A source network load storage double-layer capacity optimization configuration method and system

Abstract

The present application relates to the technical field of power system planning, and more particularly to a source-grid-load-storage double-layer capacity optimization configuration method and system. The present application uses Copula theory to construct a joint probability distribution model of wind power output and photovoltaic output, generates a multi-modal scenario cluster through Monte Carlo sampling, constructs a three-dimensional comprehensive complementary rate including fluctuation intensity, complementary depth and probability density, and solves the optimal wind-solar capacity ratio in the sense of probability; the upper layer is based on distribution robust optimization, constructs a fuzzy set, introduces a probabilistic new energy consumption rate constraint, optimizes the source-grid-load-storage capacity configuration under the most adverse distribution, and maximizes the economic benefits in the whole life cycle; the double layers are connected through a nested iteration feedback mechanism, the lower layer ratio search interval is corrected according to the upper layer energy storage compensation cost and power abandonment rate feedback, and the global economic benefits converge until convergence. The present application improves the economy, fluctuation suppression robustness and extreme scenario adaptability of source-grid-load-storage.

Description

Technical Field

[0001] This invention relates to the field of power system planning technology, and in particular to a method and system for optimizing the capacity allocation of a two-layer system of power source, grid, load and storage. Background Technology

[0002] The global energy transition is accelerating, with renewable energy sources such as wind and solar power becoming the core force in optimizing the energy structure. Integrated generation, grid, load, and storage (GPS) systems, as a key technological paradigm for integrating diverse energy resources, have been widely applied in energy bases, industrial parks, and other scenarios. my country's wind and solar power installed capacity continues to grow rapidly. The large-scale implementation of GPS projects not only injects a large amount of clean energy but also promotes the transformation of the power grid from "source following load" to "source interacting with load," providing crucial support for the low-carbon and intelligent upgrading of the energy system.

[0003] However, the inherent intermittency and volatility of renewable energy pose serious challenges to the safe operation and resource optimization of the power system. Solar power output is significantly affected by sunlight, exhibiting significant intraday and seasonal fluctuations, while wind power output is affected by wind speed variations, leading to short-term fluctuations and cross-period instability. Mismatches between these two and load curves can easily lead to problems such as increased curtailment rates and increased peak-shaving pressure. Fortunately, wind and solar power have different output characteristics across multiple time scales (hourly, daily, and monthly), providing a natural basis for synergistically suppressing fluctuations. Furthermore, energy storage, as a core regulation tool, is directly related to the wind-solar power ratio—an unreasonable ratio makes it difficult for energy storage to balance fluctuations and benefits, while capacity mismatch results in resource waste or insufficient suppression.

[0004] Current wind-solar capacity allocation optimization methods often employ a linear weighted approach using fluctuation coefficients across multiple time scales. Wind and solar power output fluctuates constantly: short-term fluctuations at the second level, intraday fluctuations at the daily level, and long-term seasonal fluctuations. Existing optimization methods calculate a fluctuation coefficient for each timescale fluctuation, then simply multiply it by a weight and sum them (linear weighting) to arrive at the optimal wind-solar capacity allocation. Linear weighting naturally smooths out fluctuation differences, assuming that fluctuations of different times and intensities are of equal importance. This method introduces recursive redundancy between indicators and is prone to equalization compensation. Furthermore, it ignores extreme fluctuation events, leading to a systematic deviation in the optimal wind-solar capacity allocation value, thereby reducing the accuracy of source-grid-load-storage capacity optimization. Summary of the Invention

[0005] Therefore, the technical problem to be solved by the present invention is to overcome the defects of using the linear weighted method of multi-time scale fluctuation coefficient for wind and solar capacity ratio optimization, which has recursive redundancy and is prone to equalization compensation, while ignoring extreme fluctuation events, thereby reducing the accuracy of source-grid-load-storage capacity optimization.

[0006] To address the aforementioned technical problems, this invention provides a method for optimizing the capacity configuration of a two-tiered system of source, grid, load, and storage, comprising: Based on the joint probability distribution model of wind power output and photovoltaic output, Monte Carlo sampling is used to generate wind power output and photovoltaic output at various times under different scenarios. Based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments, the fluctuation intensity component of each scenario is constructed. Based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, the complementary deep components of wind power output and photovoltaic power output at each time in each scenario are constructed. Based on the probability of each scenario appearing in the sampling results, probability density components are constructed. By fusing the probability density component, the fluctuation intensity component of each scene, and the complementarity depth component, the comprehensive complementarity rate of each scene is obtained. Based on the candidate wind and solar capacity ratio, the wind power output and photovoltaic output at each time under different scenarios are scaled up and down. The optimal wind and solar capacity ratio is solved with the goal of maximizing the sum of the comprehensive complementarity rate of each scenario multiplied by the probability of occurrence of the scenario. Based on the optimal wind and solar capacity ratio, an optimized configuration scheme for source-grid-load-storage capacity is obtained.

[0007] Preferably, the process of obtaining the joint probability distribution model of wind power output and photovoltaic power output includes: Based on the cumulative distribution functions of wind power output and photovoltaic power output, a joint probability distribution model of wind power output and photovoltaic power output is constructed using the Frank Copula function, and the parameters of the joint probability distribution model are estimated using the Kendall rank correlation coefficient.

[0008] Preferably, the process of obtaining the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output includes: Using the kernel density estimation method, the historical data of wind power output and photovoltaic power output are fitted with marginal distributions to obtain the cumulative distribution functions of wind power output and photovoltaic power output.

[0009] Preferably, the formula for the fluctuation intensity component of each scene is: , in, This represents the fluctuation intensity component of the current scene. The total number of moments. For time index, For the current scenario Wind power output at all times For the current scenario Solar power output at all times This represents the average sum of wind power output and solar power output at all times in the current scenario. It represents the absolute value.

[0010] Preferably, the method for constructing complementary depth components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, and the wind power output and photovoltaic power output at each time in each scenario, includes: Using the cumulative distribution function of photovoltaic power output, the photovoltaic power output at each time point in each scenario is mapped to the photovoltaic power output quantile, and the time when the photovoltaic power output quantile is less than a set threshold is taken as the target time. By using the cumulative distribution function of wind power output, the wind power output at each time point in each scenario is mapped to the wind power output quantile. For all target times, the corresponding wind power output quantiles are substituted into the conditional Copula function to obtain the conditional probability. The expected value of this set of conditional probabilities is then calculated to obtain the complementary depth components for each scenario.

[0011] Preferably, the formula for calculating the complementary depth component for each scene is: , in, These are complementary depth components for the current scene. Represents the mathematical expectation. This represents the mathematical expectation of the conditional probability of wind power output at each target time in the current scenario. Represents the set of target times.

[0012] Preferably, the method for obtaining an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio includes: A two-layer distributed bar optimization model is constructed, and the initial source-grid-load-storage capacity optimization configuration scheme is obtained by iterative solution. Among them, the outer layer of the two-layer split-bar optimization model takes minimizing the expected total cost of the source-grid-load-storage whole life cycle corresponding to the worst output distribution as the optimization objective, and sets wind-solar ratio coupling constraints, energy storage charging and discharging power and capacity constraints, system power balance constraints, transmission channel capacity constraints, and probabilistic new energy consumption rate constraints. Based on the source-grid-load-storage capacity optimization configuration scheme obtained from each iteration of the outer layer, the worst output distribution that maximizes the expected total cost of the source-grid-load-storage throughout its entire life cycle is searched within the distribution range defined by the fuzzy set. The initial source-grid-load-storage capacity optimization configuration schemes are scored, and the scheme with the highest score is selected as the final source-grid-load-storage capacity optimization configuration scheme.

[0013] Preferably, the method for constructing a two-layer bipolar bar optimization model and iteratively solving to obtain the initial source-grid-load-storage capacity optimization configuration scheme includes: The column and constraint generation algorithm is used to iteratively solve for the initial source-grid-load-storage capacity optimization configuration scheme.

[0014] Preferably, the process of obtaining the preset offset includes: The target comprehensive complementarity rate is the sum of the products of the comprehensive complementarity rate of each scenario and the probability of occurrence of each scenario. Using the candidate wind and solar capacity ratio as the independent variable and the target comprehensive complementarity rate as the dependent variable, a target comprehensive complementarity rate curve is constructed. A preliminary selection interval centered on the peak of the target comprehensive complementarity rate curve is defined, and the preset offset is determined by calculating the rate of change of the comprehensive complementarity rate of adjacent targets within this interval.

[0015] This invention also provides a source-grid-load-storage dual-layer capacity optimization configuration system, comprising: The Monte Carlo module is used to generate wind power output and photovoltaic output at various times under different scenarios based on the joint probability distribution model of wind power output and photovoltaic output. The fluctuation intensity component construction module is used to construct the fluctuation intensity component for each scenario based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments. The complementary deep component construction module is used to construct the complementary deep components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, as well as the wind power output and photovoltaic power output at each time in each scenario. The probability density component construction module is used to construct probability density components based on the probability of occurrence of each scenario in the sampling results. The fusion module is used to fuse the probability density component, the fluctuation intensity component of each scene, and the complementary depth component to obtain the comprehensive complementarity rate of each scene. The matching optimization module is used to scale the wind power output and photovoltaic output at various times under different scenarios based on the candidate wind and solar capacity matching ratio. The optimization objective is to maximize the sum of the comprehensive complementarity rate of each scenario multiplied by the probability of occurrence of the scenario, and solve for the optimal wind and solar capacity matching ratio. The configuration optimization module is used to obtain an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio.

[0016] Compared with the prior art, the above-described technical solution of the present invention has the following advantages: The present invention discloses a source-grid-load-storage dual-layer capacity optimization configuration method and system. Based on the joint probability distribution of wind and solar power output and Monte Carlo multi-scenario time-series data, it constructs a physically interpretable and highly targeted three-dimensional quantitative evaluation system to achieve multi-dimensional precise characterization: A fluctuation intensity component is constructed based on the deviation between the time-series fluctuation amplitude of wind and solar power output and the global mean in each scenario, which can accurately quantify the intensity of fluctuations in joint wind and solar power output at different time scales; a complementary depth component is constructed based on the edge cumulative distribution function of wind and solar power output and the time-series output matching relationship, which can quantitatively capture the dynamic complementary adaptation capability of wind and solar power increasing and decreasing, and peak offsetting, effectively identifying factors neglected by traditional methods. Extreme complementary / mismatched operating conditions; combining the discrete characteristics of power output distribution in various scenarios, information entropy is introduced to construct probability density components, which can quantify the differences in the probability of occurrence of different fluctuating scenarios and the weight of uncertainty in operating conditions, effectively characterizing the dynamic changes in wind and solar power output across multiple time scales; based on this, Monte Carlo multi-scenario sampling is combined to cover the fluctuation characteristics of all operating conditions, and the comprehensive complementarity rate of a single scenario is fused with scenario probability weighting. The optimal wind and solar capacity ratio is solved with the goal of maximizing the global expected comprehensive complementarity rate. This effectively avoids the equalization defects and omissions of extreme operating conditions caused by linear weighting, eliminates the systematic bias in the ratio solution, and significantly improves the rationality of the wind and solar capacity ratio and subsequent source... Furthermore, the configuration of source-grid-load-storage capacity generally adopts a sequential decision-making model of first determining the wind-solar ratio and then optimizing the storage-grid-load. This lack of a feedback mechanism between the two layers leads to a lack of coordination when the minimum fluctuation ratio at the lower level does not match the economically optimal ratio at the upper level, resulting in global optimization mismatch. To address this, this invention constructs a feasible interval for the optimal wind-solar ratio with the optimal capacity ratio as the center and a preset offset as the radius, preserving adjustable margins for both complementary and economic performance in the upper-level optimization. Simultaneously, it introduces Wasserstein distance to construct an uncertain fuzzy set, comprehensively characterizing the multidimensional random fluctuations of wind and solar output and load. Furthermore, it establishes a bidirectional iterative two-layer sub-Bruker optimization coordination model, achieving information interaction and feedback between the upper and lower layers through iterative closed loops of the inner and outer layers: the outer layer aims at the optimal total expected cost over the entire life cycle, coupling constraints and energy storage operation within the wind-solar ratio interval. Under constraints of power balance and renewable energy consumption probability, the source-grid-load-storage capacity configuration scheme is iteratively updated. The inner layer relies on the uncertain operating condition space of fuzzy sets to dynamically search for the worst output distribution for each round of capacity scheme updates in the outer layer and feed it back to the outer layer to continuously correct the capacity configuration strategy. Through the design of the ratio range margin and the two-layer iterative feedback mechanism, the dynamic coordination and matching of the wind-solar complementarity characteristics of the lower layer and the economic robustness of the upper layer system are realized. This completely solves the global optimization mismatch problem caused by the solidification of single-point ratios and effectively improves the global optimality and operating condition self-adaptation capability of source-grid-load-storage coordinated configuration. Attached Figure Description

[0017] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein:

[0018] Figure 1 This is a flowchart illustrating a dual-layer capacity optimization configuration method for source-grid-load-storage according to the present invention.

[0019] Figure 2 This is a trend chart of the overall complementarity rate as a function of wind and solar capacity ratio.

[0020] Figure 3 It is a characteristic diagram of new energy output that only considers wind power scenarios.

[0021] Figure 4 This is a characteristic diagram of new energy output in a scenario where the ratio of wind power to solar power is 1:1.

[0022] Figure 5 This is a characteristic diagram of new energy output in a scenario where the ratio of wind power to solar power is 2:1.

[0023] Figure 6 It is a characteristic diagram of new energy output only considering photovoltaic scenarios.

[0024] Figure 7 This is a typical daily power balance diagram under the recommended scheme. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0026] Reference Figure 1 As shown, this embodiment provides a method for optimizing the capacity configuration of a two-tiered system of source, grid, load, and storage, including: Step S1: Based on the joint probability distribution model of wind power output and photovoltaic output, Monte Carlo sampling is used to generate wind power output and photovoltaic output at various times under different scenarios; In this embodiment, preferably, the process of obtaining the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output includes: Using the kernel density estimation method, the historical data of wind power output and photovoltaic power output are fitted with marginal distributions to obtain the cumulative distribution functions of wind power output and photovoltaic power output.

[0027] In this embodiment, specifically, the historical measured output data of wind power stations and photovoltaic power stations for 8760 hours throughout the year are obtained. The data dimensions cover hourly power generation, cumulative power generation and output time distribution characteristics, ensuring that the data covers the output characteristics of wind and solar resources during both peak and off-peak periods.

[0028] Based on historical measured power output data of wind power plants and photovoltaic power plants, the wind power output sequence was analyzed. and photovoltaic output sequence Perform edge distribution fitting, where, Indicates time, This represents the total number of all historical sampling points. For the first The wind power output corresponding to each sampling time. For the first Photovoltaic output corresponding to each sampling moment.

[0029] The kernel density estimation (KDE) method does not require pre-setting the distribution type and can more flexibly capture the characteristics of the actual output distribution. The KDE estimate of the probability density function of wind power output is: , The KDE estimate of the probability density function of photovoltaic power output is: , in, It is a Gaussian function. , For bandwidth, the Silverman rule of thumb is used to determine: , , in, Let p be the kernel density estimation probability density at a wind power output of p, where p is any value of the output power. This is a Gaussian kernel function, with standardized variables as input. That is, the difference between the original data points and the estimated points divided by the bandwidth. For bandwidth, This is an estimate of the sample standard deviation.

[0030] This leads to the cumulative distribution function (CDF) of wind power output: , in, Let be the cumulative distribution function of wind power output. This represents the upper limit of the integral value of wind power generation output.

[0031] Similarly, the cumulative distribution function of photovoltaic power output can be obtained. .

[0032] In this embodiment, preferably, the process of obtaining the joint probability distribution model of wind power output and photovoltaic output includes: Based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, a joint probability distribution model of wind power output and photovoltaic power output is constructed by using the Frank Copula function. The parameters of the joint probability distribution model are estimated by using the Kendall rank correlation coefficient, and a goodness-of-fit test is performed to ensure that the model can accurately describe the correlation structure of wind and solar power output. The Frank Copula function is used to describe data where wind and solar power outputs exhibit a slight negative correlation, no significant tail dependence, and a positive correlation. Its cumulative distribution function expression (the joint probability distribution model of wind power output and solar power output) is as follows: , in, This is a joint probability distribution model of wind power output and photovoltaic power output. The parameters to be determined for the joint probability distribution model of wind power output and photovoltaic power output are as follows. , .

[0033] Frank Copula's probability density function for: , Kendall's rank correlation coefficient was used. The monotonic correlation of wind and solar power output, estimated based on sample data, is as follows: , in, This is an approximation of the Kendall rank correlation coefficient. Indicates time, for The wind power output corresponding to each sampling time. For the first Photovoltaic output corresponding to each sampling time. It is a sign function, taking the value 1 when the independent variable is greater than 0, -1 when it is less than 0, and 0 when it is equal to 0.

[0034] For Frank Copula, Kendall, and the parameters, the following functional relationship holds: , in, Kendall's rank correlation coefficient. The equation is solved numerically at time [time], and the Copula parameters are obtained through inverse kinematics. , for The estimated value.

[0035] The goodness-of-fit test is used to examine whether the selected function can accurately describe the correlation structure of wind and solar power output. , in, This is the Cramer-von Mises test statistic; the smaller the value, the better the model fit. The empirical Copula function is obtained directly from the original sample. , For indicator functions, To fit the obtained joint probability distribution model of wind power output and photovoltaic output.

[0036] The p-value of the test was calculated using the Bootstrap method.

[0037] This invention employs Monte Carlo sampling to generate multimodal power output scenario clusters covering both wet and dry years and extreme weather scenarios, i.e., wind power output and photovoltaic power output at various times under different scenarios. Specifically, based on a calibrated joint probability distribution model of wind power output and photovoltaic power output, a large number of joint output samples are generated using Monte Carlo sampling, and then reduced to a finite scenario cluster covering typical scenarios such as wet and dry years and extreme weather using the K-means clustering algorithm, with each scenario assigned a corresponding probability.

[0038] This invention constructs a comprehensive complementarity rate evaluation index, as the traditional single fluctuation coefficient cannot fully assess the complementary characteristics of wind-solar combined output. This invention proposes a three-dimensional comprehensive complementarity rate (CCR), which comprehensively evaluates the merits of wind-solar ratio schemes from three dimensions: fluctuation intensity, complementarity depth, and probability coverage.

[0039] Step S2: Based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments, construct the fluctuation intensity component for each scenario, define the normalized fluctuation intensity of wind and solar combined output, and reflect the overall output stability. The formula for the fluctuation intensity component of each scene is: , in, This represents the fluctuation intensity component of the current scene. The larger the value, the smaller the fluctuation in combined output. The total number of moments. For time index, For the current scenario Wind power output at all times For the current scenario Solar power output at all times This represents the average sum of wind power output and solar power output at all times in the current scenario. It represents the absolute value.

[0040] Step S3: Based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, construct the complementary depth component for each scenario based on the wind power output and photovoltaic power output at each time point in each scenario; the complementary depth component is based on the Copula conditional probability quantification of the wind-solar temporal complementary capability, defined as the conditional probability expectation of wind power output being higher under the condition of low photovoltaic power output.

[0041] The method for constructing complementary deep components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, and the wind power output and photovoltaic power output at each time in each scenario, includes: Using the cumulative distribution function of photovoltaic power output, the photovoltaic power output at each time point in each scenario is mapped to the photovoltaic power output quantile, and the time when the photovoltaic power output quantile is less than a set threshold is taken as the target time. By using the cumulative distribution function of wind power output, the wind power output at each time point in each scenario is mapped to the wind power output quantile. For all target times, the corresponding wind power output quantiles are substituted into the conditional Copula function to obtain the conditional probability. The expected value of this set of conditional probabilities is then calculated to obtain the complementary depth components for each scenario.

[0042] The formula for calculating the complementary depth component for each scene is: , in, These are complementary depth components for the current scene. , Represents the mathematical expectation. This represents the mathematical expectation of the conditional probability of wind power output at each target time in the current scenario. This represents the target time set, i.e., the interval where photovoltaic power output is relatively low.

[0043] The formula for calculating the conditional Copula function is: .

[0044] Step S4: Based on the probability of the information entropy of each scenario appearing in the sampling results, construct the probability density component of each scenario; Information entropy is used to characterize the uniformity of probability coverage of scene clusters, avoiding excessive reliance on a few extreme scenarios for optimization results. The formula is as follows: , in, For probability density components, The total number of scenes, For scene indexing, For the first The probability of a scenario appearing in the Monte Carlo sampling results is calculated as the ratio of the total number of samples for that scenario to the total number of samples in the Monte Carlo sampling. When all scenarios have equal probabilities of appearance in the Monte Carlo sampling results, When there is only one scene, .

[0045] Step S5: Fuse the probability density component, the fluctuation intensity component of each scene, and the complementary depth component to obtain the comprehensive complementarity rate of each scene; In this embodiment, optionally, the probability density component, the fluctuation intensity component of each scene, and the complementary depth component are fused, as shown in the following formula: , in, To achieve the overall complementarity rate in the current scenario, , , These are the weighting coefficients. .

[0046] Step S6: Based on the candidate wind and solar capacity ratio, scale the wind power output and photovoltaic output at each time under different scenarios, and take the maximum sum of the comprehensive complementarity rate of each scenario multiplied by the probability of the scenario as the optimization objective to solve for the optimal wind and solar capacity ratio. Define the ratio of wind and solar capacity ,in For wind power rated installed capacity, This refers to the rated installed capacity of photovoltaic power.

[0047] Given the overall planning scale of the landscape ,but: , For each candidate wind and solar capacity allocation In various scenarios Down-press rated capacity scaling output sequence: , in, For the first In each scenario Wind power output scaled down in real time. For the first In each scenario Photovoltaic output scaled up and down in real time For the first The maximum output of wind power generation in each scenario For the first The maximum value of photovoltaic output in each scenario.

[0048] Calculate the overall complementarity rate of each scene under the candidate wind-solar capacity ratio, and then calculate the probability-weighted expectation: , in, The target comprehensive complementarity rate corresponding to the candidate wind-solar capacity ratio r. For the first The overall complementarity rate of each scenario.

[0049] Optimal wind and solar capacity ratio : .

[0050] Step S7: Based on the optimal wind-solar capacity ratio, obtain the source-grid-load-storage capacity optimization configuration scheme.

[0051] In this embodiment, preferably, the method for obtaining an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio includes: Considering the uncertainties in project implementation, the optimal wind-solar capacity ratio is determined with the optimal ratio as the center and a preset offset as the radius. ; A fuzzy set is constructed using Wassertein distance to describe the multidimensional uncertainty of wind and solar power output and load; A two-layer bibliometric optimization model is constructed, and a column and constraint generation algorithm is used to iteratively solve for an initial source-grid-load-storage capacity optimization configuration scheme with economic robustness under uncertain environment. Among them, the outer layer takes minimizing the expected total cost of the entire life cycle of the source-grid-load-storage corresponding to the worst output distribution as the optimization objective, and sets constraints on wind and solar power ratio coupling, energy storage charging and discharging power and capacity, system power balance, transmission channel capacity, and probabilistic new energy consumption rate. Based on the source-grid-load-storage capacity optimization configuration scheme obtained from each iteration of the outer layer, the worst output distribution that maximizes the expected total cost of the source-grid-load-storage throughout its entire life cycle is searched within the distribution range defined by the fuzzy set. The initial source-grid-load-storage capacity optimization configuration schemes are scored, and the scheme with the highest score is selected as the final source-grid-load-storage capacity optimization configuration scheme.

[0052] In this embodiment, preferably, the process of obtaining the preset offset includes: The target comprehensive complementarity rate is the sum of the products of the comprehensive complementarity rate of each scenario and the probability of occurrence of each scenario. Using the candidate wind and solar capacity ratio as the independent variable and the target comprehensive complementarity rate as the dependent variable, a target comprehensive complementarity rate curve is constructed. A preliminary selection interval centered on the peak of the target comprehensive complementarity rate curve is defined, and the preset offset is determined by calculating the rate of change of the comprehensive complementarity rate of adjacent targets within this interval.

[0053] Preset offset The flatness of the target comprehensive complementarity rate curve near its maximum value can be determined (e.g., selecting...). (The range of proportions).

[0054] The optimal ratio range As input constraints for higher-level optimization, based on this, and combined with the joint probability distribution model of wind power output and photovoltaic output, a data-driven fuzzy set describing the multidimensional uncertainties of wind power, photovoltaic output, and load is constructed using Wassertein distance. Describing uncertain parameters The probability distribution, ,in, For empirical distribution, For Wasserstein distance, For radius, Let be the distribution space formed by all probability distributions. To contribute to wind power generation, Contribute to photovoltaic power This represents the random power of the load.

[0055] Decision variables of a two-level split-bar optimization model This includes the rated installed capacity of wind power and photovoltaic power (constrained by the lower-level allocation range), the rated capacity and maximum charging and discharging power of energy storage, the grid connection capacity, and the flexible load regulation capability. The optimization objective is to minimize the expected total cost of the entire life cycle of source-grid-load-storage corresponding to the worst-case output distribution. , in, It represents the expected total cost over the entire lifecycle of the power generation, grid, load, and storage system, including investment costs, operation and maintenance costs, wind and solar curtailment losses, electricity purchase costs, and minus carbon trading revenue.

[0056] Wind-solar ratio coupling constraints: , in, The optimal wind and solar capacity ratio is obtained from the lower-level optimization solution. This is the preset offset.

[0057] Energy storage operation constraints: , , , in, This refers to the upper limit of the rated maximum charge and discharge power of the energy storage. for Real-time energy storage capacity, for Energy storage charge at any given time for Energy storage charge at any given time For energy storage charging efficiency, For energy storage and discharge efficiency, for Real-time energy storage and charging power, for Energy storage and discharge power at all times For time step, This refers to the rated total capacity of energy storage.

[0058] Power balance and transmission constraints: , , in, for Wind power output at all times for Photovoltaic power output at all times for Energy storage and discharge power at all times For grid interaction power, for Total electrical load power at all times for Real-time energy storage and charging power, This refers to the grid connection capacity.

[0059] Probabilistic constraints on renewable energy consumption rate (confidence level) ): , in, For probability operators, for Constantly curtailing wind and solar power to reduce output. To preset a minimum renewable energy consumption rate threshold, To allow for the probability of risk that the requirements for waste disposal will not be met.

[0060] Traditional sequential decision-making, which proceeds from the bottom up, suffers from inconsistencies between the two levels of objectives: the bottom level seeks to minimize fluctuation coefficients, while the top level focuses on economic benefits and overall optimization, resulting in a lack of coordination between the two. This invention proposes an interactive, nested, iterative feedback mechanism that allows the top level to adjust the wind-solar ratio search range of the bottom level based on economic evaluation results, forming a closed-loop optimization process. The overall process is as follows: Initialization parameters: Pre-set the global search range for candidate landscape capacity ratio. Step length Convergence threshold Maximum number of iterations ; Lower-level landscape ratio optimization: Call the optimal landscape capacity ratio solution process, traverse the landscape capacity ratio interval to complete the optimization, output the optimal landscape capacity ratio, and generate the optimal ratio interval matching the optimal landscape capacity ratio based on the preset offset. Upper-layer source-grid-load-storage capacity optimization: optimizing the optimal ratio range of the lower-layer output. As a hard constraint coupled with wind and solar installed capacity, a two-layer distributed bar optimization process is invoked to solve for the source-grid-load-storage capacity configuration scheme and key evaluation indicators such as system economy and absorption level under the current iteration; among them; Indicates the first The optimal wind and solar capacity ratio in the next iteration. Indicates the first The preset offset for the next iteration.

[0061] Feasibility and convergence assessment: If the energy storage compensation cost corresponding to the initial source-grid-load-storage capacity configuration scheme under the current iteration exceeds the preset threshold or the curtailment rate is higher than the lower limit of the new energy consumption rate, mark the current wind-solar ratio as an infeasible ratio and narrow the search range for the wind-solar ratio; compare the difference of the objective function between two consecutive iterations with the preset convergence threshold. If the convergence condition is not met, conduct a local densification search on the current ratio interval and return to repeat the iteration; until the relative change in economic benefits between two consecutive iterations is less than the convergence threshold and all operational constraints and indicator constraints are met, output the optimized configuration scheme of the initial source-grid-load-storage capacity.

[0062] The qualified initial candidate source-grid-load-storage capacity optimization configuration schemes that have passed verification are comprehensively evaluated from multiple dimensions. The evaluation dimensions include system economics (return on investment, cost per kilowatt-hour), volatility suppression effect (reduction in the combined output volatility coefficient), environmental benefits (carbon emission reduction), and operational reliability (load shortage rate, wind and solar curtailment rate). The weights of each evaluation indicator are determined by the analytic hierarchy process or the entropy weight method, and a comprehensive score is calculated. The scheme with the highest comprehensive score is selected as the final source-grid-load-storage capacity optimization configuration scheme.

[0063] In this embodiment, the weights of the three dimensions of the comprehensive complementarity rate can be adjusted according to the system's emphasis on different performance aspects; the radius of the Wasserstein fuzzy set can be determined through cross-validation or by setting a confidence level; and the convergence threshold is set according to engineering accuracy requirements.

[0064] To address the shortcomings of existing linear weighted methods for fluctuation coefficients, such as redundant indicator information, neglect of extreme fluctuations, and lack of feedback between the two layers leading to optimization mismatch, this invention proposes a new approach. The lower layer uses Copula theory to establish a joint probability distribution model of wind and solar power output, characterizing the correlation structure with the Kendall rank correlation coefficient. Based on a multimodal scenario cluster, a comprehensive complementarity rate is defined, incorporating fluctuation intensity, complementarity depth, and probability density, to solve for the probabilistically optimal wind-solar capacity ratio. The upper layer constructs a source-grid-load-storage capacity configuration model based on bibloc robust optimization, introducing conditional risk value constraints and probabilistic absorption rate constraints. It identifies the worst-case operating scenario within a data-driven fuzzy set and performs robust optimization with the goal of achieving full life-cycle economic benefits. Information exchange between the upper and lower layers is achieved through a nested iterative feedback mechanism: the energy storage compensation cost and curtailment rate calculated by the upper layer are fed back to the lower layer to correct the wind-solar capacity ratio search interval until convergence. This invention overcomes the limitations of traditional sequential decision-making, significantly improving system economy, fluctuation suppression robustness, and adaptability to extreme scenarios.

[0065] This invention establishes a two-layer coupling and iterative feedback mechanism, creating a two-layer architecture of lower-layer Copula probabilistic modeling and upper-layer sub-Bruker optimization. Information interaction between the upper and lower layers is achieved through a nested iterative feedback mechanism, solving the mismatch between "fluctuation suppression" and "economic benefits" objectives in traditional sequential decision-making. It provides a breakthrough wind-solar allocation method: introducing Copula theory to construct a joint wind-solar probability distribution, using Kendall's rank correlation coefficient to characterize correlation, and defining a three-dimensional comprehensive complementarity rate (fluctuation intensity, complementarity depth, probability density), overcoming the redundancy of indicators and the neglect of extreme events in the traditional linear weighted method of fluctuation coefficients. It achieves robust upper-layer optimization: constructing a data-driven fuzzy set based on Wasserstein distance, using sub-Bruker optimization to minimize the expected cost over the entire lifecycle under the worst-case distribution, and introducing a probabilistic absorption rate constraint, improving the system's economy and reliability under strong wind-solar uncertainty. It possesses universality and practicality, with a clear methodology and adjustable parameters, flexibly adaptable to source-grid-load-storage system planning of different scales and scenarios, providing a reference technical path for engineering practice.

[0066] Based on Example 1, Example 2 collects 8760 hours of measured power output data from wind and photovoltaic power plants throughout the year, and clarifies the system load scale, load curve characteristics, and planning boundary conditions of the source-grid-load-storage system (including total installed capacity constraints, safe operation standards, etc.).

[0067] The present invention is implemented in the following scenarios, as shown in Table 1, which contains the system boundary conditions and index requirements.

[0068] Table 1

[0069] Based on the Copula comprehensive complementarity optimization, the lower-level wind-solar synergistic optimization configuration is optimized. Based on the data in Table 1, kernel density estimation is used to fit the edge distribution of wind power and photovoltaic output.

[0070] The Frank Copula function was used to construct a joint probability distribution model of wind power output and photovoltaic output. The Copula parameters were estimated using the Kendall rank correlation coefficient. The test showed that the p-value was >0.05, indicating a good fit.

[0071] Based on the joint probability distribution model of wind power output and photovoltaic output, 10,000 joint output samples were generated, and K-means clustering was used to reduce them into 20 typical scenario clusters, covering wet and dry years and extreme weather.

[0072] Construct a three-dimensional comprehensive complementarity rate (fluctuation intensity, complementarity depth, probability density), with the optimization objective being to maximize the sum of the comprehensive complementarity rates of each scenario multiplied by the probability of occurrence of the scenario. Iterate through the wind and light ratio schemes (range 1:5~7:1) and calculate the expected value of the comprehensive complementarity rate of each candidate wind and light capacity ratio under multiple scenario clusters.

[0073] The optimal wind-solar capacity ratio is 2:1, resulting in the highest overall complementarity rate.

[0074] Calculate the overall complementarity rate under 10 different landscape and light landscape scenarios, and statistically analyze the fluctuation patterns. Figure 2 , Figure 2 The overall complementarity rate trend chart with wind and solar capacity ratio is shown below. Figure 2 It can be seen that when the wind-solar ratio is low and the solar ratio is high, the overall complementarity rate is relatively low. As the wind-solar ratio increases, the overall complementarity rate gradually increases. The overall complementarity rate is the highest when the wind-solar ratio is 2:1. Subsequently, the overall complementarity rate decreases as the wind-solar ratio increases.

[0075] This embodiment employs the bibliometric optimization method proposed in this invention for upper-layer capacity configuration. Using a defined optimal wind-solar ratio range (e.g., around 2:1) as a constraint, a data-driven fuzzy set based on Wasserstein distance is constructed to describe the uncertainties in wind and solar power output and load. The goal is to maximize the economic benefits throughout the entire lifecycle, with a probabilistic constraint on renewable energy absorption rate. A column-constraint generation algorithm is used to solve the problem, resulting in a source-grid-load-storage capacity configuration scheme.

[0076] Four wind-solar configuration schemes were developed to verify the effectiveness of the proposed model. Scheme 1: Wind power only; Scheme 2: Wind power to solar power ratio of 1:1; Scheme 3: Wind power to solar power ratio of 2:1; Scheme 4: Solar power only.

[0077] See the characteristic diagrams of new energy power output under various schemes. Figure 3 , Figure 4 , Figure 5 and Figure 6 , Figure 3 It only considers the output characteristics of new energy sources in the wind power scenario. Figure 4 This is a characteristic graph of renewable energy output in a scenario where the wind power to solar power ratio is 1:1. Figure 5 This is a characteristic graph of renewable energy output in a scenario where the wind power to solar power ratio is 2:1. Figure 6 It is a characteristic diagram of new energy output only considering photovoltaic scenarios.

[0078] Option 1 and Option 4 have large power output fluctuations. Option 2 and Option 3 have complementary wind and solar power output characteristics, which reduces the fluctuation characteristics. However, compared with Option 3, Option 2 has greater overall external fluctuations.

[0079] The above four different wind-solar ratio schemes were used as inputs to the upper-level optimization model for calculation, and the calculation results are shown in Table 2. Table 2 shows the capacity optimization configuration scale and production simulation of different schemes.

[0080] Table 2

[0081] Option 4 does not meet the requirements of all indicators. Among the remaining options, Option 3 (wind-solar ratio of 2:1) has the best performance in terms of new energy consumption, carbon emission reduction, reliability, and green electricity ratio, which verifies the effectiveness of the proposed model.

[0082] The optimal wind power capacity ratio (2:1) output in this example meets the requirements for green electricity ratio, new energy consumption and other indicators after upper-level optimization.

[0083] Multi-dimensional comprehensive evaluation and optimal solution output: From four dimensions—economic efficiency, volatility, environmental benefits, and operational reliability—the scheme with the highest comprehensive indicators is selected as the final optimized configuration scheme for source-grid-load-storage capacity. Scheme 3 (wind-solar ratio 2:1) is optimal in terms of renewable energy absorption rate (89.28%), carbon emission reduction (2.7744 million tons), and green electricity ratio (66.65%), verifying the effectiveness of the method proposed in this invention.

[0084] like Figure 7 As shown, Figure 7 This is a typical daily power balance diagram under the recommended scheme. As shown in the figure, under the upper-level source-grid-load-storage optimization, energy storage charges when wind and solar output exceeds the load requirement and discharges when wind and solar output is insufficient, and ensures the reliability of power supply through external power input.

[0085] This third embodiment provides a source-grid-load-storage dual-layer capacity optimization configuration system, including: The Monte Carlo module is used to generate wind power output and photovoltaic output at various times under different scenarios based on the joint probability distribution model of wind power output and photovoltaic output. The fluctuation intensity component construction module is used to construct the fluctuation intensity component for each scenario based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments. The complementary deep component construction module is used to construct the complementary deep components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, as well as the wind power output and photovoltaic power output at each time in each scenario. The probability density component construction module is used to construct probability density components based on the probability of occurrence of each scenario in the sampling results. The fusion module is used to fuse the probability density component, the fluctuation intensity component of each scenario, and the complementarity depth component to obtain the comprehensive complementarity rate of each scenario; the ratio optimization module is used to scale the wind power output and photovoltaic output at each time under different scenarios according to the candidate wind and solar capacity ratio, and to solve for the optimal wind and solar capacity ratio by maximizing the sum of the comprehensive complementarity rate of each scenario multiplied by the probability of occurrence of the scenario. The configuration optimization module is used to obtain an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio.

[0086] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for optimizing the capacity allocation of a two-tiered system of source, grid, load, and storage, characterized in that, include: Based on the joint probability distribution model of wind power output and photovoltaic output, Monte Carlo sampling is used to generate wind power output and photovoltaic output at various times under different scenarios. Based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments, the fluctuation intensity component of each scenario is constructed. Based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, the complementary deep components of wind power output and photovoltaic power output at each time in each scenario are constructed. Based on the probability of each scenario appearing in the sampling results, probability density components are constructed. By fusing the probability density component, the fluctuation intensity component of each scene, and the complementarity depth component, the comprehensive complementarity rate of each scene is obtained. Based on the candidate wind and solar capacity ratio, the wind power output and photovoltaic output at each time under different scenarios are scaled up and down. The optimal wind and solar capacity ratio is solved with the goal of maximizing the sum of the comprehensive complementarity rate of each scenario multiplied by the probability of occurrence of the scenario. Based on the optimal wind and solar capacity ratio, an optimized configuration scheme for source-grid-load-storage capacity is obtained.

2. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 1, characterized in that, The process of obtaining the joint probability distribution model of wind power output and photovoltaic power output includes: Based on the cumulative distribution functions of wind power output and photovoltaic power output, a joint probability distribution model of wind power output and photovoltaic power output is constructed using the Frank Copula function, and the parameters of the joint probability distribution model are estimated using the Kendall rank correlation coefficient.

3. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 1, characterized in that, The process of obtaining the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output includes: Using the kernel density estimation method, the historical data of wind power output and photovoltaic power output are fitted with marginal distributions to obtain the cumulative distribution functions of wind power output and photovoltaic power output.

4. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 1, characterized in that, The formula for the fluctuation intensity component of each scene is: ， in, This represents the fluctuation intensity component of the current scene. The total number of moments. For time index, For the current scenario Wind power output at all times For the current scenario Solar power output at all times This represents the average sum of wind power output and solar power output at all times in the current scenario. It represents the absolute value.

5. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 1, characterized in that, The method for constructing complementary deep components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, and the wind power output and photovoltaic power output at each time in each scenario, includes: Using the cumulative distribution function of photovoltaic power output, the photovoltaic power output at each time point in each scenario is mapped to the photovoltaic power output quantile, and the time when the photovoltaic power output quantile is less than a set threshold is taken as the target time. By using the cumulative distribution function of wind power output, the wind power output at each time point in each scenario is mapped to the wind power output quantile. For all target times, the corresponding wind power output quantiles are substituted into the conditional Copula function to obtain the conditional probability. The expected value of this set of conditional probabilities is then calculated to obtain the complementary depth components for each scenario.

6. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 5, characterized in that, The formula for calculating the complementary depth component for each scene is: ， in, These are complementary depth components for the current scene. Represents the mathematical expectation. This represents the mathematical expectation of the conditional probability of wind power output at each target time in the current scenario. Represents the set of target times.

7. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 1, characterized in that, The method for obtaining an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio includes: Using the optimal wind and solar capacity ratio as the center and the preset offset as the radius, the optimal ratio range is determined; Fuzzy sets are constructed using Wassertein distance; A two-layer distributed bar optimization model is constructed, and the initial source-grid-load-storage capacity optimization configuration scheme is obtained by iterative solution. Among them, the outer layer of the two-layer split-bar optimization model takes minimizing the expected total cost of the source-grid-load-storage whole life cycle corresponding to the worst output distribution as the optimization objective, and sets wind-solar ratio coupling constraints, energy storage charging and discharging power and capacity constraints, system power balance constraints, transmission channel capacity constraints, and probabilistic new energy consumption rate constraints. Based on the source-grid-load-storage capacity optimization configuration scheme obtained from each iteration of the outer layer, the worst output distribution that maximizes the expected total cost of the source-grid-load-storage throughout its entire life cycle is searched within the distribution range defined by the fuzzy set. The initial source-grid-load-storage capacity optimization configuration schemes are scored, and the scheme with the highest score is selected as the final source-grid-load-storage capacity optimization configuration scheme.

8. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 7, characterized in that, The methods for constructing a two-layer distributed bar optimization model and iteratively solving it to obtain the initial source-grid-load-storage capacity optimization configuration scheme include: The column and constraint generation algorithm is used to iteratively solve for the initial source-grid-load-storage capacity optimization configuration scheme.

9. The source-grid-load-storage dual-layer capacity optimization configuration method according to claim 7, characterized in that, The process of obtaining the preset offset includes: The target comprehensive complementarity rate is the sum of the products of the comprehensive complementarity rate of each scenario and the probability of occurrence of each scenario. Using the candidate wind and solar capacity ratio as the independent variable and the target comprehensive complementarity rate as the dependent variable, a target comprehensive complementarity rate curve is constructed. A preliminary selection interval centered on the peak of the target comprehensive complementarity rate curve is defined, and the preset offset is determined by calculating the rate of change of the comprehensive complementarity rate of adjacent targets within this interval.

10. A dual-layer capacity optimization configuration system for source-grid-load-storage, characterized in that, include: The Monte Carlo module is used to generate wind power output and photovoltaic output at various times under different scenarios based on the joint probability distribution model of wind power output and photovoltaic output. The fluctuation intensity component construction module is used to construct the fluctuation intensity component for each scenario based on the sum of wind power output and photovoltaic output at each moment in each scenario, and the average of the sum of wind power output and photovoltaic output at all moments. The complementary deep component construction module is used to construct the complementary deep components for each scenario based on the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic power output, as well as the wind power output and photovoltaic power output at each time in each scenario. The probability density component construction module is used to construct probability density components based on the probability of occurrence of each scenario in the sampling results. The fusion module is used to fuse the probability density component, the fluctuation intensity component of each scene, and the complementary depth component to obtain the comprehensive complementarity rate of each scene. The matching optimization module is used to scale the wind power output and photovoltaic output at various times under different scenarios based on the candidate wind and solar capacity matching ratio. The optimization objective is to maximize the sum of the comprehensive complementarity rate of each scenario multiplied by the probability of occurrence of the scenario, and solve for the optimal wind and solar capacity matching ratio. The configuration optimization module is used to obtain an optimized configuration scheme for source-grid-load-storage capacity based on the optimal wind-solar capacity ratio.

Images