Wind and light storage micro-grid capacity optimization configuration method and system

Abstract

The invention discloses a wind and light storage micro-grid capacity optimization configuration method and system, and relates to the technical field of new energy micro-grid energy management and optimization configuration, and the method comprises the steps: building a wind and light combined power data set through employing the historical output data of a wind power and photovoltaic system; performing power signal decomposition and power distribution based on the wind-solar combined power data set to obtain a power distribution result; establishing an energy storage capacity optimization model according to the power distribution result; and solving the energy storage capacity optimization model by adopting an improved whale optimization algorithm to obtain an optimal configuration result of the wind and light storage micro-grid. According to the method, the response speed and the energy utilization efficiency of an energy storage system can be improved under the multi-source fluctuation power condition, the system operation cost is reduced, the stability and the economical efficiency of micro-grid operation are enhanced, and a new technical approach is provided for optimal scheduling and planning design of a wind-light-storage integrated system.

Description

Technical Field

[0001] This invention relates to the field of energy management and optimization configuration technology for new energy microgrids, and in particular to a method and system for optimizing the capacity configuration of wind-solar-storage microgrids. Background Technology

[0002] With the increasing penetration of renewable energy in power systems, wind and solar power are widely used due to their cleanliness and renewability. However, wind speed and irradiance are significantly affected by the natural environment, resulting in obvious intermittency and fluctuations in wind and solar power output. This leads to instability in microgrid power, affecting power quality and supply reliability. To mitigate power fluctuations and improve system stability, energy storage systems are typically configured in microgrids for time-shifting and power balancing. However, existing energy storage configuration methods are mostly based on experience or single optimization strategies, failing to fully consider the multi-timescale characteristics of wind and solar power output. Furthermore, they suffer from low convergence accuracy and insufficient global optimization capabilities during capacity configuration, making it difficult to achieve a balance between economy and stability.

[0003] Therefore, there is an urgent need for a method and system for optimizing the capacity configuration of wind, solar and energy storage microgrids to address the shortcomings of existing technologies. Summary of the Invention

[0004] The purpose of this invention is to propose a method and system for optimizing the capacity configuration of a wind-solar-storage microgrid. By performing cluster analysis and signal decomposition on the characteristics of wind and solar power output, the power fluctuation characteristics at different frequencies are accurately identified, a hierarchical energy storage power allocation mechanism is established, and an improved intelligent optimization solution strategy is combined to achieve the coordinated optimization configuration of batteries and supercapacitors.

[0005] On the one hand, to achieve the above objectives, the present invention provides a method for optimizing the capacity configuration of a wind-solar-storage microgrid, comprising:

[0006] S1. Construct a combined wind and solar power dataset using historical power output data from wind and solar power systems;

[0007] S2. Based on the wind-solar combined power dataset, perform power signal decomposition and power allocation to obtain the power allocation results;

[0008] S3. Based on the power allocation results, establish an energy storage capacity optimization model;

[0009] S4. The improved whale optimization algorithm is used to solve the energy storage capacity optimization model to obtain the optimized configuration results of the wind-solar-storage microgrid.

[0010] Optionally, S1, using historical power output data from wind and solar power systems, construct a combined wind and solar power dataset, including:

[0011] Utilize a wind-solar-storage microgrid system to collect historical power output data from wind and solar power systems;

[0012] The historical output data of wind power and photovoltaic systems are imputed for missing values ​​and outliers are removed. The time step is used for resampling to obtain the power time series of wind power and photovoltaic systems.

[0013] The power time series of the wind power and photovoltaic systems are normalized to obtain a normalized power series;

[0014] Based on the normalized power sequence, the fluctuation characteristics of the normalized power sequence are extracted, and a multi-dimensional feature vector is constructed.

[0015] Clustering algorithms are used to cluster the multi-dimensional feature vectors according to operating conditions, and power time series corresponding to several typical operating conditions are obtained as a wind-solar joint power dataset.

[0016] Optionally, S2, based on the wind-solar joint power dataset, perform power signal decomposition and power allocation to obtain power allocation results, including:

[0017] Based on the aforementioned wind-solar combined power dataset, multi-scale decomposition is performed using wavelet packet transformation to obtain node components;

[0018] Based on the node components, obtain the frequency division line;

[0019] The node components are reconstructed using the frequency division lines to obtain a power signal sequence;

[0020] The power signal sequence is allocated based on the hybrid energy storage allocation rule to obtain the power allocation result.

[0021] Optionally, obtaining nodal components by performing multi-scale decomposition using wavelet packet transformation based on the wind-solar joint power dataset further includes:

[0022] Based on the node components, the energy percentage of each node component is obtained, and the energy percentage of each node component is:

[0023]

[0024] Where, η j,k E represents the energy percentage of the nodal components. j,k E represents the energy corresponding to the k-th node component of the j-th layer after wavelet packet transform. total This represents the total energy of all node components after wavelet packet transform.

[0025] Optionally, S3, based on the power allocation results, establish an energy storage capacity optimization model, including:

[0026] Using the power allocation results, the annual comprehensive economic cost of the system is obtained;

[0027] Based on the system's annual comprehensive economic cost, the minimum annual comprehensive economic cost of the system is obtained as the optimization objective function;

[0028] Based on the aforementioned objective function and the constraints of the hybrid energy storage system, an energy storage capacity optimization model is established.

[0029] Optionally, the annual comprehensive economic cost of the system includes the cost of the battery energy storage system and the cost of the supercapacitor system. The cost of the battery energy storage system includes the investment cost and operation and maintenance cost of the battery energy storage unit, and the cost of the supercapacitor system includes the investment cost and operation and maintenance cost of the supercapacitor energy storage unit.

[0030] The investment cost of the battery energy storage unit is:

[0031]

[0032] The operation and maintenance cost of the battery energy storage unit is:

[0033]

[0034] The investment cost of the supercapacitor energy storage unit is:

[0035]

[0036] The operation and maintenance cost of the supercapacitor energy storage unit is:

[0037]

[0038] in, R is the investment cost of the battery energy storage unit, r is the discount rate, β is the lifespan, and R BP P is the unit power of the battery. BESS R is the rated power of the battery. BF For the unit capacity cost of batteries, E BESS For the battery's rated capacity, Let λ represent the operation and maintenance cost of the battery energy storage unit, and λ be the battery operation and maintenance coefficient. R represents the investment cost of a supercapacitor energy storage unit. SCP For the unit power cost of supercapacitors, P SC R is the rated power of the supercapacitor. SCE E represents the unit capacity cost of a supercapacitor. SC This refers to the rated capacitance of the supercapacitor. ρ represents the operation and maintenance cost of the supercapacitor energy storage unit, and ρ is the supercapacitor operation and maintenance coefficient.

[0039] Optionally, the constraints of the hybrid energy storage system include charge-discharge power balance constraints and energy storage state constraints. The charge-discharge power balance constraints include battery charge-discharge power constraints and supercapacitor charge-discharge power constraints. The energy storage state constraints include battery energy storage state constraints and supercapacitor energy storage state constraints.

[0040] The battery charging and discharging power constraint is:

[0041]

[0042] The charging and discharging power constraint of the supercapacitor is:

[0043]

[0044] The battery energy storage state constraint is as follows:

[0045]

[0046]

[0047] The energy storage state constraint of the supercapacitor is:

[0048]

[0049]

[0050] in, This refers to the battery's discharge power. This refers to the maximum rated power of the battery energy storage unit. P is the charging power of the battery. BESS This is the battery's rated power. This represents the discharge power of the supercapacitor. This represents the maximum rated power of the supercapacitor energy storage unit. These are the charging power of the supercapacitor, P. SC E represents the rated power of the supercapacitor. BESS (t+1) represents the energy state of the battery at time t+1, E BESS (t) represents the energy state of the battery at time t, and η BESS For the charge and discharge efficiency of the battery, P BESS (t) represents the net power output of the battery at the current moment, and Δt is the time step. This represents the lower limit of battery capacity. E represents the upper limit of battery capacity. SC (t+1) represents the energy state of the supercapacitor at time t+1, E SC (t) represents the energy state of the supercapacitor at time t, η SC For the charge and discharge efficiency of a supercapacitor, PSC (t) represents the net power output of the supercapacitor at the current moment. This represents the lower limit of the capacitance of a supercapacitor. This represents the upper limit of the supercapacitor's capacity.

[0051] Optionally, S4, the improved whale optimization algorithm is used to solve the energy storage capacity optimization model to obtain the optimized configuration results of the wind-solar-storage microgrid, including:

[0052] Set the initial parameters for the improved whale optimization algorithm;

[0053] The initial parameters of the improved whale optimization algorithm are chaotically initialized using a Logistic mapping to obtain the initial population.

[0054] Using the initial population, the core parameters of the algorithm are obtained;

[0055] The initial population is initially updated based on the core parameters of the algorithm and the search mechanism to obtain the initially updated initial population.

[0056] Based on the initial population update, a spiral search update is performed using a spiral update factor to obtain the whale fitness value.

[0057] The global optimal individual is updated based on the whale's fitness value to obtain the optimized configuration result of the wind-solar-storage microgrid.

[0058] Optionally, the search mechanism includes a first search mechanism and a second search mechanism;

[0059] The first search mechanism is:

[0060]

[0061] The second search mechanism is:

[0062]

[0063] Among them, X i|(t+1) Let X be the updated position vector of the i-th individual whale in the (t+1)-th iteration. * (t) represents the position of the globally optimal individual in the current iteration, A is the control coefficient for position vector update, C is the coefficient affecting the search direction, and X... i|(t) Let X be the position vector of the i-th individual whale at the t-th iteration. rand (t) represents the position of an individual randomly selected from the population.

[0064] On the other hand, to achieve the above objectives, the present invention provides a wind-solar-storage microgrid capacity optimization configuration system, including: a data preprocessing module, a power signal decomposition and distribution module, an optimization modeling module and an algorithm solving module;

[0065] The data preprocessing module is used to construct a combined wind and solar power dataset using historical power output data from wind and solar power systems.

[0066] The power signal decomposition and allocation module is used to perform power signal decomposition and power allocation based on the wind-solar joint power dataset, and obtain the power allocation result.

[0067] The optimization modeling module is used to establish an energy storage capacity optimization model based on the power allocation results.

[0068] The algorithm solving module is used to solve the energy storage capacity optimization model using the improved whale optimization algorithm to obtain the optimized configuration results of the wind-solar-storage microgrid.

[0069] Compared with the closest existing technology, the present invention has the following advantages:

[0070] This invention proposes a method and system for optimizing the capacity configuration of wind, solar, and energy storage microgrids, achieving coordinated configuration and dynamic optimization of wind power, photovoltaic, and energy storage units. Based on typical operating condition identification, this invention extracts representative operating data, combines wavelet packet decomposition technology to separate power fluctuation characteristics in different frequency bands, and achieves reasonable power allocation between batteries and supercapacitors according to frequency division principles. Finally, an improved whale optimization algorithm is used to perform global optimization with the goal of minimizing overall economic cost. Compared with traditional methods, this invention can improve the response speed and energy utilization efficiency of energy storage systems under multi-source fluctuating power conditions, reduce system operating costs, and enhance the stability and economy of microgrid operation, providing a new technical approach for the optimized scheduling and planning design of integrated wind, solar, and energy storage systems. This invention's method can optimize the energy storage capacity allocation ratio, reduce microgrid operating costs, and improve the utilization efficiency of renewable energy and the overall economic operation of the system while ensuring system stability. Attached Figure Description

[0071] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0072] Figure 1 This is a flowchart illustrating a method for optimizing the capacity configuration of a wind-solar-storage microgrid according to an embodiment of the present invention;

[0073] Figure 2 This is a schematic diagram of the structure of the wind-solar-storage microgrid system proposed in an embodiment of the present invention;

[0074] Figure 3 This is a flowchart of the IWOA proposed in an embodiment of the present invention;

[0075] Figure 4 This is a schematic diagram of a wind-solar-storage microgrid capacity optimization configuration system according to an embodiment of the present invention. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0077] The terminology used in the embodiments section of this invention is for the purpose of explaining specific embodiments of the invention only, and is not intended to limit the invention.

[0078] like Figure 1 As shown in the figure, this invention provides a method for optimizing the capacity configuration of a wind-solar-storage microgrid, used to achieve coordinated operation and reasonable capacity configuration of wind power, photovoltaic, and energy storage systems in a microgrid. The method specifically includes the following steps:

[0079] S1. Construct a combined wind and solar power dataset using historical power output data from wind and solar power systems;

[0080] This step constructs a high-quality basic dataset through data preprocessing and operating condition clustering. First, historical power output data for wind and solar power, along with corresponding meteorological and load data, are collected. Missing value imputation, outlier removal, unified time-step resampling, and normalization are performed to eliminate data interference and dimensional differences. Then, based on time windows, multidimensional fluctuation features such as mean, standard deviation, and kurtosis of the power series are extracted. The K-means++ clustering algorithm is used, and an improved initial centroid selection strategy minimizes the intra-cluster squared error. The silhouette coefficient is used to determine the optimal number of clusters. Finally, several typical operating conditions' power time series are selected to form a joint wind and solar power dataset, providing accurate and representative data support for subsequent signal decomposition and capacity optimization.

[0081] S2. Based on the wind-solar combined power dataset, perform power signal decomposition and power allocation to obtain the power allocation results;

[0082] This step achieves precise power allocation through signal decomposition and strategy design. For the obtained typical operating condition power sequence, an appropriate wavelet basis and decomposition level are selected. Wavelet packet transform is used to perform multi-scale decomposition of the power signal, obtaining nodal components in different frequency bands and calculating their energy proportions. Frequency division lines are determined based on the center frequencies and energy distributions of each component, reconstructing low-frequency and high-frequency power signals. A hybrid energy storage power allocation strategy is then designed, with battery energy storage units handling low-frequency, large-amplitude fluctuations and supercapacitors handling high-frequency, small-amplitude fluctuation suppression, ultimately outputting a clear power allocation result.

[0083] S3. Based on the power allocation results, establish an energy storage capacity optimization model;

[0084] This step constructs an optimization model that balances economy and safety based on the power allocation results. The model's core objective is to minimize the system's annual comprehensive economic cost, which includes the investment and operation and maintenance costs of the battery and supercapacitor. The annualized conversion is completed by combining the discount rate and life cycle. The battery capacity and supercapacitor capacity are the main decision variables, while incorporating two types of constraints: first, the charge and discharge power balance constraint, which ensures that the energy storage device charges and discharges in only one direction at any given time, avoiding bidirectional power conflicts; second, the energy storage state constraint, which limits the energy state of the battery and supercapacitor to within a preset upper and lower limit range and conforms to the energy change law related to charge and discharge efficiency and time step, ultimately forming a multi-objective nonlinear optimization problem.

[0085] S4. The improved whale optimization algorithm is used to solve the energy storage capacity optimization model to obtain the optimized configuration results of the wind-solar-storage microgrid;

[0086] This step achieves efficient model solving through improved optimization algorithms. An improved whale optimization algorithm incorporating chaotic initialization and adaptive spiral update mechanisms is adopted. Chaotic initialization enhances the diversity and search space coverage of the initial population through Logistic mapping, while the adaptive spiral update mechanism balances global search and local convergence capabilities. Individuals in the population are defined with battery and supercapacitor capacities as parameters. During iteration, the individual positions are updated dynamically based on a random factor, choosing either a shrinking encirclement strategy or a random search strategy, while the spiral update factor optimizes local convergence. In each iteration, the overall system economic cost corresponding to each individual is calculated as the fitness, continuously updating the globally optimal individual. When the iteration reaches a threshold or the objective function converges, the optimal capacity configuration result for the battery and supercapacitor is output, i.e., the optimized configuration scheme for the wind-solar-storage microgrid.

[0087] In summary, steps S1 to S4 first collect historical power output and related data for wind and solar power. After preprocessing, typical operating conditions are identified using the K-means++ clustering algorithm. Then, wavelet packet transform is used to decompose the power signals of typical operating conditions into multi-scale data. Based on the energy distribution characteristics, high and low frequency signals are reconstructed, and a hybrid energy storage power allocation strategy is formulated, in which "batteries bear the large fluctuations in low frequencies, and supercapacitors bear the small fluctuations in high frequencies." Subsequently, a capacity optimization model is constructed with the goal of minimizing overall economic cost, encompassing equipment investment and operation and maintenance costs, charge and discharge power balance, and energy storage state constraints. Finally, an improved whale optimization algorithm with chaotic initialization and adaptive spiral update mechanisms is used to solve the problem and obtain the optimal capacity configuration. Compared with traditional solutions, this process effectively improves the response speed and energy utilization efficiency of the energy storage system, significantly reduces the operating cost of the microgrid, and enhances the stability and economy of system operation, providing a new technical approach for the optimized scheduling and planning design of integrated wind, solar, and energy storage systems.

[0088] As one possible implementation, in the above embodiments, step S1 may specifically include the following steps:

[0089] S1-1. Utilize a wind-solar-storage microgrid system to collect historical power output data from wind and solar power systems;

[0090] By relying on the wind-solar-storage microgrid system, historical output data of wind power and photovoltaic systems are collected comprehensively, including wind power and photovoltaic output, as well as corresponding meteorological and load data. This ensures that the data covers different operating scenarios, provides a complete and comprehensive basic data source for subsequent data analysis and processing, and lays the foundation for the construction of a wind-solar joint power dataset.

[0091] like Figure 2 As shown, the wind-solar-storage microgrid system mainly consists of wind power generation units, photovoltaic power generation units, battery energy storage units, supercapacitor energy storage units, loads, and grid connection interfaces. The power output from the wind and photovoltaic units is uniformly dispatched through the energy management system, and power balancing and fluctuation suppression are achieved through energy storage devices, thereby realizing the stable and efficient operation of the microgrid.

[0092] S1-2. Impute missing values ​​and remove outliers from the historical output data of wind power and photovoltaic systems, and resample using the time step to obtain the power time series of wind power and photovoltaic systems.

[0093] The collected wind power and photovoltaic power output, as well as corresponding meteorological and load data, are preprocessed in a targeted manner: Appropriate interpolation methods (such as linear interpolation and nearest neighbor interpolation) are used to impute missing values ​​in the data, preventing them from affecting the accuracy of subsequent analysis; outliers are identified and removed through statistical analysis and threshold judgment, such as data that deviates extremely from the normal range due to equipment failure or measurement errors, ensuring data reliability; subsequently, the processed data is resampled according to a uniform time step Δt to maintain consistency and regularity in the time dimension, ultimately forming a regular and continuous wind power and photovoltaic system power time series, providing high-quality data support for subsequent normalization processing and feature extraction.

[0094] S1-3. Normalize the power time series of the wind power and photovoltaic system to obtain a normalized power series;

[0095] To eliminate the influence of different dimensions in the power time series of wind and photovoltaic systems and ensure the accuracy of subsequent fluctuation feature extraction and cluster analysis, the power time series is normalized to obtain a normalized power series, calculated as follows:

[0096]

[0097] Wherein, P0(t) is the normalized power sequence, P(t) is the power time sequence of the wind power and photovoltaic system, min(P) is the minimum value of the power time sequence of the wind power and photovoltaic system, and max(P) is the maximum value of the power time sequence of the wind power and photovoltaic system.

[0098] By using a specific normalization calculation method, the values ​​of the original power time series are mapped to a unified range of values, and a normalized power series is finally obtained. This series can achieve horizontal comparison and fusion analysis of power data from different sources while preserving the fluctuation trend and relative relationship of the original data, thus creating conditions for the subsequent construction of multidimensional feature vectors.

[0099] S1-4. Extract the fluctuation characteristics of the normalized power sequence based on the normalized power sequence, and construct a multi-dimensional feature vector;

[0100] Fluctuation features are extracted from the normalized power sequence, and a multidimensional feature vector x is constructed using a time window w. The calculation formula is as follows:

[0101] x=[μ,σ,κ,Δ w ,skew] T

[0102] Where μ is the window mean, σ is the standard deviation, κ is the kurtosis, and Δ w is the maximum change over time w, skew is the skewness used to characterize the symmetry of fluctuations and abrupt changes, and T is the transpose.

[0103] These extracted feature parameters are integrated to construct a multi-dimensional feature vector that can comprehensively characterize the fluctuation characteristics of the normalized power sequence. This vector fully preserves the fluctuation information of wind and solar power output at different time scales, providing a core analytical basis for subsequent working condition clustering.

[0104] S1-5. Use a clustering algorithm to cluster the multi-dimensional feature vectors according to operating conditions, and obtain the power time series corresponding to several typical operating conditions as a wind-solar joint power dataset.

[0105] The K-means++ clustering algorithm is used to cluster multi-dimensional feature vectors according to their working conditions. This algorithm minimizes the intra-cluster squared error J through an improved initial centroid selection strategy, as shown in the following formula:

[0106]

[0107] Where K is the number of clusters, C i Let x be the set of samples of class i. j Let μ be the multidimensional feature vector corresponding to the j-th sample to be clustered. i These correspond to the cluster centers. To determine the optimal number of clusters K... * The silhouette coefficient is introduced to evaluate the clustering effect of different K values, and the K value that maximizes the silhouette coefficient is selected. * The final cluster number is used to obtain several typical operating conditions. The power time series corresponding to each operating condition is then input into the subsequent wavelet packet decomposition process.

[0108] In summary, steps S1-1 to S1-5 first collect historical power output data of wind and solar power based on the wind-solar-storage microgrid system. After preprocessing, the K-means++ clustering algorithm is used to cluster the multi-dimensional feature vector under different operating conditions. The optimal number of clusters is determined by the silhouette coefficient, and power time series with similar fluctuation characteristics are classified. Finally, several power time series corresponding to typical operating conditions are obtained as a joint wind-solar power dataset. This process not only effectively ensures the integrity, reliability, and consistency of the data and eliminates invalid interference information, but also simplifies the data scale and reduces subsequent computational redundancy by extracting typical operating conditions. At the same time, it accurately captures the core fluctuation patterns of wind and solar power output, providing high-quality and representative basic data support for subsequent wavelet packet transform signal decomposition, hybrid energy storage hierarchical power allocation, and capacity optimization modeling. This significantly improves the efficiency and accuracy of microgrid capacity optimization configuration and lays a solid data foundation for achieving collaborative configuration of energy storage systems and balancing system economy and stability.

[0109] As one possible implementation, in the above embodiments, step S2 may specifically include the following steps:

[0110] S2-1. Based on the wind-solar power dataset, perform multi-scale decomposition using wavelet packet transformation to obtain node components;

[0111] Wavelet Packet Transform (WPT) is used to decompose the representative power sequence under typical operating conditions into multi-scale components, which are then divided into high-frequency and low-frequency components in different frequency bands. This allows for the extraction of features in different frequency bands, analysis of allocable and adjustable energy components, and provides input for subsequent energy storage unit configuration.

[0112] First, a suitable wavelet basis and decomposition level j are selected, and the multi-scale decomposition depth is determined based on the sampling time step and the target frequency range. The power signal s(t) is decomposed into j-level wavelet packets, yielding 2... j Each node component W j,k (t), where j is the decomposition level and k is the node index.

[0113] Simultaneously through 2 j Each node component W j,k (t) Defines the energy percentage of the node components, which reflects the energy share of each frequency band in the total power signal. The energy percentage η of the node components. j,k The calculation formula is as follows:

[0114]

[0115]

[0116]

[0117] Among them, E j,k E represents the energy corresponding to the k-th node component of the j-th layer after wavelet packet transform. total t represents the total energy of all node components after wavelet packet transform, and t is the time index.

[0118] S2-2. Based on the node components, obtain the frequency division line;

[0119] Based on the obtained components of each node, the frequency division line f is determined according to the center frequency or energy distribution characteristics of each node. th This is used to distinguish between high-frequency and low-frequency components in the power signal. The low-frequency component corresponds to the slow change trend of the system, with a larger energy proportion and gentle fluctuations; the high-frequency component corresponds to short-term disturbance components, with a smaller energy proportion and rapid changes.

[0120] S2-3. Reconstruct the node components using the frequency division lines to obtain a power signal sequence;

[0121] Based on the energy distribution characteristics, the components of each node are reconstructed to obtain power signal sequences representing fluctuations at different time scales. Specifically:

[0122] Based on the defined frequency division line f th Based on this, for all node components W j,k (t) can be categorized to obtain the low-frequency signal s. L (t) and high-frequency signal s H (t), where the low-frequency signal s L (t) reflects the slow power change of the system, and the high-frequency signal s H (t) reflects the rapid fluctuation characteristics of the system.

[0123] low frequency signal s L (t) and high-frequency signal s H The reconstruction expression for (t) is:

[0124]

[0125] Wherein, set L represents the node frequency f j,k ≤f th The low-frequency portion, where set H represents the node frequency f j,k >f th The high-frequency part.

[0126] S2-4. Based on the hybrid energy storage allocation rule, perform power allocation on the power signal sequence and obtain the power allocation result;

[0127] Based on the functional division and power allocation strategy of the hybrid energy storage system, directional allocation is performed on the reconfigured high and low frequency signals: low frequency signal s L (t) corresponds to the low-frequency, high-amplitude power fluctuation portion, which is handled by the battery energy storage unit. Its core function is to achieve energy balance in the microgrid and smooth out instability caused by long-term power variations; high-frequency signal s H The high-frequency, low-amplitude power fluctuation portion corresponding to (t) is handled by the supercapacitor energy storage unit. Its main function is to quickly respond to short-term power disturbances and suppress rapid power fluctuations. Through the above division of labor and cooperation, a power allocation scheme for the coordinated operation of the battery and supercapacitor is formed, ultimately outputting a clear correspondence between "low-frequency signal - battery energy storage" and "high-frequency signal - supercapacitor" and the specific power allocation result.

[0128] In summary, steps S2-1 to S2-4 first use wavelet packet transform multi-scale decomposition based on the combined wind and solar power dataset to accurately separate node components with different frequency band characteristics. Then, based on the frequency distribution and energy proportion of the node components, a suitable frequency division line is determined. Subsequently, the node components are classified and reconstructed according to this division line to obtain a clearly layered power signal sequence. Finally, following the hybrid energy storage allocation rules, the reconstructed power signal sequence is specifically allocated, outputting the final power allocation result. This process achieves accurate decomposition of the fluctuation characteristics of the combined wind and solar power signal through wavelet packet multi-scale decomposition, ensures the reconstructed signal conforms to the fluctuation patterns of wind and solar power based on the frequency division line, and achieves precise division of labor—"batteries stabilize low frequencies, capacitors suppress high frequencies"—through hybrid energy storage targeted allocation. This fully leverages the performance advantages of different energy storage devices, effectively improves the operating efficiency of the hybrid energy storage system, reduces energy storage device losses, extends battery life, and significantly smooths out multi-scale fluctuations in combined wind and solar power, improving the stability and controllability of power output. This provides reliable technical support for efficient grid connection of wind and solar power generation and the safe and stable operation of microgrids.

[0129] As one possible implementation, in the above embodiments, step S3 may specifically include the following steps:

[0130] S3-1. Using the power allocation results, obtain the system's annual comprehensive economic cost;

[0131] Based on the power allocation results, a capacity optimization model is established with the goal of minimizing the overall economic cost. The annual overall economic cost C of the system consists of two parts: the cost of the battery energy storage system and the cost of the supercapacitor system, expressed as:

[0132]

[0133] Among them, C BESS For the cost of battery energy storage systems, C SC Cost of supercapacitor systems.

[0134] The cost of a battery energy storage system includes two parts: the investment cost of the battery energy storage unit and the operation and maintenance cost, expressed as follows:

[0135]

[0136] in, The investment cost of battery energy storage units, The operating and maintenance costs of the battery energy storage unit;

[0137] Investment cost of battery energy storage units The calculation formula is as follows:

[0138]

[0139] Operation and maintenance costs of battery energy storage units The calculation formula is as follows:

[0140]

[0141] Where r is the discount rate, β is the lifespan, and R BP P is the unit power of the battery. BESS R is the rated power of the battery. BF For the unit capacity cost of batteries, E BESS λ represents the battery's rated capacity, and λ represents the battery's maintenance coefficient.

[0142] The cost of a supercapacitor system includes two parts: the investment cost of the supercapacitor energy storage unit and the operation and maintenance cost, expressed as:

[0143]

[0144] in, The investment cost of a supercapacitor energy storage unit. The operating and maintenance costs of supercapacitor energy storage units;

[0145] Investment cost of the supercapacitor energy storage unit The calculation formula is as follows:

[0146]

[0147] The operation and maintenance cost of the supercapacitor energy storage unit The calculation formula is as follows:

[0148]

[0149] Among them, R SCP For the unit power cost of supercapacitors, P SC R is the rated power of the supercapacitor. SCE E represents the unit capacity cost of a supercapacitor. SC ρ represents the rated capacity of the supercapacitor, and ρ is the operation and maintenance coefficient of the supercapacitor.

[0150] S3-2. Based on the system's annual comprehensive economic cost, obtain the minimum annual comprehensive economic cost of the system as the optimization objective function;

[0151] The system's annual comprehensive economic cost is the sum of the costs of the battery energy storage system and the supercapacitor system. This forms the basis for an optimization objective function that minimizes the system's annual comprehensive economic cost. This objective function integrates the total lifecycle investment and operation and maintenance costs of both types of energy storage equipment. Its core is to optimize the two key decision variables—battery capacity and supercapacitor capacity—to achieve the overall optimal balance between system investment and operating costs, taking into account both the cost-effectiveness of equipment investment and long-term operational economics. This forms the core direction of nonlinear optimization involving multiple variables.

[0152] Combining the above equations, the system's objective function is:

[0153]

[0154] Wherein, min F represents the minimum annual comprehensive economic cost of the system. The objective is to minimize the annual comprehensive economic cost of the system, thereby achieving the overall optimization of system investment and operating costs.

[0155] S3-3. Based on the aforementioned objective function and the constraints of the hybrid energy storage system, establish an energy storage capacity optimization model;

[0156] To ensure the safe, stable, and economical operation of hybrid energy storage systems, the energy storage capacity optimization model must satisfy two types of conditions: charging and discharging power balance constraints and energy storage state constraints.

[0157] Charge and discharge power balance constraints include battery charge and discharge power constraints and supercapacitor charge and discharge power constraints.

[0158] The charging and discharging power constraint of the battery at any time t is the battery charging and discharging power constraint, and its calculation formula is as follows:

[0159]

[0160] in, This refers to the battery's discharge power. This refers to the maximum rated power of the battery energy storage unit. P is the charging power of the battery. BESS This is the battery's rated power. This constraint ensures that the battery is only in a charging or discharging state at any given time, without bidirectional power conflict.

[0161] The charging and discharging power constraint of a supercapacitor at any time t is called the supercapacitor charging and discharging power constraint, and its calculation formula is as follows:

[0162]

[0163] in, This represents the discharge power of the supercapacitor. This represents the maximum rated power of the supercapacitor energy storage unit. These are the charging power of the supercapacitor, P.SC This is the rated power of the supercapacitor. This constraint ensures that the supercapacitor is only in a charging or discharging state at any given time, without bidirectional power conflict.

[0164] Energy storage state constraints include battery energy storage state constraints and supercapacitor energy storage state constraints. The energy state of supercapacitors and battery energy storage units must be kept within the allowable range.

[0165] The energy state of the battery storage unit must be kept within an allowable range. The formula for calculating the battery energy storage state constraint is as follows:

[0166]

[0167]

[0168] Among them, E BESS (t+1) represents the energy state of the battery at time t+1, E BESS (t) represents the energy state of the battery at time t, and η BESS For the charge and discharge efficiency of the battery, P BESS (t) represents the net power output of the battery at the current moment, and Δt is the time step. This represents the lower limit of battery capacity. This represents the upper limit of the battery's capacity.

[0169] The energy state of a supercapacitor energy storage unit must be maintained within an allowable range. The formula for calculating the energy state constraint of a supercapacitor energy storage unit is as follows:

[0170]

[0171]

[0172] Among them, E SC (t+1) represents the energy state of the supercapacitor at time t+1, E SC (t) represents the energy state of the supercapacitor at time t, η SC For the charge and discharge efficiency of a supercapacitor, P SC (t) represents the net power output of the supercapacitor at the current moment. This represents the lower limit of the capacitance of a supercapacitor. This represents the upper limit of the supercapacitor's capacity.

[0173] With the minimum annual comprehensive economic cost of the system as the optimization objective function and the constraints of charge-discharge power balance and energy storage state as the constraints, a multi-objective nonlinear energy storage capacity optimization model that combines economy, safety and stability is formed, providing a complete mathematical framework for subsequent algorithm solutions.

[0174] In summary, steps S3-1 to S3-3 construct an energy storage capacity optimization model based on the power allocation results, with the optimal comprehensive economic cost as the core optimization objective, and clearly defining battery capacity and supercapacitor capacity as the main decision variables. During the modeling process, multiple constraints are comprehensively incorporated, including economic constraints such as equipment investment cost and full life cycle operation and maintenance cost, as well as operational constraints such as output power fluctuation constraints, power balance constraints, charging and discharging safety constraints, and capacity limit constraints, ultimately forming a multi-objective nonlinear optimization problem. This modeling approach closely integrates power allocation and capacity configuration, ensuring that capacity optimization remains aligned with actual power response requirements. It precisely manages the economic input throughout the entire lifecycle by targeting optimal cost, focusing on core optimization variables based on battery and supercapacitor capacities. Combined with multi-dimensional constraints, it guarantees the model's engineering feasibility. This approach not only fully leverages the synergistic advantages of the two types of energy storage devices to meet core functional requirements such as power regulation and fluctuation mitigation, but also achieves a scientific allocation of energy storage capacity through multi-objective nonlinear optimization. This effectively reduces economic losses and operational risks caused by over- or under-configuration, significantly improving the economy, reliability, and operational stability of hybrid energy storage systems. Furthermore, it provides a logically rigorous and practically sound quantitative analysis framework for subsequent optimization algorithms.

[0175] like Figure 3 As shown, after modeling the objective function and constraints, the improved Whale Optimization Algorithm (IWOA) is used to solve for the capacity configuration of the battery and supercapacitor. This algorithm improves upon the standard Whale Optimization Algorithm (WOA) in two ways: first, it introduces a chaotic initialization strategy to enhance the diversity of the initial population; second, it introduces an adaptive spiral update mechanism to balance the algorithm's capabilities in the global search and local convergence phases, thereby improving global optimization performance and avoiding premature convergence.

[0176] As one possible implementation, in the above embodiments, step S4 may specifically include the following steps:

[0177] S4-1. Set the initial parameters for the improved whale optimization algorithm;

[0178] To meet the practical needs of optimizing energy storage capacity in wind-solar-storage microgrids, we first set the initial parameters for the improved whale optimization algorithm, including population size, maximum number of iterations, and time step. Simultaneously, we mapped the decision variables (energy storage power and capacity) of the energy storage capacity optimization model to the positions of individual whales and defined the feasible region of these variables to prepare for subsequent solutions.

[0179] In this system, the whale population size is set to N, which corresponds to the number of candidate energy storage configuration schemes. The maximum number of iterations is used to control the time cost of the solution. The time step matches the scheduling cycle of the microgrid. Each individual represents a set of parameter vectors X to be optimized. i :

[0180] X i =[C BESS,i C CS,i ]

[0181] Among them, C BESS,i C represents the battery capacity corresponding to the i-th individual. CS,i This represents the supercapacitor capacity corresponding to the i-th individual.

[0182] S4-2. The initial parameters of the improved whale optimization algorithm are chaotically initialized using Logistic mapping to obtain the initial population;

[0183] Using the defined feasible region of the variables as the boundary, chaotic initialization is performed using the Logistic mapping to generate a chaotic sequence, calculated as follows:

[0184] μ=4

[0185] Among them, z n+1 The values ​​of the chaotic variables generated in the (n+1)th iteration are given by μ, which is the chaos control parameter, and z. n The value of the chaotic variable at the nth iteration.

[0186] The generated chaotic sequence is mapped to the parameter space, i.e., the range of values ​​for energy storage decision variables, to obtain an initial population. Each "whale individual" in the population corresponds to a set of candidate energy storage configuration schemes for a wind-solar-storage microgrid, including specific values ​​for energy storage power and capacity. Chaotic initialization can enhance the coverage of the search space and improve the uniformity of the initial solution. At the same time, it can avoid the problem of uneven distribution of schemes in traditional random initialization, allowing the initial candidate schemes to more comprehensively cover the possible solution space of energy storage configurations, and adapt to the needs of microgrid nonlinear optimization.

[0187] S4-3. Using the initial population, obtain the core parameters of the algorithm;

[0188] For the initial population obtained, the core parameters of the improved whale optimization algorithm are calculated based on the current iteration progress. These parameters include the control coefficient A for position vector updates and the coefficient C affecting the search direction. These parameters serve as the basis for subsequent selection of the search mechanism, enabling the algorithm to cover more potential optimal solutions and focus on refining high-quality solutions in solving the energy storage configuration of wind-solar-storage microgrids.

[0189] S4-4. Based on the core parameters of the algorithm and the search mechanism, the initial population is initially updated to obtain the initially updated initial population.

[0190] Position updates are performed based on two main search mechanisms optimized by whales:

[0191] First search mechanism: When the random factor |A| < 1, individuals tend to shrink the encirclement towards the optimal solution.

[0192]

[0193] Second search mechanism: When |A| ≥ 1, individuals randomly explore distant solution spaces to maintain global search.

[0194]

[0195] Among them, X i|(t+1) Let X be the updated position vector of the i-th individual whale in the (t+1)-th iteration. * (t) represents the position of the globally optimal individual in the current iteration, X i|(t) Let X be the position vector of the i-th individual whale at the t-th iteration, representing the current candidate solution. rand (t) represents the position of an individual randomly selected from the population.

[0196] S4-5. Based on the initial population update, a spiral search update is performed using a spiral update factor to obtain the whale fitness value.

[0197] Based on the initial update, a spiral update factor θ(t) is introduced for local convergence control:

[0198]

[0199] Where b is the spiral contraction coefficient and θ(t) is the spiral update factor, which is a parameter that changes dynamically with iteration and is used to balance exploration and development capabilities.

[0200] Further spiral search updates are performed on the initially updated population to refine the values ​​of candidate energy storage configurations. In each iteration, each updated set of candidate energy storage configurations is substituted into the energy storage capacity optimization model of the wind-solar-storage microgrid to calculate the corresponding objective function value F(X). i The overall economic cost of the system, which is the fitness value of the "whale individual", is used to evaluate the merits of candidate energy storage configurations.

[0201] S4-6. Update the global optimal individual based on the whale fitness value to obtain the optimized configuration result of the wind-solar-storage microgrid.

[0202] Update the globally optimal individual X based on the optimal fitness (whale fitness value).* When the number of iterations reaches a preset threshold or the objective function converges, the optimal parameters are output. and This refers to the optimal capacity configuration scheme for batteries and supercapacitors. Among them, This is the optimal capacity configuration for the battery. This represents the optimal capacity configuration for supercapacitors.

[0203] In summary, steps S4-1 to S4-6 employ an improved whale optimization algorithm to solve the capacity optimization model. By introducing chaotic initialization, an adaptive convergence factor, and a spiral search update mechanism, the algorithm's global optimization capability and convergence accuracy are enhanced. The algorithm continuously updates the optimal solution during the iteration process, ultimately obtaining the optimal capacity configuration of the battery and supercapacitor.

[0204] Further reference Figure 4 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of a wind-solar-storage microgrid capacity optimization configuration system, which is similar to... Figure 1 The method embodiments shown correspond to those described.

[0205] like Figure 4 As shown, a wind-solar-storage microgrid capacity optimization configuration system in this embodiment includes: a data preprocessing module, a power signal decomposition and distribution module, an optimization modeling module, and an algorithm solving module;

[0206] The data preprocessing module is used to construct a combined wind and solar power dataset using historical power output data from wind and solar power systems.

[0207] The core function of this module is to construct a high-quality, representative combined wind and solar power dataset based on historical power output data of wind and solar power systems, providing fundamental data support for subsequent analysis. First, historical power output data of wind and solar power systems, along with corresponding meteorological and load data for the time periods, are collected. Missing values ​​are imputed using interpolation, outlier data is removed using statistical anomaly detection, and resampling is performed at a uniform time step Δt. To eliminate the influence of different units on the analysis results, the power time series is normalized to obtain a standardized power series. Subsequently, using a preset time window w as the unit, the fluctuation characteristics of the normalized power series are extracted, including the window mean μ, standard deviation σ, kurtosis κ, and maximum variation Δ within the window. w Skewness was used to construct multi-dimensional feature vectors to comprehensively characterize the symmetry and abrupt change trends of force fluctuations. Finally, the K-means++ clustering algorithm was used to cluster the feature vectors according to working conditions. An improved initial centroid selection strategy was adopted to minimize the intra-cluster squared error, and a silhouette coefficient was introduced to evaluate the clustering effect of different cluster sizes K. The K that maximizes the silhouette coefficient was selected. *As the optimal number of clusters, the power time series corresponding to several typical operating conditions are finally output, forming a wind-solar joint power dataset.

[0208] The power signal decomposition and allocation module is used to perform power signal decomposition and power allocation based on the wind-solar joint power dataset, and obtain the power allocation result.

[0209] This module, based on a combined wind and solar power dataset, performs multi-scale decomposition of power signals and hybrid energy storage power allocation, outputting clear power allocation results. For the power sequences under various typical operating conditions obtained from the data preprocessing module, an appropriate wavelet basis and decomposition level j are first selected. Then, WPT is used to perform multi-scale decomposition of the power signal s(t), yielding 2... j Each node component W j,k (t), and calculate the energy proportion of each node component to clarify the energy share of different frequency bands in the total power signal; then, determine the frequency division line f based on the center frequency and energy distribution characteristics of each node. th The decomposed components are reconstructed into two types of signals: low-frequency power signal s L (t) (corresponding to a slow changing trend in the system, with a large energy proportion and gentle fluctuations) and high-frequency power signal s H (t) (corresponding to short-term disturbance components, with a small energy proportion and rapid changes); finally, based on the signal characteristics, a hybrid energy storage power allocation strategy is designed: the battery energy storage unit undertakes the power regulation task of the low-frequency, large-amplitude fluctuation part to achieve energy time shift and system energy balance; the supercapacitor energy storage unit undertakes the power suppression task of the high-frequency, small-amplitude fluctuation part to quickly respond to short-term power disturbances and finally output the power allocation result.

[0210] The optimization modeling module is used to establish an energy storage capacity optimization model based on the power allocation results.

[0211] This module establishes an energy storage capacity optimization model that balances economic efficiency and constraints based on power allocation results. The model's core optimization objective is to minimize the system's annual comprehensive economic cost, with decision variables set as the battery energy storage system capacity and the supercapacitor capacity. The annual comprehensive economic cost encompasses two parts: first, the battery energy storage system cost, including the investment and operation / maintenance costs of the battery energy storage units; and second, the supercapacitor system cost, including the investment and operation / maintenance costs of the supercapacitor energy storage units. Simultaneously, the model sets two types of constraints to ensure the system's safe and stable operation: first, a charge / discharge power balance constraint, limiting the battery and supercapacitor to only one state (charge or discharge) at any given time to avoid bidirectional power conflicts; and second, an energy storage state constraint, limiting the energy state of the battery and supercapacitor to within a preset upper and lower limit range at any given time, and ensuring that energy state changes conform to the correlation between charge / discharge efficiency and time step, ultimately forming a multi-objective nonlinear energy storage capacity optimization model.

[0212] The algorithm solving module is used to solve the energy storage capacity optimization model using the improved whale optimization algorithm to obtain the optimized configuration results of the wind-solar-storage microgrid.

[0213] This module employs an improved whale optimization algorithm to solve the energy storage capacity optimization model, outputting the optimal capacity configuration result for a wind-solar-storage microgrid. First, algorithm parameters are set: the whale population size is N, each individual in the population corresponds to a set of parameter vectors to be optimized, and parameters such as the maximum number of iterations and time step are also set. To enhance the diversity and search space coverage of the initial population, chaotic initialization of the population is achieved through Logistic mapping, generating a uniformly distributed initial population. During iteration, the algorithm combines two search mechanisms and an adaptive spiral update mechanism to update the population position: when the random factor |A| < 1, individuals shrink towards the current global optimal solution to improve local search accuracy; when |A| ≥ 1, individuals randomly explore distant solution spaces to maintain global search capability; simultaneously, a spiral update factor θ(t) (including the spiral shrinkage coefficient b) is introduced, dynamically adjusted with iteration to balance the algorithm's exploration and development capabilities, avoiding premature convergence. In each iteration, the objective function value corresponding to each individual is calculated, i.e., the system's annual comprehensive economic cost, and the global optimal individual is updated based on the optimal fitness. When the number of iterations reaches a preset threshold or the objective function meets the convergence condition, the globally optimal parameters are output, which is the optimal capacity configuration result of the battery and supercapacitor, thus completing the capacity optimization configuration of the wind-solar-storage microgrid.

[0214] In summary, this system focuses on the multi-timescale characteristics of wind and solar power output. Through a closed-loop architecture of "data processing - signal decomposition - model building - intelligent solution," it achieves coordinated optimization of battery and supercapacitor configuration, balancing the stability and economy of microgrid operation. Specifically, it comprises four core modules: data preprocessing, power signal decomposition and allocation, optimization modeling, and algorithm solution. Through the coordinated operation of these four modules, the system achieves fully automated processing from data preprocessing to optimal configuration. It fully considers the multi-timescale fluctuations in wind and solar power output and improves the accuracy and economy of the configuration results through an improved intelligent optimization algorithm, effectively solving the problems of insufficient global optimization capability and difficulty in balancing stability and economy in traditional configuration methods.

[0215] In this embodiment, the specific processing of a wind-solar-storage microgrid capacity optimization configuration system and its resulting technical effects can be referred to respectively. Figure 1 The relevant descriptions of steps S1, S2, S3 and S4 in the corresponding embodiments will not be repeated here.

[0216] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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.

[0217] This invention 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0218] 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.

[0219] 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.

[0220] Finally, 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 the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing the capacity configuration of a wind-solar-storage microgrid, characterized in that, include: S1. Construct a combined wind and solar power dataset using historical power output data from wind and solar power systems; S2. Based on the wind-solar combined power dataset, perform power signal decomposition and power allocation to obtain the power allocation results; S3. Based on the power allocation results, establish an energy storage capacity optimization model; S4. The improved whale optimization algorithm is used to solve the energy storage capacity optimization model to obtain the optimized configuration results of the wind-solar-storage microgrid.

2. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 1, characterized in that, S1. Using historical power output data from wind and solar power systems, construct a combined wind and solar power dataset, including: Utilize a wind-solar-storage microgrid system to collect historical power output data from wind and solar power systems; The historical output data of wind power and photovoltaic systems are imputed for missing values ​​and outliers are removed. The time step is used for resampling to obtain the power time series of wind power and photovoltaic systems. The power time series of the wind power and photovoltaic systems are normalized to obtain a normalized power series; Based on the normalized power sequence, the fluctuation characteristics of the normalized power sequence are extracted, and a multi-dimensional feature vector is constructed. Clustering algorithms are used to cluster the multi-dimensional feature vectors according to operating conditions, and power time series corresponding to several typical operating conditions are obtained as a wind-solar joint power dataset.

3. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 1, characterized in that, S2. Based on the aforementioned wind-solar combined power dataset, perform power signal decomposition and power allocation to obtain power allocation results, including: Based on the aforementioned wind-solar combined power dataset, multi-scale decomposition is performed using wavelet packet transformation to obtain node components; Based on the node components, obtain the frequency division line; The node components are reconstructed using the frequency division lines to obtain a power signal sequence; The power signal sequence is allocated based on the hybrid energy storage allocation rule to obtain the power allocation result.

4. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 3, characterized in that, Based on the aforementioned wind-solar joint power dataset, multi-scale decomposition using wavelet packet transformation is performed to obtain node components, which also includes: Based on the node components, the energy percentage of each node component is obtained, and the energy percentage of each node component is: ， Where, η j,k E represents the energy percentage of the nodal components. j,k E represents the energy corresponding to the k-th node component of the j-th layer after wavelet packet transform. total This represents the total energy of all node components after wavelet packet transform.

5. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 1, characterized in that, S3. Based on the power allocation results, establish an energy storage capacity optimization model, including: Using the power allocation results, the annual comprehensive economic cost of the system is obtained; Based on the system's annual comprehensive economic cost, the minimum annual comprehensive economic cost of the system is obtained as the optimization objective function; Based on the aforementioned objective function and the constraints of the hybrid energy storage system, an energy storage capacity optimization model is established.

6. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 5, characterized in that, The annual comprehensive economic cost of the system includes the cost of the battery energy storage system and the cost of the supercapacitor system. The cost of the battery energy storage system includes the investment cost and operation and maintenance cost of the battery energy storage unit. The cost of the supercapacitor system includes the investment cost and operation and maintenance cost of the supercapacitor energy storage unit. The investment cost of the battery energy storage unit is: ， The operation and maintenance cost of the battery energy storage unit is: ， The investment cost of the supercapacitor energy storage unit is: ， The operation and maintenance cost of the supercapacitor energy storage unit is: ， in, R is the investment cost of the battery energy storage unit, r is the discount rate, β is the lifespan, and R BP P is the unit power of the battery. BESS R is the rated power of the battery. BF For the unit capacity cost of batteries, E BESS For the battery's rated capacity, Let λ represent the operation and maintenance cost of the battery energy storage unit, and λ be the battery operation and maintenance coefficient. R represents the investment cost of a supercapacitor energy storage unit. SCP For the unit power cost of supercapacitors, P SC R is the rated power of the supercapacitor. SCE E represents the unit capacity cost of a supercapacitor. SC This refers to the rated capacitance of the supercapacitor. ρ represents the operation and maintenance cost of the supercapacitor energy storage unit, and ρ is the supercapacitor operation and maintenance coefficient.

7. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 5, characterized in that, The constraints of the hybrid energy storage system include charge and discharge power balance constraints and energy storage state constraints. The charge and discharge power balance constraints include battery charge and discharge power constraints and supercapacitor charge and discharge power constraints. The energy storage state constraints include battery energy storage state constraints and supercapacitor energy storage state constraints. The battery charging and discharging power constraint is: ， The charging and discharging power constraint of the supercapacitor is: ， The battery energy storage state constraint is as follows: ， ， The energy storage state constraint of the supercapacitor is: ， ， in, This refers to the battery's discharge power. This is the maximum rated power of the battery energy storage unit. P is the charging power of the battery. BESS This is the battery's rated power. This represents the discharge power of the supercapacitor. This represents the maximum rated power of the supercapacitor energy storage unit. These are the charging power of the supercapacitor, P. SC E represents the rated power of the supercapacitor. BESS (t+1) represents the energy state of the battery at time t+1, E BESS (t) represents the energy state of the battery at time t, and η BESS For the charge and discharge efficiency of the battery, P BESS (t) represents the net power output of the battery at the current moment, and Δt is the time step. This represents the lower limit of the battery's capacity. E represents the upper limit of battery capacity. SC (t+1) represents the energy state of the supercapacitor at time t+1, E SC (t) represents the energy state of the supercapacitor at time t, η SC For the charge and discharge efficiency of a supercapacitor, P SC (t) represents the net power output of the supercapacitor at the current moment. This represents the lower limit of the capacitance of a supercapacitor. This represents the upper limit of the supercapacitor's capacity.

8. The method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 1, characterized in that, S4. The improved whale optimization algorithm is used to solve the energy storage capacity optimization model to obtain the optimized configuration results of the wind-solar-storage microgrid, including: Set the initial parameters for the improved whale optimization algorithm; The initial parameters of the improved whale optimization algorithm are chaotically initialized using a Logistic mapping to obtain the initial population. Using the initial population, the core parameters of the algorithm are obtained; The initial population is initially updated based on the core parameters of the algorithm and the search mechanism to obtain the initially updated initial population. Based on the initial population update, a spiral search update is performed using a spiral update factor to obtain the whale fitness value. The global optimal individual is updated based on the whale's fitness value to obtain the optimized configuration result of the wind-solar-storage microgrid.

9. A method for optimizing the capacity configuration of a wind-solar-storage microgrid according to claim 8, characterized in that, The search mechanism includes a first search mechanism and a second search mechanism; The first search mechanism is: ， The second search mechanism is: ， Among them, X i|(t+1) Let X be the updated position vector of the i-th individual whale in the (t+1)-th iteration. * (t) represents the position of the globally optimal individual in the current iteration, A is the control coefficient for position vector update, C is the coefficient affecting the search direction, and X... i|(t) Let X be the position vector of the i-th individual whale at the t-th iteration. rand (t) represents the position of an individual randomly selected from the population.

10. A wind-solar-storage microgrid capacity optimization configuration system, implementing the method as described in any one of claims 1-9, characterized in that, include: Data preprocessing module, power signal decomposition and distribution module, optimization modeling module and algorithm solving module; The data preprocessing module is used to construct a combined wind and solar power dataset using historical power output data from wind and solar power systems. The power signal decomposition and allocation module is used to perform power signal decomposition and power allocation based on the wind-solar joint power dataset, and obtain the power allocation result. The optimization modeling module is used to establish an energy storage capacity optimization model based on the power allocation results. The algorithm solving module is used to solve the energy storage capacity optimization model using the improved whale optimization algorithm to obtain the optimized configuration results of the wind-solar-storage microgrid.

Images