WOA-VMD-based capacity optimization method for deep and far sea wind power combined hybrid energy storage power generation system

Abstract

The invention discloses a WOA-based capacity optimization method for a deep and far sea wind power combined hybrid energy storage power generation system, and the method comprises the steps: obtaining grid-connected power, and taking the unbalanced power between the grid-connected power and wind power as reference power; a WOA algorithm is utilized to optimize by taking a sample entropy SE as a fitness function to obtain an optimal parameter combination, the SE is an index for measuring the confusion degree of the time sequence, the value of the SE is inversely proportional to effective information contained in the sequence, the SE is substituted into a specified algorithm to decompose reference power, and power distribution is carried out; and establishing an evaluation model, and carrying out joint optimization solution on the power distribution and cost model in the hybrid energy storage system by using a WOA algorithm, thereby determining a high and low frequency optimal demarcation point of a hybrid energy storage power instruction and a corresponding configuration scheme. According to the method, the parameters are adaptively determined through the whale optimization algorithm, cooperative division of labor of the storage battery and the super capacitor is achieved, and the economical efficiency and reliability of the energy storage system are optimal while the requirement for wind power grid connection stability is met.

Description

Technical Field

[0001] This invention belongs to the field of new energy power generation and energy storage optimization configuration, specifically involving a capacity optimization method for a deep-sea wind power combined with hybrid energy storage power generation system based on WOA-VMD. Background Technology

[0002] Driven by the dual-carbon development goals and the pressure on grid stability brought about by large-scale wind power grid integration, energy storage technology has developed rapidly. Due to the high energy density of batteries and the high power density and fast response characteristics of supercapacitors, hybrid energy storage systems composed of these two technologies have been widely used in wind-storage combined power generation. Currently, how to achieve the economic viability of energy storage systems while ensuring the safe operation of the power grid has become a research hotspot in this field.

[0003] However, existing energy storage capacity configuration methods suffer from numerous problems, including reliance on experience in parameter selection, insufficient accuracy of decomposition results, and imperfect economic optimization. Therefore, a new technical solution is needed to address these issues. Summary of the Invention

[0004] Purpose of the invention: To overcome the problems of existing energy storage capacity configuration methods, such as parameter selection relying on experience, insufficient accuracy of decomposition results, and imperfect economic optimization, this invention provides a capacity optimization method for deep-sea wind power combined with hybrid energy storage power generation system based on WOA-VMD. This method adaptively determines VMD parameters through the whale optimization algorithm, combines adaptive moving average and power decomposition to achieve collaborative division of labor between batteries and supercapacitors, and uses the total life cycle cost as the objective function. Thus, while meeting the stability requirements of wind power grid connection, it achieves optimal economy and reliability of the energy storage system.

[0005] Technical Solution: To achieve the above objectives, this invention provides a capacity optimization method for a deep-sea wind power combined with hybrid energy storage power generation system based on WOA-VMD, comprising the following steps:

[0006] S1: Determine the reference power based on the grid-connected power;

[0007] S2: The variational mode decomposition algorithm is optimized using the WOA algorithm with sample entropy SE as the fitness function to obtain the optimal parameter combination. SE is an indicator of the disorder of the time series.

[0008] S3: Based on the optimal parameter combination, the WOA algorithm is used to solve the evaluation model established based on LCC theory to determine the system capacity optimization scheme.

[0009] Further, step S1 specifically includes:

[0010] The moving average method involves selecting a time window width L, averaging the data within L, and replacing the center data with the average. This window moves from front to back with the time series; for each new sampled data point added, one older data point is removed from the front of the window. Its expression is:

[0011]

[0012] In the formula, Let h be the value of the h-th input data; This represents the value of the h-th filtered output data.

[0013]

[0014]

[0015] In the formula: This represents the actual power output of the wind power. This refers to the grid-connected power of wind power. This represents the total power of HESS. and These are the power ratings of the storage battery and the supercapacitor, respectively.

[0016] Furthermore, in step S2, WOA-VMD optimizes VMD parameters based on WOA, and the specific process is as follows:

[0017] A1: Set the initial parameters for WOA;

[0018] A2: Randomly initialize the location of the whale population - i.e., the VMD parameter combination, and calculate the prey objective function value for each individual whale;

[0019] A3: Determine the current optimal whale position based on the fitness value;

[0020] A4: The whale's position is updated according to WOA's two main update mechanisms (shrinking encirclement and bubble web attack);

[0021] A5: The updated whale position is used to recalculate the objective function value and compare it with the original value, retaining the better solution;

[0022] A6: If the termination condition is met (maximum number of iterations or error threshold is reached), output the optimal solution; otherwise, return to step A3.

[0023] Furthermore, in step S2, the WOA algorithm is used to optimize VMD with sample entropy SE as the fitness function to obtain the optimal parameter combination [K, α], thereby improving the VMD decomposition effect; SE is an indicator of the disorder of a time series, and its value is inversely proportional to the effective information contained in the series; for a time series with a finite number of data points N... SE is represented as:

[0024]

[0025] In the formula, q is the dimension; r is the similarity tolerance; , Let q and q+1 be the probabilities of matching points in the lower region r, respectively;

[0026] In the algorithm, the position of each whale represents a feasible solution in the solution space, and the whale finds the optimal solution by adjusting its position; its position is updated by the following formula;

[0027]

[0028] In the formula: n is the number of iterations, The position vector of the optimal solution; This is the current position vector; Let be the position vector at iteration number n+1; D simulates the distance between the target and the target; A and c are coefficients.

[0029] Let p be a random number, p ,when When p > 0.5, the whale is in the development stage of a bubble web attack, and its position is updated by the following formula;

[0030]

[0031] In the formula, b is the helical constant; l is Random numbers;

[0032] when If p < 1 and p < 0.5, the whale is in a shrinking enclosure during the development phase, and its position can be updated using the following formula;

[0033]

[0034] when At position 1, the whale is in the exploration phase and updates its position using the following formula;

[0035]

[0036] Compared to traditional VMD methods, this application uses the WOA algorithm to optimize VMD parameters, overcoming the subjectivity of parameter selection and providing a good foundation for hybrid energy storage power allocation.

[0037] Furthermore, in step S2, the WOA-VMD algorithm is used to calculate the total power of the HESS. The system is decomposed into K IMFs, and reconstructed into high- and low-frequency power commands by selecting a high- and low-frequency boundary point f. Based on the characteristics of hybrid energy storage, the supercapacitor handles high-frequency power, and the battery handles low-frequency power, i.e.:

[0038]

[0039] Considering the energy storage charging and discharging efficiency, the rated power configuration of the energy storage should be as follows:

[0040]

[0041] In the formula, This refers to the rated power of the battery. Rated power of supercapacitors; The charging power of the battery; The charging power of the supercapacitor.

[0042] Furthermore, the establishment of the energy storage economic evaluation model in step S3 includes:

[0043] An economic evaluation model based on LCC theory is established with the objective function of minimizing the annual comprehensive cost of the energy storage system, namely:

[0044]

[0045] In the formula: The investment cost of a hybrid energy storage system; For operation and maintenance costs; Cost of battery replacement; To recover residual value costs.

[0046] Investment costs The expression is as follows:

[0047]

[0048] In the formula: , These are the investment costs for batteries and supercapacitors, respectively. , , , These represent the investment cost per unit of power and capacity of a battery and a supercapacitor, respectively. The discount rate; , These are the rated capacities of the storage battery and the supercapacitor, respectively. This refers to the operating life of a supercapacitor.

[0049] Operation and maintenance costs The expression is as follows:

[0050]

[0051] In the formula: , These are the operating and maintenance costs for batteries and supercapacitors, respectively. , , , These are the power and capacity operation and maintenance costs per unit of battery and supercapacitor, respectively.

[0052] Recovery of residual value cost The expression is as follows:

[0053]

[0054] In the formula: The recovery residual value rate is generally taken as 3% to 5%.

[0055] Furthermore, the constraints of the energy storage economic evaluation model in step S3 are expressed as follows:

[0056] Power balance constraints:

[0057]

[0058] Energy storage charging and discharging power constraints:

[0059]

[0060] Energy storage SOC constraints:

[0061]

[0062] In the formula, This is the lower limit of the battery's state of charge (SOC). This represents the upper limit of the battery's state of charge (SOC). This represents the lower limit of the state of charge (SOC) of a supercapacitor. This represents the upper limit of the SOC (State of Charge) of a supercapacitor.

[0063] The method of this invention optimizes the number of modes and the second-order penalty factor of variational mode decomposition (VMD) using the Whale Optimization Algorithm (WOA), and combines it with the adaptive moving average algorithm to smooth and decompose wind power, thereby constructing a hybrid energy storage capacity optimization model. With the goal of minimizing the system's annual comprehensive cost, the method obtains the most economically optimal energy storage capacity configuration scheme.

[0064] The method of this invention can be summarized into the following steps:

[0065] The first step is to use an adaptive moving average algorithm to obtain the grid-connected wind power and hybrid energy storage reference power that meet the national wind power grid connection standards.

[0066] The second step involves using the whale optimization algorithm to optimize the modality number K and the quadratic penalty factor α in VMD, with the sample entropy used as the fitness function.

[0067] The third step is to substitute the optimized parameters into VMD to decompose the hybrid energy storage reference power and obtain the multi-scale intrinsic mode functions.

[0068] The fourth step involves allocating high-frequency components to the supercapacitor and low-frequency components to the battery, based on the characteristics of power-type and energy-type energy storage, to achieve coordinated operation of the energy storage devices.

[0069] The fifth step is to establish a hybrid energy storage life assessment and full life cycle cost model, which includes investment costs, operation and maintenance costs, battery replacement costs, and residual value recovery costs.

[0070] The sixth step involves using the whale optimization algorithm to iteratively solve the above capacity optimization model, obtaining energy storage configuration results at different boundary points, and determining the capacity configuration scheme with the optimal overall system cost.

[0071] The seventh step involves using typical daily wind power data and numerical examples to verify the effectiveness of the proposed method in reducing system operating costs and improving grid stability.

[0072] The eighth step involves using the VMD optimized by WOA to decompose the hybrid energy storage reference power, determining the high and low frequency power components based on the boundary point, and dynamically allocating the power among different energy storage units by combining the charging and discharging efficiency, rated power, and SOC constraints of the battery and supercapacitor.

[0073] The ninth step is that the method can reduce the operating cost of the hybrid energy storage system, extend the service life of the energy storage equipment, and achieve a balance between the system's economy and reliability while ensuring that the grid-connected power of wind power meets national standards.

[0074] Beneficial Effects: The WOA algorithm is used to optimize VMD, obtaining the optimal parameter combination [K, α], avoiding the subjectivity of manually selecting VMD parameters. Compared to other algorithms, the WOA algorithm has superior computational accuracy and speed. After WOA-VMD decomposition and reconstruction, the hybrid energy storage reference power can achieve the effect of smoothing low-frequency fluctuations in batteries and smoothing high-frequency fluctuations in supercapacitors, realizing the complementary advantages of power-type and energy-type energy storage. Based on the WOA algorithm, the internal power allocation problem and cost optimization problem of HESS are jointly optimized and solved. Compared with other configuration methods, this invention can obtain the most economically optimal HESS power command and its corresponding capacity configuration scheme. Attached Figure Description

[0075] Figure 1 This is a flowchart of the method of the present invention;

[0076] Figure 2 Diagram of a wind-storage combined power generation system;

[0077] Figure 3 Flowchart of the adaptive moving average filtering method;

[0078] Figure 4 The flowchart shows the solution process based on the WOA algorithm. Detailed Implementation

[0079] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0080] like Figure 1 As shown, this invention provides a capacity optimization method for a deep-sea wind power combined with hybrid energy storage power generation system based on WOA-VMD, comprising the following steps:

[0081] S1: The grid-connected power is obtained using the adaptive moving average algorithm, and the unbalanced power between it and the original wind power is regarded as the reference power of HESS.

[0082] A combined wind-storage power generation system, consisting of a power grid, a wind power generation system, and a hybrid energy storage system, is such a system. Figure 2 As shown, it is expressed as follows:

[0083]

[0084]

[0085] In the formula: This represents the actual power output of the wind power. This refers to the grid-connected power of wind power. This represents the total power of HESS. and These are the power ratings of the storage battery and the supercapacitor, respectively.

[0086] To overcome Traditional hybrid energy storage power allocation methods that directly decompose power can lead to insufficient or excessive power smoothing, hindering the safe and stable operation of the power grid after wind power is connected. This embodiment selects an adaptive moving average filtering method to smooth wind power output, based on China's wind power grid connection power requirements, with specific limits shown in Table 1.

[0087] Table 1: Wind Power Active Power Fluctuation Requirements

[0088]

[0089] The moving average method involves selecting a time window width L, averaging the data within L, and replacing the center data with the average. This window moves from front to back with the time series; for each new sampled data point added, one older data point is removed from the front of the window. Its expression is:

[0090]

[0091] In the formula, Let h be the value of the h-th input data; This represents the value of the h-th filtered output data.

[0092] When the wind power output meets the requirements, it can be directly connected to the grid; otherwise, filtering is required to ultimately obtain wind power grid-connected power that meets national standards. The specific process is as follows: Figure 3 As shown.

[0093] S2: The optimal parameter combination [K,α] is obtained by using WOA-VMD, and then the HESS reference power is decomposed by substituting it into the VMD algorithm. The power is then allocated to the battery and supercapacitor in combination with their performance.

[0094] The VMD algorithm overcomes the mode aliasing problem inherent in the EMD algorithm and is a novel adaptive, fully non-recursive signal processing and analysis method. It can effectively separate intrinsic mode functions (IMFs), and its overall approach involves constructing and solving a variational problem. The constrained variational problem model is as follows:

[0095]

[0096] In the formula, K is the number of modal decompositions; This is the k-th IMF after decomposition; The original signal; The center frequency corresponding to each IMF; A unit pulse signal;

[0097] By introducing the Lagrange multiplication operator λ and the quadratic penalty factor α, the constrained variational problem is transformed into an unconstrained variational problem, yielding the Lagrange formula:

[0098]

[0099] The above equation is solved using the alternating direction multiplier algorithm, i.e., for... , , Iterate and update continuously until the update stops when the conditions are met. At that time, the final output includes all IMFs and their corresponding center frequencies.

[0100] As the above analysis shows, K determines the number of IMF decompositions, and an inappropriate value can lead to over- or under-decomposition of the signal; α determines the bandwidth of the IMF components, and an inappropriate selection can cause information loss or redundancy. The selection of these two values ​​will seriously affect the decomposition effect of the original signal. Therefore, when applying VMD to decompose a signal, it is necessary to set an appropriate number of modes K and a quadratic penalty factor α. However, the traditional method of selecting these parameters based on empirical values ​​cannot guarantee that they are optimal.

[0101] To address the aforementioned problems, this invention utilizes the WOA algorithm with sample entropy SE as the fitness function to optimize VMD, obtaining the optimal parameter combination [K, α], thereby improving the VMD decomposition effect. SE is an indicator of the disorder level of a time series, and its value is inversely proportional to the effective information contained in the series. For a time series with a finite number of data points N... SE is represented as:

[0102]

[0103] In the formula, q is the dimension; r is the similarity tolerance; , Let q and q+1 be the probabilities of matching points in the lower region r, respectively;

[0104] like Figure 4 As shown, in the algorithm's solution, the position of each whale represents a feasible solution in the solution space, and the whale finds the optimal solution by adjusting its position; its position is updated by the following formula;

[0105]

[0106] In the formula: n is the number of iterations, The position vector of the optimal solution; This is the current position vector; Let be the position vector at iteration number n+1; D simulates the distance between the target and the target; A and c are coefficients.

[0107] Let p be a random number, p ,when When p > 0.5, the whale is in the development stage of a bubble web attack, and its position is updated by the following formula;

[0108]

[0109] In the formula, b is the helical constant; l is Random numbers;

[0110] when If p < 1 and p < 0.5, the whale is in a shrinking enclosure during the development phase, and its position can be updated using the following formula;

[0111]

[0112] when At position 1, the whale is in the exploration phase and updates its position using the following formula;

[0113]

[0114] WOA-VMD optimizes VMD parameters based on WOA. The specific process is as follows:

[0115] A1: Set the initial parameters for WOA;

[0116] A2: Randomly initialize the location of the whale population - i.e., the VMD parameter combination, and calculate the prey objective function value for each individual whale;

[0117] A3: Determine the current optimal whale position based on the fitness value;

[0118] A4: The whale's position is updated according to WOA's two main update mechanisms (shrinking encirclement and bubble web attack);

[0119] A5: The updated whale position is used to recalculate the objective function value and compare it with the original value, retaining the better solution;

[0120] A6: If the termination condition is met (maximum number of iterations or error threshold is reached), output the optimal solution; otherwise, return to step A3.

[0121] Compared to traditional VMD methods, this application uses the WOA algorithm to optimize VMD parameters, overcoming the subjectivity of parameter selection and providing a good foundation for hybrid energy storage power allocation.

[0122] The total power of HESS is calculated using the WOA-VMD algorithm. The system is decomposed into K IMFs, and reconstructed into high- and low-frequency power commands by selecting a high- and low-frequency boundary point f. Based on the characteristics of hybrid energy storage, the supercapacitor handles high-frequency power, and the battery handles low-frequency power, i.e.:

[0123]

[0124] Considering the energy storage charging and discharging efficiency, the rated power configuration of the energy storage should be as follows:

[0125]

[0126] In the formula, This refers to the rated power of the battery. Rated power of supercapacitors; The charging power of the battery; The charging power of the supercapacitor.

[0127] S3: Considering the economics of the energy storage system throughout its entire life cycle and using energy storage charging and discharging power, power balance, and state of charge (SOC) as constraints, an energy storage economics evaluation model is established. The WOA algorithm is used to jointly optimize and solve the power allocation and cost model within the hybrid energy storage system, thereby determining the optimal high- and low-frequency boundary points of the hybrid energy storage power command and their corresponding configuration schemes.

[0128] The establishment of the energy storage economic evaluation model includes:

[0129] An economic evaluation model based on LCC theory is established with the objective function of minimizing the annual comprehensive cost of the energy storage system, namely:

[0130]

[0131] In the formula: The investment cost of a hybrid energy storage system; For operation and maintenance costs; Cost of battery replacement; To recover residual value costs.

[0132] Investment costs The expression is as follows:

[0133]

[0134] In the formula: , These are the investment costs for batteries and supercapacitors, respectively. , , , These represent the investment cost per unit of power and capacity of a battery and a supercapacitor, respectively. The discount rate; , These are the rated capacities of the storage battery and the supercapacitor, respectively. This refers to the operating life of a supercapacitor.

[0135] Operation and maintenance costs The expression is as follows:

[0136]

[0137] In the formula: , These are the operating and maintenance costs for batteries and supercapacitors, respectively. , , , These are the power and capacity operation and maintenance costs per unit of battery and supercapacitor, respectively.

[0138] Recovery of residual value cost The expression is as follows:

[0139]

[0140] In the formula: The recovery residual value rate is generally taken as 3% to 5%.

[0141] Regarding the operating life of supercapacitors The calculation is as follows:

[0142] Because the charging and discharging process of supercapacitors is reversible, their cycle life is typically hundreds of thousands to millions of times, far exceeding that of batteries. Therefore, this invention sets their lifespan as a constant. Battery lifespan, however, is affected by internal and external factors such as temperature, depth of discharge (DOD), and the number of charge-discharge cycles. Since the depth of discharge has a significant impact on its lifespan, to simplify calculations, this invention only considers the impact of DoD on the battery's cycle life. The calculation method for DoD is as follows:

[0143]

[0144] In the formula, Let t be the actual depth of discharge of the battery at time t.

[0145] When assessing battery life using the rainflow counting method, the functional relationship between DoD (Dosage of Damping) and the number of cycles is as follows:

[0146]

[0147] In the formula, This refers to the rated depth of charge and discharge of the battery. This refers to the number of cycles a battery can perform at its rated depth of charge and discharge.

[0148] The battery life within one working cycle T is:

[0149]

[0150] The constraints of the energy storage economics assessment model are expressed as follows:

[0151] Power balance constraints:

[0152]

[0153] Energy storage charging and discharging power constraints:

[0154]

[0155] Energy storage SOC constraints:

[0156]

[0157] In the formula, This is the lower limit of the battery's state of charge (SOC). This represents the upper limit of the battery's state of charge (SOC). This represents the lower limit of the state of charge (SOC) of a supercapacitor. This represents the upper limit of the SOC (State of Charge) of a supercapacitor.

[0158] In this embodiment, it is considered that the results of hybrid energy storage power allocation and hybrid energy storage capacity configuration will be affected by the high-frequency and low-frequency boundary points of hybrid energy storage power. The above-mentioned hybrid energy storage capacity optimization configuration model is a nonlinear, multi-constraint optimization problem. The WOA algorithm can be used to calculate the annual comprehensive cost of the system under different boundary points. By comparing the results, the optimal boundary point and its corresponding energy storage configuration scheme can be determined.

[0159] This invention first combines wind power grid connection power smoothing demand with an adaptive moving average algorithm to obtain wind power grid connection power and hybrid energy storage reference power that meet national standards. Second, it introduces the Whale Optimization Algorithm (WOA) to optimize the mode number K and quadratic penalty factor α in Variational Mode Decomposition (VMD), thereby achieving effective decomposition of fluctuating wind power and completing a reasonable power allocation between batteries and supercapacitors. Then, it establishes a hybrid energy storage capacity optimization model with the objectives of system investment, operation and maintenance, equipment replacement, and residual value recovery costs to obtain the economically optimal configuration scheme for the system. Finally, simulation analysis using actual operating data from typical wind farms verifies the effectiveness and practicality of this method in reducing system operating costs and improving wind power grid connection stability.

Claims

1. A capacity optimization method for a deep-sea wind power combined with hybrid energy storage power generation system based on WOA, characterized in that: Includes the following steps: S1: Determine the reference power based on the grid-connected power; S2: The variational mode decomposition algorithm is optimized using the WOA algorithm with sample entropy SE as the fitness function to obtain the optimal parameter combination. SE is an indicator of the disorder of the time series. S3: Based on the optimal parameter combination, the WOA algorithm is used to solve the evaluation model established based on LCC theory to determine the system capacity optimization scheme.

2. The capacity optimization method according to claim 1, characterized in that: Step S1 specifically includes: The moving average method involves selecting a time window width L, averaging the data within L, and replacing the center data with the average. This window moves from front to back with the time series; for each new sampled data point added, one older data point is removed from the front of the window. Its expression is: ； In the formula, Let h be the value of the h-th input data; This represents the value of the h-th filtered output data. ； ； In the formula: This represents the actual power output of the wind power. This refers to the grid-connected power of wind power. This represents the total power of HESS. and These are the power ratings of the storage battery and the supercapacitor, respectively.

3. The capacity optimization method according to claim 2, characterized in that: In step S2, WOA-VMD optimizes VMD parameters based on WOA. The specific process is as follows: A1: Set the initial parameters for WOA; A2: Randomly initialize the location of the whale population - i.e., the VMD parameter combination, and calculate the prey objective function value for each individual whale; A3: Determine the current optimal whale position based on the fitness value; A4: Whale location is updated according to WOA's two main update mechanisms; A5: The updated whale position is used to recalculate the objective function value and compare it with the original value, retaining the better solution; A6: If the termination condition is met, output the optimal solution; otherwise, return to step A3.

4. The capacity optimization method according to claim 3, characterized in that: In step S2, the WOA algorithm is used to optimize VMD with sample entropy SE as the fitness function to obtain the optimal parameter combination [K, α], thereby improving the VMD decomposition effect; for time series with a finite number of data N. SE is represented as: ； In the formula, q is the dimension; r is the similarity tolerance; , Let q and q+1 be the probabilities of matching points in the lower region r, respectively; In the algorithm, the position of each whale represents a feasible solution in the solution space, and the whales find the optimal solution by adjusting their positions. Its position is updated by the following formula; ； In the formula: n is the number of iterations, The position vector of the optimal solution; This is the current position vector; Let be the position vector at iteration number n+1; D simulates the distance between the target and the target; A and c are coefficients. Let p be a random number, p ,when When p > 0.5, the whale is in the development stage of a bubble web attack, and its position is updated by the following formula; ； In the formula, b is the helical constant; l is... Random numbers; when If p < 1 and p < 0.5, the whale is in a shrinking enclosure during the development phase, and its position can be updated using the following formula; ； when At position 1, the whale is in the exploration phase and updates its position using the following formula; 。 5. The capacity optimization method according to claim 4, characterized in that: In step S2, the WOA-VMD algorithm is used to calculate the total power of HESS. The system is decomposed into K IMFs, and reconstructed into high- and low-frequency power commands by selecting a high- and low-frequency boundary point f. Based on the characteristics of hybrid energy storage, the supercapacitor handles high-frequency power, and the battery handles low-frequency power, i.e.: ； Considering the energy storage charging and discharging efficiency, the rated power configuration of the energy storage should be as follows: ； In the formula, This refers to the rated power of the battery. Rated power of supercapacitors; The charging power of the battery; The charging power of the supercapacitor.

6. The capacity optimization method according to claim 5, characterized in that: The establishment of the energy storage economic evaluation model in step S3 includes: An economic evaluation model based on LCC theory is established with the objective function of minimizing the annual comprehensive cost of the energy storage system, namely: ； In the formula: The investment cost of a hybrid energy storage system; For operation and maintenance costs; Cost of battery replacement; To recover residual value costs.

7. The capacity optimization method for a deep-sea wind power combined with hybrid energy storage power generation system based on WOA-VMD according to claim 6, characterized in that: Investment costs The expression is as follows: ； In the formula: , These are the investment costs for batteries and supercapacitors, respectively. , , , These represent the investment cost per unit of power and capacity of a battery and a supercapacitor, respectively. The discount rate; , These are the rated capacities of the storage battery and the supercapacitor, respectively. This refers to the operating life of a supercapacitor.

8. The capacity optimization method according to claim 6, characterized in that: Operation and maintenance costs The expression is as follows: ； In the formula: , These are the operating and maintenance costs for batteries and supercapacitors, respectively. , , , These are the power and capacity operation and maintenance costs per unit of battery and supercapacitor, respectively.

9. The capacity optimization method according to claim 6, characterized in that: Recovery of residual value cost The expression is as follows: ； In the formula: The recovery residual value rate.

10. The capacity optimization method according to claim 6, characterized in that: The constraints of the energy storage economic evaluation model in step S3 are expressed as follows: Power balance constraints: ； Energy storage charging and discharging power constraints: ； Energy storage SOC constraints: ； In the formula, This is the lower limit of the battery's state of charge (SOC). This represents the upper limit of the battery's state of charge (SOC). This represents the lower limit of the state of charge (SOC) of a supercapacitor. This represents the upper limit of the SOC (State of Charge) of a supercapacitor.

Images