A new energy cluster deviation complementary energy storage capacity demand optimization method, system, device and medium

Abstract

This invention relates to the fields of energy storage system planning and optimization and power system operation and control technology, and discloses a method, system, equipment, and medium for optimizing energy storage capacity demand based on complementary deviations in new energy clusters. The method includes: collecting historical actual and predicted power output data of each power station within the new energy cluster, and calculating the time-series power output deviation; establishing a probability distribution model of the power output deviation of each power station using kernel density estimation; constructing a joint distribution model of the power output deviation of each power station using a Copula function; combining the probability distribution and the joint distribution, generating a set of deviation scenarios for each power station using Monte Carlo simulation and calculating the total cluster deviation; determining the energy storage technology capacity demand based on the scenario set according to a pre-set confidence level; constructing an optimization model with technical capacity as a constraint and maximizing the net present value over the entire life cycle as the objective, and solving for the optimal solution. This method utilizes the complementary effect of cluster deviations to improve the accuracy and economy of energy storage configuration, providing a scientific basis for energy storage planning in new energy clusters.

Description

Technical Field

[0001] This invention relates to the fields of energy storage system planning and optimization and power system operation and control technology, specifically to a method, system, equipment and medium for optimizing energy storage capacity demand for complementary deviations in new energy clusters. Background Technology

[0002] With the large-scale grid connection of new energy sources such as wind power and photovoltaics, the volatility and uncertainty of their output pose a severe challenge to the stable operation of the power system. To mitigate the fluctuations in new energy output and reduce prediction errors, configuring energy storage systems has become an important technical means. Current energy storage capacity configuration methods have the following shortcomings: First, the configuration methods are simplistic, failing to fully utilize the complementary output effects between different power plants in the new energy cluster, leading to redundant energy storage capacity and high investment costs; second, the configuration accuracy is insufficient, lacking detailed modeling of the probability distribution characteristics and correlations of output deviations, resulting in overly conservative or aggressive capacity configurations; third, there is a lack of comprehensive economic optimization, failing to achieve a balance between energy storage investment costs and operational benefits. Currently, the configuration methods for energy storage capacity in new energy clusters have not systematically considered the complementary characteristics of deviations between power plants, joint probability distributions, and full life-cycle economic optimization, and related research remains significantly lacking. Summary of the Invention

[0003] In view of the existing problems mentioned above, the present invention provides a method, system, equipment and medium for optimizing energy storage capacity demand based on the complementary deviation of new energy clusters.

[0004] Therefore, the technical problem solved by this invention is: how to achieve a multi-objective balance optimization of improving energy storage configuration accuracy, reducing investment costs and increasing operating benefits in the configuration of energy storage capacity in new energy clusters by modeling the complementary characteristics and joint probability distribution of the output deviation between power stations in the system, and constructing a comprehensive economic optimization model for the whole life cycle.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for optimizing energy storage capacity demand based on complementary deviations in new energy clusters, comprising,

[0006] Collect historical actual and predicted power output data of each power station in the new energy cluster, and calculate the time-series power output deviation data of each power station;

[0007] Based on the time-series power output deviation data, a probability distribution model of the power output deviation of each station is established using the kernel density estimation method;

[0008] Using time-series power output deviation data, a joint distribution model of the correlation structure among the power output deviations of each station is established through the Copula function;

[0009] By combining probability distribution models and joint distribution models, Monte Carlo simulations are used to generate a set of deviation scenarios for each station, and the total cluster deviation under each scenario is calculated.

[0010] Based on the scenario set of total cluster deviation, the technical capacity requirements of the energy storage system are determined according to the preset confidence level.

[0011] An economic optimization model is constructed with the technical capacity requirements of the energy storage system as a constraint and the maximization of the net present value of the energy storage system throughout its entire life cycle as the objective function. The optimal energy storage configuration scheme is obtained by solving the model.

[0012] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters described in this invention, the step of collecting historical actual and predicted power output data of each power station within the new energy cluster, and calculating the time-series power output deviation data of each power station, includes:

[0013] The data period covers at least one full year to include the characteristics of all four seasons, and the absolute deviation is normalized to the relative deviation, calculated using the following formula:

[0014] ,

[0015] in, To normalize the relative bias, dimensionless. For the first The venue is at the first The output deviation at any moment, For the first The rated installed capacity of each station;

[0016] After preprocessing, a complete time series dataset of power output deviation for each station is obtained, which serves as the basic input for subsequent probabilistic modeling.

[0017] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters described in this invention, the step of establishing a probability distribution model of the output deviation of each power station based on time-series output deviation data using a kernel density estimation method includes:

[0018] The kernel density estimation method is used to establish the nonparametric probability density function of the power output deviation of each station. For the th Deviation samples from individual stations The calculation expression is:

[0019] ,

[0020] in, For the first The probability density function of the deviation of each station The total number of samples, For bandwidth parameters, For kernel function, For the input variables of the probability density function, For the first The venue is at the first Normalized relative deviation values ​​for each sample point;

[0021] The Silverman criterion is used for automatic selection, and the continuous probability distribution of the deviation of each station is obtained through kernel density estimation.

[0022] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters described in this invention, the method involves: using time-series output deviation data and establishing a joint distribution model of the correlation structure between output deviations of each power station through a Copula function, including...

[0023] By analyzing the correlation between the output deviations of different power plants within the new energy cluster, the correlation coefficient matrix between the deviations of each power plant is calculated. The calculation formula is as follows:

[0024] ,

[0025] in, For station Hechang The correlation coefficient of the deviation For station and station The covariance of the bias and Stations and station The standard deviation of the deviation;

[0026] The Copula function is introduced for modeling to separate the marginal distribution from the relevant structure, construct a multivariate joint distribution, and generate joint samples of deviations for each station.

[0027] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters as described in this invention, the method combines a probability distribution model and a joint distribution model to generate a set of deviation scenarios for each site through Monte Carlo simulation, and calculates the total cluster deviation under each scenario, including...

[0028] Extract relevant uniformly distributed random number vectors from the Copula function, and transform them using the inverse cumulative distribution function of the deviation distribution of each station to obtain deviation samples with true marginal distribution and related structure;

[0029] For each scenario The computing cluster in the The total deviation at time is given by the formula:

[0030] ,

[0031] in, For the scene Next Total cluster bias at time 10:00 The total number of stations, For the scene Next Deviation of each station.

[0032] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters as described in this invention, the method for determining the technical capacity requirements of the energy storage system based on a scenario set of total cluster deviations and a preset confidence level includes:

[0033] Based on the generated cluster total deviation scenario data, a mapping relationship between energy storage capacity demand and confidence level is established, and the confidence level is extracted. The corresponding deviation value is used as the energy storage capacity requirement, and the calculation formula is as follows:

[0034] ,

[0035] in, For the first At all times with confidence The energy storage capacity requirement is expressed in megawatt-hours. Confidence level of the total bias distribution of the cluster Quantiles For time intervals.

[0036] As a preferred embodiment of the energy storage capacity demand optimization method for complementary deviations in new energy clusters described in this invention, the step of constructing an economic optimization model with the technical capacity demand of the energy storage system as a constraint and the maximization of the net present value of the energy storage system over its entire life cycle as the objective function, and solving for the optimal energy storage configuration scheme, includes:

[0037] The total investment cost of an energy storage system is calculated using the following formula:

[0038] ,

[0039] in, The total investment cost is expressed in ten thousand yuan. Cost per unit capacity battery Configure the capacity of the energy storage system, in megawatt-hours. For the unit power conversion system cost, The rated power of the energy storage system, Other costs;

[0040] The net present value of the system is:

[0041] ,

[0042] in, Net present value, For the operating revenue in year y, The maintenance cost for year y is... The discount rate is... Design lifespan for energy storage systems, The total investment cost is expressed in ten thousand yuan.

[0043] By solving the optimization model using a genetic algorithm, the most economically efficient energy storage capacity configuration scheme can be obtained.

[0044] This invention collects power output deviation data from new energy clusters and uses kernel density estimation and Copula function to establish probability distribution and correlation structure models, which can accurately characterize the deviation complementarity characteristics of each power station and optimize energy storage capacity requirements.

[0045] This invention provides a new energy cluster deviation complementary energy storage capacity demand optimization system, comprising:

[0046] The data acquisition and processing module collects historical actual and predicted power output data from each power station in the new energy cluster, and performs data cleaning and normalization to generate time-series power output deviation data for each power station.

[0047] The probability distribution modeling module, based on the time-series output deviation data of each station, uses the kernel density estimation method to establish a probability distribution model of the output deviation of each station;

[0048] The correlation modeling and joint distribution construction module analyzes the correlation between the output deviations of each station through the Copula function and establishes a joint distribution model of the deviations of each station.

[0049] The scene generation and total deviation calculation module, based on the probability distribution model and the joint distribution model, uses Monte Carlo simulation to generate a set of deviation scenarios for each station and calculates the total cluster deviation under each scenario.

[0050] The energy storage technology capacity requirement module determines the technical capacity requirement of the energy storage system based on the total deviation scenario set of the cluster and the preset confidence level.

[0051] The optimization model construction and solution module uses the technical capacity requirements of the energy storage system as a constraint and the net present value over the entire life cycle as the objective to construct an economic optimization model and use a genetic algorithm to solve it, thereby obtaining the optimal energy storage configuration scheme.

[0052] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a method for optimizing energy storage capacity demand based on the deviation complementarity of new energy clusters.

[0053] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a method for optimizing energy storage capacity demand based on the offset complementarity of new energy clusters are implemented.

[0054] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention establishes a method for optimizing energy storage capacity demand based on complementary deviations in new energy clusters. Through probability statistics and optimization theory, it comprehensively evaluates the joint distribution characteristics and complementary effects of output deviations at each site, achieving intensive configuration and life-cycle economic optimization of energy storage capacity. First, historical and predicted output data of each site in the new energy cluster are collected, and a normalized deviation sequence is calculated. Then, a non-parametric probability distribution model of the deviations at each site is constructed using kernel density estimation, and the Copula function is used to characterize the correlation structure of deviations between sites. Next, a large-scale deviation scenario with realistic statistical characteristics is generated through Monte Carlo simulation, calculating the total deviation distribution of the cluster and quantifying the complementary effect. Based on this, a mapping relationship between energy storage capacity demand and confidence level is established, and the lower limit of technical capacity that meets a given confidence level is determined. Finally, a life-cycle economic optimization model covering investment costs, operation and maintenance costs, savings in assessment costs, and ancillary service revenue is constructed, with the goal of maximizing net present value, to solve for the optimal configuration scheme of energy storage capacity. This method can significantly reduce the energy storage capacity demand caused by the complementary deviations between power stations, improve configuration accuracy and economy, and provide a scientific basis for the planning and investment of new energy cluster energy storage systems. Attached Figure Description

[0055] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 The flowchart illustrates a method for optimizing energy storage capacity requirements based on the complementary bias of new energy clusters, as provided in one embodiment of the present invention. Detailed Implementation

[0057] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0058] Example 1, referring to Figure 1 The first embodiment of the present invention provides a method for optimizing energy storage capacity demand based on complementary deviations in new energy clusters, comprising:

[0059] S1: Collect historical actual and predicted power output data of each power station in the new energy cluster, and calculate the time-series power output deviation data of each power station.

[0060] S2: Based on the time-series output deviation data, a probability distribution model of the output deviation of each station is established using the kernel density estimation method.

[0061] S3: Using time-series output deviation data, a joint distribution model of the correlation structure between output deviations of each station is established through the Copula function.

[0062] S4: Combining the probability distribution model and the joint distribution model, Monte Carlo simulation is used to generate a set of deviation scenarios for each station, and the total cluster deviation under each scenario is calculated.

[0063] S5: Based on the scenario set of total cluster deviation, determine the technical capacity requirements of the energy storage system according to the preset confidence level.

[0064] S6: Construct an economic optimization model with the technical capacity requirements of the energy storage system as a constraint and the maximization of the net present value of the energy storage system throughout its entire life cycle as the objective function, and solve for the optimal energy storage configuration scheme.

[0065] It should be noted that this embodiment establishes an energy storage capacity optimization model for complementary output deviations in a new energy cluster. This eliminates the need for separate energy storage capacity configuration for each new energy power station and can simulate the statistical characteristics and complementary effects of output deviations within the new energy cluster. This invention derives the probability distribution of the total cluster deviation by modeling the joint probability distribution of output deviations from each power station in the new energy cluster and using its Monte Carlo scene generation method. This allows for the determination of the energy storage capacity requirement that meets the confidence level and the optimization of economic efficiency throughout the entire lifecycle.

[0066] This embodiment also addresses the issue of energy storage capacity configuration in new energy clusters. To overcome the limitations of traditional methods, such as insufficient utilization of complementary effects, limited accuracy of probabilistic modeling, and lack of economic optimization, a deviation-complementary energy storage capacity demand optimization model is established based on the analysis of the statistical characteristics and related structures of the output deviations of each power station. A joint probability distribution of deviations for each power station is constructed using kernel density estimation and a Copula function. Monte Carlo simulation is then used to generate a total deviation scenario for the cluster, establishing a mapping relationship between energy storage capacity demand and confidence level. Finally, a full life-cycle economic optimization model is constructed, aiming to maximize net present value and synergistically optimize energy storage capacity and power configuration to achieve the optimal balance between technical feasibility and economic efficiency of the energy storage system. This method can effectively evaluate and utilize the deviation complementarity effect of each power station within the cluster, significantly reducing energy storage configuration capacity and investment costs, and improving the economic efficiency of new energy consumption and grid operation.

[0067] Example 2, refer to Figure 1 As an embodiment of the present invention, based on the above embodiment, a method for optimizing energy storage capacity demand based on the complementary deviation of new energy clusters is provided.

[0068] In this embodiment of the application, step S1 involves collecting historical actual and predicted power output data for each power station within the new energy cluster, and calculating the time-series power output deviation data for each power station. The specific steps include:

[0069] The system collects historical actual power output data and day-ahead predicted power output data from all power plants within the renewable energy cluster, with a time resolution of 15 minutes and a data period covering at least one full year to include seasonal characteristics. For a cluster containing M renewable energy power plants, in the... At that moment, the The power output deviation of each station is defined as the difference between the actual power output and the predicted power output:

[0070] ,

[0071] in, For the first The venue is at the first The output deviation at any given moment, in megawatts. To contribute practically, To contribute to prediction.

[0072] To eliminate the impact of differences in installed capacity, the absolute deviation is normalized to a relative deviation, and the calculation formula is as follows:

[0073] ,

[0074] in, To normalize the relative bias, dimensionless. For the first The venue is at the first The output deviation at any moment, For the first The rated installed capacity of each station.

[0075] The system performs quality checks on the collected raw data, removing outliers and missing values:

[0076] Outlier detection uses the three-standard-deviation criterion. When a data point deviates from the mean by more than three standard deviations, it is considered an outlier and removed.

[0077] Data integrity is required to be no less than 95%, and missing data should be filled using linear interpolation.

[0078] After preprocessing, a complete time series dataset of output deviation for each station is obtained, which serves as the basic input for subsequent probabilistic modeling.

[0079] In this embodiment of the invention, step S2, which involves establishing a probability distribution model of the power output deviation for each station based on time-series power output deviation data using a kernel density estimation method, specifically includes the following steps:

[0080] A nonparametric probability density function for the power output deviation of each station is established using kernel density estimation. Compared to parametric methods that assume a specific distribution type, kernel density estimation does not presuppose the distribution form and can more accurately fit the distribution characteristics of actual data.

[0081] Using the Gaussian kernel function, for the th Deviation samples from individual stations The calculation expression is:

[0082] ,

[0083] in, For the first The probability density function of the deviation of each station The total number of samples, The bandwidth parameter determines the smoothness of the density curve. For kernel function, , which is the input variable for the probability density function, used to calculate the probability density at a specific deviation value. For the first The venue is at the first Normalized relative deviation values ​​for each sample point.

[0084] Bandwidth parameters Automatic selection is performed using the Silverman criterion, and the calculation formula is as follows:

[0085] ,

[0086] in, The standard deviation of the sample is 1. The total number of samples.

[0087] By estimating kernel density, a continuous probability distribution of the deviations at each station is obtained. This distribution accurately reflects the actual statistical characteristics of the deviation data, including features such as mean, variance, skewness, and kurtosis.

[0088] The system extracts key statistical parameters of the deviation distribution of each station, which will be used for subsequent scene generation and capacity calculation.

[0089] The probability density function obtained from kernel density estimation can be transformed into a cumulative distribution function through numerical integration, providing a basis for calculating the confidence level of capacity demand.

[0090] In this embodiment of the invention, step S3 uses time-series power output deviation data and establishes a joint distribution model of the correlation structure between the power output deviations of each station through the Copula function. The specific steps include:

[0091] There is a certain correlation between the output deviations of different stations within the new energy cluster. This correlation stems from the regional characteristics of meteorological conditions. Accurately characterizing this correlation is key to assessing the complementary effect of deviations.

[0092] The system first calculates the linear correlation coefficient matrix between the deviations of each station. The calculation formula is as follows:

[0093] ,

[0094] in, For station and station The correlation coefficient of the deviation For station and station The covariance of the bias and Stations and station Standard deviation of the deviation.

[0095] The closer the absolute value of the correlation coefficient is to 1, the stronger the correlation; the closer it is to 0, the weaker the correlation.

[0096] To more comprehensively describe the joint distribution characteristics of the deviations at each station, the Copula function is introduced for modeling. The Copula function can separate the marginal distribution from the correlation structure and flexibly construct a multivariate joint distribution.

[0097] The Gaussian Copula joint distribution function is chosen to represent the multivariate normal distribution of the deviations at each station after standard normal transformation. Using the maximum likelihood estimation method, the parameter matrix of the Copula function is estimated based on historical deviation data. This parameter matrix fully describes the correlation structure of the deviations at each station. Using the Copula model, joint samples of deviations at each station with true correlation can be generated, providing a foundation for subsequent scenario analysis.

[0098] In this embodiment of the application, step S4 combines the probability distribution model and the joint distribution model to generate a set of deviation scenarios for each station through Monte Carlo simulation, and calculates the total cluster deviation under each scenario. The specific steps include:

[0099] Using the established edge distribution model and Copula related structure model, a large number of deviation scenarios for each station were generated through Monte Carlo simulation.

[0100] The scene generation process is as follows:

[0101] Relevant uniformly distributed random number vectors are extracted from the Copula function. These vectors are then transformed using the inverse cumulative distribution function of the deviation distribution at each site to obtain deviation samples with true marginal distribution and relevant structure. The number of generated scenarios is set to 10,000 to ensure the stability of the statistical results. For each scenario... The computing cluster in the The total deviation at time is given by the formula:

[0102] ,

[0103] in, For the scene Next Total cluster bias at time 10:00 The total number of stations, For the scene Next Deviation of each station.

[0104] By statistically analyzing the total cluster deviation across all scenarios, a probability distribution of the total cluster deviation is established. A deviation complementarity coefficient is defined to quantitatively evaluate the complementarity effect of the cluster, with the following formula:

[0105] ,

[0106] in, The deviation complementarity coefficient, The standard deviation of the total cluster deviation This is the sum of the standard deviations of the deviations at each station.

[0107] A larger complementarity coefficient indicates a more significant cluster complementarity effect, with the coefficient reaching its maximum when the deviations of each station are completely uncorrelated. This coefficient will serve as a key indicator for evaluating the cluster synergy effect, used to quantify the cost savings brought about by intensive configuration.

[0108] In this embodiment of the application, step S5, based on the scenario set of total cluster deviation, determines the technical capacity requirement of the energy storage system according to a preset confidence level. The specific steps include:

[0109] Based on the generated cluster total deviation scenario data, a mapping relationship between energy storage capacity demand and confidence level is established. For a given confidence level... Energy storage systems need to be able to cope with Bias fluctuations under probability.

[0110] The system calculates the total cluster deviation for all scenarios at each time step, sorts them by absolute value, and extracts the confidence score. The corresponding deviation value is taken as the energy storage capacity requirement at that moment, and the calculation formula is as follows:

[0111] ,

[0112] in, For the first At all times with confidence The energy storage capacity requirement is expressed in megawatt-hours. Confidence level of the total bias distribution of the cluster Quantiles The time interval is 0.25 hours in this method.

[0113] Because the deviations at different times are time-dependent, the actual capacity requirement of the energy storage system must also consider the cumulative effect over multiple consecutive time periods. The system calculates the cumulative deviation within a rolling time window, the length of which is determined based on the charging and discharging characteristics of the energy storage system, typically 2-4 hours. Within a window length of... Under these conditions, the energy storage capacity requirement is the difference between the maximum and minimum cumulative deviations within the window.

[0114] By traversing all time windows and all scenarios, the confidence level is obtained. The minimum required energy storage capacity. This capacity value represents the energy storage system configuration capacity that meets the deviation smoothing requirements at the technical level, serving as the lower bound of the technical constraints.

[0115] In this embodiment of the application, step S6 involves constructing an economic optimization model with the technical capacity requirement of the energy storage system as a constraint and maximizing the net present value of the energy storage system over its entire life cycle as the objective function, and solving for the optimal energy storage configuration scheme. Specific steps include:

[0116] The total investment cost of an energy storage system includes the cost of battery purchase, the cost of the power conversion system, and the cost of installation and commissioning. The calculation formula is as follows:

[0117] ,

[0118] in, The total investment cost is expressed in ten thousand yuan. Cost per unit capacity battery Configure the capacity of the energy storage system, in megawatt-hours. For the unit power conversion system cost, The rated power of the energy storage system, Other costs.

[0119] The annual operating revenue of an energy storage system mainly comes from reduced assessment costs and ancillary service revenue. Reduced assessment costs are calculated according to grid assessment rules, while ancillary service revenue is determined based on local peak-shaving and frequency regulation compensation policies. The net present value of the system is... :

[0120] ,

[0121] in , is the net present value. For the operating revenue in year y, The maintenance cost in year y is typically taken as 2% of the investment cost. The discount rate is 8%. The design life of the energy storage system is assumed to be 15 years. The total investment cost is expressed in ten thousand yuan.

[0122] The optimization model uses net present value maximization as the objective function, and the energy storage system capacity as the decision variable. and power The constraints include the lower bound of technical requirements for capacity calculation, the matching relationship between power and capacity, and the operational constraints of the energy storage system's state of charge. The optimization model is solved using a genetic algorithm to obtain the economically optimal energy storage capacity configuration. The system output optimization results include the recommended configuration capacity, corresponding investment cost, expected payback period, and net present value, providing a quantitative basis for investment decisions.

[0123] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that:

[0124] This embodiment also provides a new energy cluster deviation complementary energy storage capacity demand optimization system, including:

[0125] The data acquisition and processing module collects historical actual and predicted power output data from each power station in the new energy cluster, and performs data cleaning and normalization to generate time-series power output deviation data for each power station.

[0126] The probability distribution modeling module, based on the time-series output deviation data of each station, uses the kernel density estimation method to establish a probability distribution model of the output deviation of each station;

[0127] The correlation modeling and joint distribution construction module analyzes the correlation between the output deviations of each station through the Copula function and establishes a joint distribution model of the deviations of each station.

[0128] The scene generation and total deviation calculation module, based on the probability distribution model and the joint distribution model, uses Monte Carlo simulation to generate a set of deviation scenarios for each station and calculates the total cluster deviation under each scenario.

[0129] The energy storage technology capacity requirement module determines the technical capacity requirement of the energy storage system based on the total deviation scenario set of the cluster and the preset confidence level.

[0130] The optimization model construction and solution module uses the technical capacity requirements of the energy storage system as a constraint and the net present value over the entire life cycle as the objective to construct an economic optimization model and use a genetic algorithm to solve it, thereby obtaining the optimal energy storage configuration scheme.

[0131] This embodiment also provides an electronic device applicable to the optimization of energy storage capacity demand for complementary new energy clusters, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the energy storage capacity demand optimization method for complementary new energy clusters as proposed in the above embodiment.

[0132] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements a method for optimizing energy storage capacity requirements based on the complementary deviations of new energy clusters as proposed in the above embodiments.

[0133] The storage medium proposed in this embodiment belongs to the same inventive concept as the energy storage capacity demand optimization method for achieving complementary deviations in a new energy cluster proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0134] Example 4, the fourth embodiment of the present invention, provides a computational analysis and simulation verification of a method for optimizing energy storage capacity demand based on complementary deviations in new energy clusters, including:

[0135] This example is validated based on a real-world renewable energy cluster comprising three wind farms and two photovoltaic power plants, with a total installed capacity of 500 MW. The wind farms have installed capacities of 120 MW, 100 MW, and 80 MW respectively, while the photovoltaic power plants have installed capacities of 100 MW and 100 MW respectively. One year's actual power output data and day-ahead forecast data for this cluster were collected, totaling 35,040 data points with a time resolution of 15 minutes. The optimization method proposed in this invention was implemented using the MATLAB platform, performing complete data analysis and optimization calculations.

[0136] (1) Analysis of deviation statistical characteristics. The output deviation of each station was calculated. The statistical results show that the standard deviations of the deviations of the three wind farms are 18.5 MW, 15.2 MW and 12.8 MW, respectively, and the standard deviations of the deviations of the two photovoltaic power stations are 16.3 MW and 15.9 MW, respectively. The sum of the standard deviations of the five stations is 78.7 MW, while the standard deviation of the total deviation of the cluster is only 42.3 MW. According to the deviation complementarity coefficient formula, the deviation complementarity coefficient is calculated to be 0.46, indicating that there is a significant complementary effect among the stations in the cluster.

[0137] (2) Correlation Analysis. The correlation coefficient matrix of the deviations between the power stations was calculated. The results showed that the correlation coefficient between wind farms was between 0.35 and 0.52, the correlation coefficient between photovoltaic power stations was 0.68, and the correlation coefficient between wind farms and photovoltaic power stations was between 0.12 and 0.28. This moderate positive correlation indicates that the power stations have both a certain degree of synchronicity and sufficient complementary space. The joint distribution of the five power stations was modeled using the Gaussian Copula function, and the parameters were determined by maximum likelihood estimation. The goodness-of-fit test of the model was passed.

[0138] (3) Scene Generation and Capacity Requirement Calculation. Based on the established probability model, 10,000 scene samples were generated, each containing 35,040 time points of deviation data from 5 power stations. For each scene, the total deviation time series of the cluster was calculated, and the deviation distribution characteristics of all scenes were statistically analyzed. At a 95% confidence level, using a 2-hour rolling window, the minimum energy storage capacity required to meet the technical requirements was calculated to be 85 MWh. In comparison, if the traditional method is used to independently configure energy storage for each power station, the total capacity required at the same confidence level is 146 MWh, achieving a capacity saving of 41.8%.

[0139] (4) Economic Optimization Analysis. Based on local electricity pricing policies and assessment standards, the investment cost and operating revenue under different capacity configurations were calculated. The optimization results show that, under a 95% confidence level technical constraint, the optimal capacity is 90 MWh (slightly higher than the minimum technical requirement for better economic efficiency), corresponding to a power output of 30 MW. The total investment cost is RMB 129.6 million, the expected annual revenue is RMB 15.8 million, the payback period is 8.2 years, and the 15-year net present value is RMB 52.3 million. If the confidence level requirement is lowered to 90%, the optimal capacity decreases to 72 MWh, the investment cost decreases to RMB 103.68 million, and the payback period is shortened to 7.5 years, but the assessment risk increases. Sensitivity analysis is used to provide optimization recommendations under different confidence levels, electricity prices, and assessment standards.

[0140] As shown in Table 1, compared with the traditional independent configuration method, the method of this invention achieves a 38.4% capacity saving, a 38.4% reduction in investment cost, and a 3.3-year shorter investment payback period at a 95% confidence level. Even under the same technical specifications, this method can significantly reduce energy storage demand by fully utilizing the cluster complementarity effect.

[0141] Table 1 Comparison of Energy Storage Capacity Configurations Using Different Methods

[0142]

[0143] Table 2 shows that as the cluster size increases and the number of sites increases, the deviation complementarity coefficient gradually improves, and the energy storage capacity saving rate also increases accordingly. This verifies the superiority of clustered configuration over independent configuration, and the scale effect is significant.

[0144] Table 2 Quantitative Analysis of Deviation Complementarity Effect

[0145]

[0146] Table 3 shows that increasing the confidence level from 90% to 95% increases investment costs by 25%, but significantly reduces performance fees, resulting in better overall economic performance. Further increasing the confidence level to 99% leads to a substantial increase in investment costs, while the reduction in performance fees is limited, resulting in a decrease in net present value. A 95% confidence level is the optimal choice in this case.

[0147] Table 3 Technical and economic indicators at different confidence levels

[0148]

[0149] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0150] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A new energy cluster bias complementary energy storage capacity demand optimization method, characterized in that: include, ​ Collect historical actual and predicted power output data of each power station in the new energy cluster, and calculate the time-series power output deviation data of each power station; The process involves collecting historical actual and predicted power output data from each power station within the new energy cluster, and calculating the time-series power output deviation data for each power station, including... The data period covers at least one full year to include the characteristics of all four seasons, and the absolute deviation is normalized to the relative deviation, calculated using the following formula: ； wherein, is the normalized relative deviation, dimensionless, is the power output deviation of the th power plant at the th time instant, is the rated installed capacity of the th power plant; After preprocessing, a complete time series dataset of output deviation for each station is obtained, which serves as the basic input for subsequent probabilistic modeling. Based on the time-series power output deviation data, a probability distribution model of the power output deviation of each station is established using the kernel density estimation method; The method for establishing a probability distribution model of the power output deviation for each station based on time-series power output deviation data and using kernel density estimation includes: The kernel density estimation method is used to establish the nonparametric probability density function of the power output deviation of each station. For the th Deviation samples from individual stations The calculation expression is: ； in, For the first The probability density function of the deviation of each station The total number of samples, For bandwidth parameters, For kernel function, For the input variables of the probability density function, For the first The venue is at the first Normalized relative deviation values ​​for each sample point; The Silverman criterion is used for automatic selection, and the continuous probability distribution of the deviation of each station is obtained through kernel density estimation. Using time-series power output deviation data, a joint distribution model of the correlation structure among the power output deviations of each station is established through the Copula function; The method employs time-series output deviation data and uses the Copula function to establish a joint distribution model of the correlation structure among the output deviations of each power station, including... By analyzing the correlation between the output deviations of different power plants within the new energy cluster, the correlation coefficient matrix between the deviations of each power plant is calculated. The calculation formula is as follows: ； in, For station and station The correlation coefficient of the deviation For station and station The covariance of the bias and Stations and station The standard deviation of the deviation; The Copula function is introduced for modeling to separate the marginal distribution from the relevant structure, construct the joint distribution of each variable, and generate joint samples of the deviation of each station. By combining probability distribution models and joint distribution models, Monte Carlo simulations are used to generate a set of deviation scenarios for each station, and the total cluster deviation under each scenario is calculated. Based on the scenario set of total cluster deviation, the technical capacity requirements of the energy storage system are determined according to the preset confidence level. An economic optimization model is constructed with the technical capacity requirements of the energy storage system as a constraint and the maximization of the net present value of the energy storage system throughout its entire life cycle as the objective function. The optimal energy storage configuration scheme is obtained by solving the model.

2. The new energy cluster deviation complementary energy storage capacity demand optimization method of claim 1, wherein: The combined probability distribution model and joint distribution model generate a set of deviation scenarios for each station through Monte Carlo simulation, and calculate the total cluster deviation under each scenario, including... Extract relevant uniformly distributed random number vectors from the Copula function, and transform them using the inverse cumulative distribution function of the deviation distribution of each station to obtain deviation samples with true marginal distribution and related structure; For each scenario , the total deviation of the cluster at the moment is calculated, with the formula: ； in, For the scene Next Total cluster bias at time 10:00 The total number of stations, For the scene Next Deviation of each station.

3. The new energy cluster deviation complementary energy storage capacity demand optimization method of claim 2, wherein: The scenario set based on the total cluster deviation determines the technical capacity requirements of the energy storage system according to a preset confidence level, including: Based on the generated cluster total deviation scene data, a mapping relationship between energy storage capacity demand and confidence is established, and the confidence is extracted The corresponding deviation value is taken as the energy storage capacity demand, and the calculation formula is as follows: ； in, For the first At all times with confidence The energy storage capacity requirement is expressed in megawatt-hours. Confidence level of the total bias distribution of the cluster Quantiles For time intervals.

4. The energy storage capacity demand optimization method for complementary deviations in new energy clusters as described in claim 3, characterized in that: The construction of an economic optimization model, which takes the technical capacity requirements of the energy storage system as constraints and maximizes the net present value of the energy storage system over its entire life cycle as the objective function, yields the optimal energy storage configuration scheme. This includes... The total investment cost of an energy storage system is calculated using the following formula: ； in, The total investment cost is expressed in ten thousand yuan. Cost per unit capacity battery Configure the capacity of the energy storage system, in megawatt-hours. For the unit power conversion system cost, The rated power of the energy storage system, Other costs; The net present value of the system is: ； in, Net present value, For the operating revenue in year y, The maintenance cost for year y is... The discount rate is... Design lifespan for energy storage systems, The total investment cost is expressed in ten thousand yuan. By solving the optimization model using a genetic algorithm, the most economically efficient energy storage capacity configuration scheme can be obtained.

5. A new energy cluster deviation complementary energy storage capacity demand optimization system, applying the new energy cluster deviation complementary energy storage capacity demand optimization method as described in any one of claims 1 to 4, characterized in that, include: The data acquisition and processing module collects historical actual and predicted power output data from each power station in the new energy cluster, and performs data cleaning and normalization to generate time-series power output deviation data for each power station. The probability distribution modeling module, based on the time-series output deviation data of each station, uses the kernel density estimation method to establish a probability distribution model of the output deviation of each station; The correlation modeling and joint distribution construction module analyzes the correlation between the output deviations of each station through the Copula function and establishes a joint distribution model of the deviations of each station. The scene generation and total deviation calculation module, based on the probability distribution model and the joint distribution model, uses Monte Carlo simulation to generate a set of deviation scenarios for each station and calculates the total cluster deviation under each scenario. The energy storage technology capacity requirement module determines the technical capacity requirement of the energy storage system based on the total deviation scenario set of the cluster and the preset confidence level. The optimization model construction and solution module uses the technical capacity requirements of the energy storage system as a constraint and the net present value over the entire life cycle as the objective to construct an economic optimization model and use a genetic algorithm to solve it, thereby obtaining the optimal energy storage configuration scheme.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the energy storage capacity demand optimization method for new energy cluster deviation complementarity as described in any one of claims 1 to 4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the energy storage capacity demand optimization method for new energy cluster deviation complementarity as described in any one of claims 1 to 4.

Images