculate the capacity required for the energy storage system to provide voltage support. The calculation of reactive power compensation capacity can be expressed by the following formula:

[0032] Among them, represents the energy storage capacity required for voltage support, represents the weight coefficient of the i-th node, reflecting the importance of this node in the power grid, represents the actual reactive power of the i-th node in the j-th time period, represents the reference reactive power of the i-th node in the j-th time period, representing the reactive power demand in the ideal state, N represents the number of statistical nodes, and M represents the number of statistical time periods. represents the voltage sensitivity coefficient of the i-th node, characterizing the influence degree of reactive power change on the node voltage. In practical applications, the voltage of the end node of a distribution network has been low for a long time. Through the statistical analysis of reactive power compensation, the energy storage capacity required to improve the voltage of this node has been determined, effectively improving the voltage quality. During the statistical process of reactive power compensation, key nodes in the distribution network are selected, voltage monitoring data and reactive power demand data of these nodes are collected, and then the voltage qualification rate and voltage deviation statistical value of each node are calculated. The time period and node with the largest voltage deviation are determined, and combined with the voltage sensitivity analysis of the node, the reactive power compensation amount required to improve the voltage quality is calculated and converted into the capacity demand of the energy storage system.

[0033] Data aggregation processing refers to a technical method that comprehensively considers the capacity requirements of different application scenarios and determines the overall demand of the energy storage system through weighted synthesis and time-series superposition analysis. Specifically, analyze the time distribution characteristics of the capacity requirements of each scenario, identify the overlapping time periods of the capacity requirements between different scenarios, avoid duplicate capacity calculation through the adjustment of the sharing coefficient, and then assign weights according to the priority of each scenario, and calculate the total energy storage demand considered comprehensively. In practical applications, a power company found through data aggregation processing analysis that there is an obvious demand overlap between peak-valley regulation and frequency regulation in the afternoon, and the demand of new energy consumption and voltage support is superimposed in the evening. By reasonably setting the sharing adjustment coefficient, the total capacity configuration of the energy storage system is optimized, avoiding the capacity redundancy caused by simple accumulation. The specific steps of data aggregation processing include: time-series alignment processing, reorganizing the capacity demand data of different scenarios according to the same time scale; overlap analysis, identifying the common demand for energy storage capacity in the same time period by multiple scenarios; weight assignment, setting weight coefficients according to the importance and response priority of different scenarios; capacity synthesis, calculating the energy storage demand data by considering time-series characteristics and sharing effects. This data aggregation processing method avoids the over-configuration of energy storage capacity caused by simply adding the requirements of each scenario, and improves the economy and utilization efficiency of the energy storage system.

[0034] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Decompose the energy storage demand data into energy storage equipment investment cost data, equipment maintenance cost data, and equipment replacement cost data to obtain a cost data set; (2) Conduct a time series analysis of the peak-valley electricity price difference based on the energy storage demand data, and combine it with the frequency regulation market compensation mechanism to conduct revenue accounting to obtain a revenue data set; (3) Analyze the attenuation characteristics of the energy storage equipment life curve in the energy storage demand data, and calculate the equipment failure probability to obtain an equipment reliability data set; (4) Perform multi-dimensional constraint processing on the cost data set, revenue data set, and equipment reliability data set to obtain energy storage capacity calculation parameters; (5) Use the energy storage capacity calculation parameters to conduct total capacity optimization and state of charge analysis to obtain the total energy storage capacity; (6) Allocate and process the total energy storage capacity according to the charge and discharge power limit to obtain the power configuration; (7) Divide the total energy storage capacity and power configuration proportionally according to the demand priorities of the peak-valley regulation scenario, new energy consumption scenario, frequency regulation scenario, and voltage support scenario to obtain an energy storage configuration plan.

[0035] Specifically, decomposing the energy storage demand data into energy storage equipment investment cost data, equipment maintenance cost data, and equipment replacement cost data to obtain a cost data set involves cost decomposition and data classification processing. Cost decomposition refers to a data processing method that refines the life cycle cost of an energy storage system into three main parts: initial investment, operation and maintenance, and renewal and replacement. When specifically implemented, extract the energy storage equipment type, capacity specification, and technical parameters from the energy storage demand data, and calculate the initial investment cost in combination with market price information; estimate the annual maintenance cost based on the equipment operation parameters and maintenance procedures; predict the future replacement cost according to the equipment life characteristics and replacement cycle. The cost data set contains the full-cycle cost information of energy storage systems with different capacities and different types. Conducting a time series analysis of the peak-valley electricity price difference based on the energy storage demand data and combining it with the frequency regulation market compensation mechanism to conduct revenue accounting to obtain a revenue data set is the second key link in multi-scenario optimization calculation. Time series analysis refers to a data processing method that studies the distribution law of electricity price differences in different time periods and predicts future electricity price trends. Among them, extract the load curve and electricity price information from the energy storage demand data, calculate the electricity price difference in different time periods, analyze the seasonal and periodic change laws of the electricity price difference, and predict the future electricity price difference trend; at the same time, collect the auxiliary service compensation standards of the frequency regulation market, calculate the potential revenue of the energy storage system participating in frequency regulation services, and form a revenue data set that includes two parts: peak-valley arbitrage revenue and auxiliary service compensation.

[0036] Analyzing the attenuation characteristics of the energy storage device life curve in the energy storage demand data and calculating the device failure probability to obtain the device reliability data set is the third key link in the multi-scenario optimization calculation. Attenuation characteristic analysis refers to the data processing method of studying the law of the energy storage device capacity decreasing with the number of charge-discharge cycles and time, and predicting the device performance changes. This process can be expressed by the following formula:

[0037] Where, represents the remaining capacity ratio of the energy storage device at time t, represents the initial capacity ratio value (usually 1), represents the attenuation weight coefficient of the i-th charge-discharge cycle, represents the depth of discharge of the i-th charge-discharge cycle, represents the reference depth of discharge, represents the depth of discharge influence index, represents the ambient temperature of the i-th charge-discharge cycle, represents the reference temperature, represents the temperature influence index, represents the charge-discharge current of the i-th charge-discharge cycle, represents the reference current, represents the current influence index, represents the natural aging coefficient, represents the natural time. When implementing, extract the technical specifications and operating parameters of the energy storage device from the energy storage demand data, establish a capacity attenuation model in combination with historical operating data, analyze the attenuation rate under different operating conditions, and form a data set describing the device reliability and life expectancy. Multi-dimensional constraint processing refers to the data processing method of establishing a multi-objective optimization model by comprehensively considering economic, technical feasibility, and reliability requirements. When specifically implementing, convert the cost data into investment payback period and net present value indicators, convert the revenue data into annualized rate of return indicators, convert the reliability data into device life expectancy and failure rate indicators, and then establish a multi-objective optimization model. Through constraint condition screening and weight allocation, obtain the energy storage capacity calculation parameters that meet the requirements of all aspects.

[0038] Using the energy storage capacity calculation parameters for total capacity optimization and state of charge analysis, obtaining the total energy storage capacity is the fifth key link in multi-scenario optimization. Total capacity optimization refers to the data processing method for determining the overall scale of the energy storage system based on the previously obtained calculation parameters. When implementing, substitute the capacity calculation parameters into the optimization model, consider the return on investment, technical constraints, and application requirements, calculate the comprehensive performance indicators under different capacity configurations, and determine the optimal total energy storage capacity through comparative analysis. At the same time, conduct state of charge analysis to evaluate the charge and discharge state distribution and utilization rate level of the energy storage system under different capacities, and further verify the rationality of the total capacity configuration. Power allocation processing refers to the data processing method for determining the power-to-energy ratio relationship of the energy storage system. This process can be expressed by the following formula:

[0039] where, represents the power configuration under the j-th application scenario, represents the total energy storage capacity, represents the power distribution coefficient of the j-th application scenario, represents the power conversion efficiency, represents the power availability coefficient, represents the weight of the k-th power demand peak under the j-th application scenario, represents the time distribution function of the k-th power demand under the j-th application scenario, represents the duration of the k-th power demand under the j-th application scenario, and m represents the number of power demand peaks. Starting from the total energy storage capacity, combined with the power demand characteristics of different application scenarios, determine the upper and lower limits of the charge and discharge power, and reasonably allocate power resources according to the time distribution and response speed requirements of the power demand to obtain a power configuration plan that fully meets the requirements of each scenario. Proportionally dividing the total energy storage capacity and power configuration according to the demand priorities of the peak-valley regulation scenario, new energy consumption scenario, frequency regulation scenario, and voltage support scenario to obtain the energy storage configuration plan is the last key link in multi-scenario optimization. Proportionally dividing refers to the data processing method for reasonably allocating energy storage resources to each scenario according to the importance and demand characteristics of different application scenarios. When specifically implementing, determine the priority ranking of each scenario. Usually, the frequency regulation has the highest priority, followed by peak-valley regulation and new energy consumption, and the voltage support has a lower priority; then, according to the capacity demand characteristics and time distribution characteristics of each scenario, determine the capacity allocation ratio, and at the same time consider the possibility of resource sharing between scenarios to avoid duplicate configurations; form a configuration plan that clearly indicates the energy storage capacity allocated to each scenario.

[0040] In a energy storage planning project for a regional power grid, technicians collected power grid operation data and historical data of energy storage systems. Through feature extraction, the operation feature vectors of the power system were obtained, and then time series analysis was carried out to obtain a prediction data set. Based on the prediction data set, the energy storage requirements for different scenarios were analyzed. In the multi-scenario optimization calculation stage, the energy storage requirement data was decomposed into cost data, including the initial investment cost of the lithium battery energy storage system, the 2% annual maintenance cost, and the replacement cost after 8 years. At the same time, the economic benefits of the local peak-valley electricity price difference of 0.4 yuan / kWh and the compensation mechanism of the ancillary service market were analyzed. An energy storage device capacity attenuation model was established, and it was found that every 1000 complete charge-discharge cycles resulted in a capacity loss of about 20%. Through multi-dimensional constraint processing, economic constraints with a payback period not exceeding 6 years and an annualized rate of return not less than 8% and a reliability constraint with a device failure rate not exceeding 1% were determined. Through optimization calculation, it was concluded that the total energy storage capacity suitable for configuration in this region was 10 MWh, the power configuration was determined to be 3 MW according to the energy-power ratio of 3:1, and the energy storage resource allocation was completed according to the ratio of 30% for peak-valley regulation, 40% for new energy consumption, 20% for frequency regulation, and 10% for voltage support, forming a complete energy storage configuration plan.

[0041] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Extract the peak-valley regulation revenue data, frequency modulation compensation data, and new energy consumption revenue data from the energy storage configuration plan, and perform numerical accumulation processing to obtain energy storage revenue evaluation data; (2) Generate initial cost data, operation and maintenance cost data, and equipment replacement cost data for the total energy storage capacity in the energy storage configuration plan, and perform full-cycle calculation to obtain energy storage cost evaluation data; (3) Compare and analyze the energy storage revenue evaluation data with the energy storage cost evaluation data, and process them according to the net present value calculation method to obtain the benefit evaluation index; (4) According to the benefit evaluation index, correct and compare the total energy storage capacity and power configuration in the energy storage configuration plan to obtain the capacity adjustment parameter; (5) Perform data calibration processing on the capacity adjustment parameter and the capacity allocation ratio under different scenarios in the energy storage configuration plan to obtain the operation optimization parameter; (6) Perform boundary constraint processing on the numerical ranges of the benefit evaluation index and the capacity adjustment parameter, and perform data fusion with the operation optimization parameter to obtain the target energy storage plan.

[0042] Specifically, the process of extracting peak-valley regulation revenue data, frequency modulation compensation data, and new energy consumption revenue data from the energy storage configuration plan and performing numerical accumulation to obtain energy storage revenue evaluation data involves the extraction and integration of multi-source revenue data. Revenue data extraction refers to the data processing method of separating the economic revenue information of different application scenarios from the energy storage configuration plan. When specifically implemented, the capacity configuration and operation parameters of the peak-valley regulation scenario are extracted from the configuration plan, and combined with the peak-valley electricity price difference and the charge-discharge efficiency of the energy storage system to calculate the peak-valley arbitrage revenue; at the same time, the power configuration and response characteristic parameters of the frequency regulation scenario are extracted, and combined with the market compensation standard of frequency modulation ancillary services to calculate the frequency modulation compensation revenue; in addition, from the configuration data of the new energy consumption scenario, combined with the renewable energy consumption subsidy and the economic benefits brought by the reduction of curtailment of wind and solar power, the new energy consumption revenue is calculated. Then, these three types of revenue data are aligned according to the time dimension and numerically accumulated to form the total revenue prediction data for each year within the whole life cycle of the energy storage system, that is, the energy storage revenue evaluation data. Generating the initial cost data, operation and maintenance cost data, and equipment replacement cost data for the total energy storage capacity in the energy storage configuration plan and performing a full-cycle calculation to obtain the energy storage cost evaluation data is the second key link in the comprehensive evaluation. Full-cycle cost calculation refers to the data processing method of considering all cost expenditures during the entire life cycle of the energy storage system from construction to retirement. Among them, according to the total capacity and power configuration in the energy storage configuration plan, combined with the market price and construction and installation costs of energy storage equipment, the initial cost data is generated; then, based on the equipment operation and maintenance regulations and the labor cost standard, the operation and maintenance cost data for each future year is predicted; at the same time, considering the service life and replacement cycle of the energy storage equipment, the possible equipment replacement cost data in the future is estimated. When performing the full-cycle calculation, the time value of money also needs to be considered, and the costs incurred in each future year are discounted to finally obtain the energy storage cost evaluation data reflecting the economic burden of the energy storage system throughout its life cycle.

[0043] Comparing and analyzing the energy storage revenue evaluation data with the energy storage cost evaluation data and processing them according to the net present value calculation method to obtain the benefit evaluation index is the third key link in the comprehensive evaluation. Net present value calculation refers to the data processing method of discounting the revenues and costs in each future year to the current point in time at a certain discount rate to calculate the net benefit of the project. When implemented, the energy storage revenue evaluation data and the energy storage cost evaluation data are aligned along the time axis and corresponding to each year of the life cycle of the energy storage system; then an appropriate discount rate is selected, usually determined based on the cost of capital or the industry average investment return rate; then the net cash flow of the revenue and cost in each year is calculated, that is, the annual revenue minus the annual cost; finally, the net cash flows in each year are discounted to the starting point of the project at the discount rate and accumulated to obtain the net present value of the project. At the same time, based on the same data, supplementary indicators such as the internal rate of return and the payback period can also be calculated, jointly constituting a benefit evaluation index system that comprehensively reflects the economy of the energy storage project.

[0044] The correction comparison refers to a data processing method for determining the optimal capacity range by comparing the economic benefits under different capacity configurations. Among them, taking the original energy storage configuration plan as the benchmark, multiple adjustment plans are designed, including multiple gradients in two directions of increasing capacity and decreasing capacity; then the benefit evaluation indicators are recalculated for each adjustment plan; then the index values of each plan are compared with the preset economic target (such as the minimum acceptable rate of return); finally, the capacity configuration that meets the economic target and has the largest net present value is determined, and the difference from the original plan forms the capacity adjustment parameters, including the adjustment value of the total energy storage capacity and the adjustment value of the charge and discharge power configuration.

[0045] Data calibration processing refers to a data processing method for correspondingly adjusting the capacity allocation of each application scenario according to the adjustment of the total capacity. When implementing, the capacity adjustment parameters are decomposed according to the sensitivity and priority of the application scenarios to determine the adjustment weights of each scenario; then, according to the adjustment weights, the change amounts of the capacity allocation of each scenario are calculated; then, the change amounts are corrected with the original allocation ratio to obtain the new capacity allocation ratio; finally, based on the new allocation ratio, the operating parameters of each scenario are adjusted, including the charge and discharge time periods, power limits, and response priorities, to form a complete set of operating optimization parameters.

[0046] Performing boundary constraint processing on the numerical ranges of the benefit evaluation indicators and the capacity adjustment parameters and integrating the data with the operating optimization parameters to obtain the target energy storage plan is the last key link in the comprehensive evaluation. Boundary constraint processing refers to a data processing method for ensuring that the adjusted energy storage plan still meets the requirements of technical feasibility and economic rationality. Among them, the acceptable range of the benefit evaluation indicators is determined, such as the net present value must be positive, the investment payback period does not exceed the set number of years, etc.; at the same time, the technical boundaries of the capacity adjustment parameters are determined, such as the power-to-capacity ratio does not exceed the equipment limit, and the single charge and discharge duration meets the application requirements, etc.; then the parameters that exceed the boundary are constrained and adjusted to bring them back to the reasonable range; finally, the parameters after the boundary constraint processing are integrated with the operating optimization parameters to form a target energy storage plan that meets both economic requirements and technical standards, including the finally determined total energy storage capacity, power configuration, capacity allocation ratio, and operating parameter set.

[0047] In an energy storage planning project of a certain power grid enterprise, based on the energy storage configuration plan formed in the early stage, the technical personnel conducted a comprehensive evaluation and dynamic correction of the plan. The electricity price difference benefits that can be brought by peak-valley regulation every year, the auxiliary service compensation income from participating in the frequency regulation market, and the consumption benefits of reducing new energy curtailment were extracted from the configuration plan, and the annual income stream within the 20-year life cycle was obtained through time series accumulation. Secondly, the initial investment cost of the energy storage system, the annual fixed equipment maintenance cost, and the battery module replacement cost expected to occur in the 10th year were calculated to form complete cost evaluation data. Then, a net present value calculation was performed on the income and cost using a discount rate of 8%, and the benefit evaluation indicators with a positive net present value of the project and an internal rate of return of 9.2% were obtained. Then, through correction and comparison analysis, it was found that by reducing the total energy storage capacity from 15 MWh in the original plan to 12 MWh and the power configuration from 5 MW to 4 MW, the return on investment of the project can be improved. Subsequently, according to the principle of giving priority to frequency regulation, followed by peak-valley regulation, new energy consumption, and voltage support, the capacity allocation ratios of each scenario were adjusted accordingly to obtain more optimal operating parameters. Finally, after boundary constraint verification, it was confirmed that the adjusted plan meets the technical requirements, and the adjusted parameters were integrated to form the target energy storage plan, which not only ensures economic feasibility but also meets the technical requirements of the power grid for the energy storage system.

[0048] In a specific embodiment, the process of performing the step of data correction processing on the capacity adjustment parameter and the capacity allocation ratios in different scenarios in the energy storage configuration plan may specifically include the following steps: (1) Decompose the capacity adjustment parameter by scenario to obtain the capacity parameter for the peak-valley regulation scenario, the capacity parameter for the new energy consumption scenario, the capacity parameter for the frequency regulation scenario, and the capacity parameter for the voltage support scenario; (2) Perform numerical correction on the capacity parameter for the peak-valley regulation scenario and the capacity allocation ratio in the peak-valley regulation scenario in the energy storage configuration plan to obtain the peak-valley scenario correction parameter; (3) Calculate the correction value from the capacity parameter for the new energy consumption scenario and the capacity allocation ratio in the new energy consumption scenario in the energy storage configuration plan to obtain the consumption scenario correction parameter; (4) Use the capacity parameter for the frequency regulation scenario to perform correction calculation on the capacity allocation ratio in the frequency regulation scenario in the energy storage configuration plan to obtain the frequency regulation scenario correction parameter; (5) Perform data matching on the capacity parameter for the voltage support scenario and the capacity allocation ratio in the voltage support scenario in the energy storage configuration plan to obtain the voltage scenario correction parameter; (6) Aggregate the peak-valley scenario correction parameter, the consumption scenario correction parameter, the frequency regulation scenario correction parameter, and the voltage scenario correction parameter to obtain the operation optimization parameter.

[0049] Specifically, the capacity adjustment parameters are decomposed by scenario to obtain the capacity parameters for peak-valley regulation scenarios, new energy consumption scenarios, frequency regulation scenarios, and voltage support scenarios. This process involves the decomposition and allocation of adjustment parameters. Scenario decomposition refers to a data processing method that distributes the total capacity adjustment according to the characteristics and importance of each application scenario. When specifically implemented, first extract the total capacity adjustment value and power configuration adjustment value in the capacity adjustment parameters, then calculate the weights according to the capacity sensitivity coefficient, response priority, and adjustment elasticity of each scenario, and then decompose the total adjustment amount into each application scenario according to the weight ratio to form the exclusive capacity adjustment parameters for each scenario. The scenario decomposition process takes into account the different requirements of different application scenarios for energy storage performance. The peak-valley regulation scenario focuses on the energy capacity characteristics, the new energy consumption scenario pays attention to the ability to suppress power fluctuations, the frequency regulation scenario emphasizes the response speed, and the voltage support scenario emphasizes the reactive power compensation ability. The utilization efficiency of energy storage resources is improved through differential decomposition. Numerically correcting the capacity parameters of the peak-valley regulation scenario with the capacity allocation ratio in the peak-valley regulation scenario of the energy storage configuration scheme to obtain the peak-valley scenario correction parameter is the second key link in data correction processing. Numerical correction refers to a data processing method that precisely adjusts the original configuration according to the adjustment requirements. Among them, extract the capacity allocation ratio and operation parameters of the peak-valley regulation scenario in the energy storage configuration scheme, compare them with the capacity parameters of the peak-valley regulation scenario, calculate the absolute adjustment amount and relative adjustment ratio, and then determine the best adjustment method according to the time characteristics of peak-valley regulation and the characteristics of the load curve. Finally, generate the peak-valley scenario correction parameter, including the corrected capacity allocation ratio, adjustment period, and power configuration. The peak-valley scenario correction parameter directly affects the regulation ability of the energy storage system during periods with significant grid peak-valley differences. Reasonable correction can improve the economic benefits of peak-valley regulation.

[0050] Calculating the correction value from the capacity parameters of the new energy consumption scenario and the capacity allocation ratio in the new energy consumption scenario of the energy storage configuration scheme to obtain the consumption scenario correction parameter is the third key link in data correction processing. Correction value calculation refers to a data processing method that determines the optimal adjustment amount through a specific algorithm. This process can be expressed by the following formula:

[0051] Among them, represents the correction parameter for the new energy consumption scenario, represents the weight coefficient of the g-th type of new energy, represents the consumption adjustment elasticity coefficient, represents the capacity parameters of the new energy consumption scenario, represents the capacity allocation value of the new energy consumption scenario in the original configuration scheme, represents the new energy time series distribution characteristic coefficient, represents the power fluctuation characteristic value of the g-th type of new energy, represents the penetration rate of the g-th type of new energy, represents the amount of curtailed power of the g-th type of new energy, represents the total power generation of the g-th type of new energy, represents the number of new energy types. In practical applications, this formula comprehensively considers factors such as the power fluctuation characteristics, penetration rate, and curtailed power situation of new energy, and reflects the non-linear relationship between the curtailed power and the consumption capacity through a logarithmic function, enabling the calculated correction parameters to more accurately respond to the new energy consumption demand. Specifically, when implemented, the capacity allocation ratio of the new energy consumption scenario is extracted from the energy storage configuration plan and calculated and processed with the capacity parameter of the new energy consumption scenario to obtain the correction value of the new energy consumption scenario, forming the correction parameter for the consumption scenario, including the adjusted capacity allocation ratio and the power fluctuation suppression strategy.

[0052] Using the capacity parameter of the frequency regulation scenario to correct the calculation of the capacity allocation ratio in the frequency regulation scenario of the energy storage configuration plan to obtain the correction parameter for the frequency regulation scenario is the fourth key link in data correction processing. The correction calculation refers to a data processing method for specialized adjustment for a specific technical scenario. This process can be expressed by the following formula:

[0053] where, represents the correction parameter for the frequency regulation scenario, represents the reference correction coefficient, represents the inertia constant of the power system, represents the maximum allowable frequency deviation, represents the rated frequency, represents the frequency regulation contribution factor, represents the capacity parameter of the frequency regulation scenario, represents the capacity allocation value of the frequency regulation scenario in the original configuration plan, represents the grid damping coefficient, represents the weight of the q-th type of frequency regulation event, represents the severity coefficient of the q-th type of frequency regulation event, It represents the response time requirement for the q-th type of frequency modulation event, and L represents the number of types of frequency modulation events. The relationship between system inertia and frequency stability is reflected by an exponential function. At the same time, the characteristics and response requirements of different types of frequency modulation events are considered, so that the calculated correction parameters can meet the technical requirements of power grid frequency regulation. During implementation, the capacity allocation and response characteristics of the frequency regulation scenario are extracted from the energy storage configuration plan, combined with the capacity parameters of the frequency regulation scenario and the power grid frequency characteristics, and correction calculations are carried out to obtain the correction parameters for the frequency modulation scenario, including the adjusted capacity allocation, response time, and power configuration. Matching the capacity parameters of the voltage support scenario with the capacity allocation ratio in the voltage support scenario of the energy storage configuration plan to obtain the voltage scenario correction parameters is the fifth key link in data correction processing. Data matching refers to the data processing method of determining the most suitable configuration through comparison and analysis. Among them, the capacity allocation ratio and reactive power configuration of the voltage support scenario are extracted from the energy storage configuration plan, and then compared and analyzed with the capacity parameters of the voltage support scenario to determine the capacity value and power value that need to be adjusted. Then, combined with the voltage sensitivity of the power grid node and the characteristics of reactive power demand, the optimal adjustment plan is determined, and finally the voltage scenario correction parameters are formed, including the corrected capacity allocation ratio, reactive power configuration, and response strategy. The accurate calculation of the voltage scenario correction parameters is of great significance for ensuring the voltage quality of the power grid.

[0054] Aggregating the peak-valley scenario correction parameters, accommodation scenario correction parameters, frequency modulation scenario correction parameters, and voltage scenario correction parameters to obtain the operation optimization parameters is the last key link in data correction processing. Parameter aggregation refers to the data processing method of integrating the results of independent correction of each scenario into a unified operation strategy. During implementation, the correction parameters of the four scenarios are converted in format according to a unified data structure, and then the consistency and conflict between the correction parameters of each scenario are checked to eliminate potential operation conflicts. Then, coordinated processing is carried out according to the scenario priority and timing characteristics to form a strategy for the coordinated operation of each scenario, and finally integrated into the complete operation optimization parameters, including the capacity allocation ratio, power configuration, response priority, and operation period arrangement of each scenario. The operation optimization parameters are the guiding basis for the multi-scenario coordinated operation of the energy storage system, directly affecting the utilization efficiency of energy storage resources and the overall performance of the system.

[0055] In a energy storage planning project of a certain power enterprise, after the technical staff completed the benefit assessment, they carried out detailed data calibration processing on the energy storage configuration plan. The capacity adjustment parameter with the total capacity adjusted from the original plan of 15 MWh to 12 MWh was used for scenario decomposition. Considering the actual situation of a large peak-valley electricity price difference, a high new energy penetration rate, insufficient system inertia, and frequent voltage fluctuations, according to the weight ratios of 30%, 40%, 20%, and 10%, the sub-scenario capacity parameters were decomposed, with the peak-valley regulation scenario capacity reduced by 0.9 MWh, the new energy consumption scenario capacity reduced by 1.2 MWh, the frequency regulation scenario capacity reduced by 0.6 MWh, and the voltage support scenario capacity reduced by 0.3 MWh. Then, the numerical correction was made on the peak-valley regulation scenario capacity parameter and the 30% allocation ratio in the original configuration plan. Due to the reduction of the total capacity, the peak-valley regulation capacity was adjusted from 4.5 MWh to 3.6 MWh, but by optimizing the regulation period, the peak-valley regulation power was kept unchanged, and the corrected parameters of the peak-valley scenario including the new allocation ratio and period adjustment were obtained. Next, according to the new energy consumption scenario capacity parameter and the 40% allocation ratio in the original configuration plan, considering the characteristics of large fluctuations in photovoltaic power generation and rapid growth of wind power installed capacity in this area, the corrected parameters of the new energy consumption scenario were obtained by applying the correction value calculation formula, including the adjusted capacity allocation and power smoothing strategy. At the same time, the frequency regulation scenario capacity parameter was used to correct the 20% allocation ratio in the original configuration plan. Considering the grid frequency fluctuation characteristics and response time requirements, the corrected parameters of the frequency regulation scenario were calculated, and the frequency response strategy was optimized. In addition, the voltage support scenario capacity parameter was matched with the 10% allocation ratio in the original configuration plan, and combined with the analysis of the node voltage sensitivity of the distribution network, the corrected parameters of the voltage scenario were obtained. Finally, the corrected parameters of the four scenarios were comprehensively analyzed and conflict checked to ensure the coordinated operation among the scenarios.

[0056] The energy storage planning method for power systems in the embodiments of the present application has been described above. Next, the energy storage planning system for power systems in the embodiments of the present application will be described. Please refer to Figure 3 , an embodiment of the energy storage planning system for power systems in the embodiments of the present application includes: An extraction module, configured to obtain a power system operation feature vector including load change features, power generation output features, and grid operation features through feature extraction processing according to the real-time operation data of the power grid and the operation data of the energy storage system; An analysis module, configured to perform time series analysis processing on the load data and new energy power generation data in the power system operation feature vector to obtain a prediction data set including future load demands and power generation amounts; A statistics module, configured to perform probability statistical analysis on the operation data of the energy storage system in the peak-valley regulation scenario, new energy consumption scenario, frequency regulation scenario, and voltage support scenario based on the prediction data set to obtain energy storage demand data; A configuration module, configured to determine energy storage capacity configuration parameters through multi-scenario optimization calculation according to energy storage demand data, and obtain an energy storage configuration plan including the total energy storage capacity, power configuration, and capacity allocation ratios under different scenarios; A correction module, configured to dynamically correct the energy storage planning plan through comprehensive evaluation of the power system based on the energy storage configuration plan, and obtain a target energy storage plan including benefit evaluation indicators, capacity adjustment parameters, and operation optimization parameters.

[0057] Refer to Figure 4 , in an embodiment of the present invention, a computer device is further provided. The computer device may be a server, and its internal structure may be as Figure 4 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0058] Those skilled in the art can understand that Figure 4 the structure shown in

[0059] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0060] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0061] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, systems, and units can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0062] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc., which can store program codes.

[0063] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A method for energy storage planning for a power system, characterized in that: The energy storage planning method for the power system comprises: Based on the real-time operation data of the power grid and the operation data of the energy storage system, a power system operation feature vector including load change characteristics, power generation output characteristics and power grid operation characteristics is obtained through feature extraction processing; Perform time series analysis on the load data and renewable energy generation data in the power system operation feature vector to obtain a forecast data set containing future load demand and power generation; Based on the forecast data set, the operation data of the energy storage system in peak-valley regulation scenarios, new energy consumption scenarios, frequency regulation scenarios and voltage support scenarios are analyzed by probability statistics to obtain energy storage demand data; Based on the energy storage demand data, the energy storage capacity configuration parameters are determined through multi-scenario optimization calculations to obtain an energy storage configuration solution that includes total energy storage capacity, power configuration, and capacity allocation ratios under different scenarios; Based on the energy storage configuration plan, the energy storage planning plan is dynamically revised through comprehensive evaluation of the power system to obtain a target energy storage plan that includes benefit evaluation indicators, capacity adjustment parameters and operation optimization parameters.

2. The energy storage planning method for a power system according to claim 1, characterized in that: The power system operation feature vector including load change characteristics, power generation output characteristics and power grid operation characteristics is obtained through feature extraction processing based on the real-time operation data of the power grid and the operation data of the energy storage system, including: Collecting load curve data, renewable energy power generation output data and grid frequency fluctuation data from the real-time operation data of the grid; Performing peak-to-valley difference calculation and load factor statistics on the load curve data to obtain the load change characteristics; Calculating the power generation fluctuation rate and the power generation capacity coefficient on the renewable energy power generation output data to obtain the power generation output characteristics; Calculating the voltage qualification rate and frequency deviation statistics on the power grid frequency fluctuation data to obtain the power grid operation characteristics; The load variation characteristics, the power generation output characteristics and the power grid operation characteristics are subjected to data fusion processing to obtain the power system operation characteristic vector.

3. The energy storage planning method for a power system according to claim 1, characterized in that: The load data and new energy generation data in the power system operation feature vector are subjected to time series analysis and processing to obtain a forecast data set containing future load demand and power generation, including: Generate load time series sample data using the load data in the power system operation characteristic vector; Extracting new energy power generation data from the power system operation feature vector to generate power generation time series sample data; Performing time-stamping processing on the load time series sample data to obtain load forecast data including time period characteristics, date characteristics and holiday characteristics; Performing meteorological factor correlation processing on the power generation time series sample data to obtain power generation prediction data including temperature characteristics, humidity characteristics, light characteristics and wind speed characteristics; The load forecast data and the power generation forecast data are subjected to time series fusion processing to obtain the forecast data set including future load demand and power generation.

4. The energy storage planning method for a power system according to claim 1, characterized in that: Based on the prediction data set, the operation data of the energy storage system in the peak-valley regulation scenario, the new energy consumption scenario, the frequency regulation scenario and the voltage support scenario are subjected to probability statistical analysis to obtain the energy storage demand data, including: Using the predicted data set to perform peak shaving and valley filling analysis on the load fluctuation data in the peak-valley regulation scenario, to obtain peak-valley regulation capacity data; According to the prediction data set, the power fluctuation data in the new energy consumption scenario is smoothed to obtain power capacity data; Extracting features of frequency response data in a frequency regulation scenario based on the predicted data set to obtain frequency regulation capacity data; Performing statistical processing on reactive power compensation data in the voltage support scenario in the prediction data set to obtain voltage support capacity data; The peak-valley regulation capacity data, the power capacity data, the frequency regulation capacity data and the voltage support capacity data are aggregated to obtain the energy storage demand data.

5. The energy storage planning method for a power system according to claim 1, characterized in that: The energy storage capacity configuration parameters are determined through multi-scenario optimization calculation according to the energy storage demand data to obtain an energy storage configuration solution including total energy storage capacity, power configuration and capacity allocation ratio under different scenarios, including: Decomposing the energy storage demand data into energy storage equipment investment cost data, equipment maintenance cost data and equipment replacement cost data to obtain a cost data set; Based on the energy storage demand data, a time series analysis is performed on the peak-valley electricity price difference, and a revenue calculation is performed in combination with the frequency regulation market compensation mechanism to obtain a revenue data set; Performing attenuation characteristic analysis on the life curve of the energy storage equipment in the energy storage demand data, and calculating the equipment failure probability, to obtain an equipment reliability data set; Perform multi-dimensional constraint processing on the cost data set, the benefit data set, and the equipment reliability data set to obtain energy storage capacity calculation parameters; Utilizing the energy storage capacity calculation parameters to perform total capacity optimization and state of charge analysis to obtain the total energy storage capacity; Allocating the total energy storage capacity according to the charge and discharge power limit to obtain the power configuration; The total energy storage capacity and the power configuration are proportionally divided according to the demand priorities of the peak-valley regulation scenario, the new energy consumption scenario, the frequency regulation scenario and the voltage support scenario to obtain the energy storage configuration plan.

6. The energy storage planning method for a power system according to claim 1, characterized in that: The energy storage configuration scheme is based on the energy storage planning scheme, which is dynamically modified through comprehensive evaluation of the power system to obtain a target energy storage scheme including benefit evaluation indicators, capacity adjustment parameters and operation optimization parameters, including: Extract peak-valley regulation revenue data, frequency regulation compensation data, and new energy consumption revenue data from the energy storage configuration scheme, perform numerical accumulation processing, and obtain energy storage revenue evaluation data; Generating initial cost data, operation and maintenance cost data, and equipment replacement cost data for the total energy storage capacity in the energy storage configuration scheme, and performing full-cycle calculations to obtain energy storage cost assessment data; Comparing and analyzing the energy storage benefit assessment data with the energy storage cost assessment data, and processing them according to a net present value calculation method to obtain the benefit assessment index; According to the benefit evaluation index, the total energy storage capacity and power configuration in the energy storage configuration scheme are corrected and compared to obtain the capacity adjustment parameter; Performing data correction processing on the capacity adjustment parameters and the capacity allocation ratios under different scenarios in the energy storage configuration scheme to obtain the operation optimization parameters; Boundary constraint processing is performed on the numerical range of the benefit evaluation index and the capacity adjustment parameter, and data fusion is performed with the operation optimization parameter to obtain the target energy storage solution.

7. The energy storage planning method for a power system according to claim 6, characterized in that: The step of performing data correction processing on the capacity adjustment parameter and the capacity allocation ratio under different scenarios in the energy storage configuration scheme to obtain the operation optimization parameter includes: Decomposing the capacity adjustment parameters by scenarios to obtain capacity parameters for peak-valley regulation scenarios, capacity parameters for new energy consumption scenarios, capacity parameters for frequency regulation scenarios, and capacity parameters for voltage support scenarios; Numerically correct the peak-valley regulation scenario capacity parameter and the capacity allocation ratio under the peak-valley regulation scenario in the energy storage configuration scheme to obtain a peak-valley scenario correction parameter; Calculate a correction value from the capacity parameter of the new energy consumption scenario and the capacity allocation ratio under the new energy consumption scenario in the energy storage configuration scheme to obtain a consumption scenario correction parameter; Using the frequency regulation scenario capacity parameter, a correction calculation is performed on the capacity allocation ratio under the frequency regulation scenario in the energy storage configuration scheme to obtain a frequency regulation scenario correction parameter; Performing data matching on the voltage support scenario capacity parameter and the capacity allocation ratio under the voltage support scenario in the energy storage configuration scheme to obtain a voltage scenario correction parameter; The peak-valley scenario correction parameter, the absorption scenario correction parameter, the frequency modulation scenario correction parameter, and the voltage scenario correction parameter are aggregated to obtain the operation optimization parameter.

8. An energy storage planning system for a power system, used to implement the energy storage planning method for a power system as described in any one of claims 1 to 7, characterized in that: The energy storage planning system for a power system comprises: An extraction module is used to obtain a power system operation feature vector including load change characteristics, power generation output characteristics and power grid operation characteristics through feature extraction processing based on real-time power grid operation data and energy storage system operation data; An analysis module is used to perform time series analysis on the load data and new energy generation data in the power system operation feature vector to obtain a forecast data set containing future load demand and power generation; The statistical module is used to perform probabilistic statistical analysis on the operating data of the energy storage system in peak-valley regulation scenarios, new energy consumption scenarios, frequency regulation scenarios, and voltage support scenarios based on the predicted data set to obtain energy storage demand data; A configuration module is used to determine the energy storage capacity configuration parameters through multi-scenario optimization calculation according to the energy storage demand data, and obtain an energy storage configuration plan including the total energy storage capacity, power configuration and capacity allocation ratio under different scenarios; The correction module is used to dynamically correct the energy storage planning scheme based on the energy storage configuration scheme through comprehensive evaluation of the power system, and obtain a target energy storage scheme that includes benefit evaluation indicators, capacity adjustment parameters and operation optimization parameters.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program executable on the processor, and is characterized in that when the processor executes the computer program, the energy storage planning method for a power system described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is enabled to execute the energy storage planning method for a power system according to any one of claims 1 to 7.

Images