Power cooperative regulation and control method and device based on heterogeneous energy storage cluster

Abstract

The invention discloses a power cooperative regulation and control method and device based on a heterogeneous energy storage cluster, and the method comprises the steps: carrying out the determination and optimization processing according to a user demand, the output power of a renewable energy subsystem, and a variational mode algorithm, and obtaining a target high-frequency signal and a target low-frequency signal; determining the charging and discharging power of each single super capacitor according to the target high-frequency signal and the health state and the charge state of the single super capacitor; determining the charging and discharging power of each single battery according to the target low-frequency signal and the health state and the charge state of the single battery; judging whether the charging and discharging power of each single super capacitor and the charging and discharging power of each single battery meet a preset charging and discharging constraint condition, an electric power balance constraint condition and a cost constraint condition or not; and if all constraint conditions are met, taking the scheme as a final charging and discharging scheme. According to the application, the systematized and refined response to the power grid total dispatching instruction is realized by optimizing the distribution of the power among different stations and heterogeneous energy storage units.

Description

Technical Field

[0001] This application relates to the field of power regulation technology, and in particular to a power coordinated regulation method and device based on heterogeneous energy storage clusters. Background Technology

[0002] Against the backdrop of a global energy structure transitioning towards low-carbon and clean energy, renewable energy sources, represented by wind and solar power, are developing at an unprecedented pace. However, the power output of these energy sources inherently possesses intermittency, volatility, and randomness. Their large-scale grid connection poses a severe challenge to the stability, reliability, and power quality of traditional power grids, often causing frequency fluctuations, voltage disturbances, and even wind and solar power curtailment in localized areas. In this context, Energy Storage Systems (ESS), with their dynamic compensation and optimization capabilities, have become a key technology for solving the grid integration and consumption of renewable energy and supporting the stable operation of the power grid.

[0003] With technological advancements and the expansion of project scale, energy bases are evolving towards clustering, integration, and diversification, forming complex "new energy clusters" that include multiple geographically dispersed wind and solar power plants and supporting energy storage facilities. To maximize system efficiency and economy, these clusters typically employ a variety of energy storage technologies with different characteristics, i.e., heterogeneous energy storage. However, this also brings about a core technological contradiction—there are inherent constraints on performance and cost between different energy storage units and individual energy storage devices.

[0004] Specifically, this technological contradiction manifests itself in the following aspects:

[0005] Contradiction 1: Power-type energy storage units, represented by supercapacitors (SC), possess millisecond-level instantaneous response speed, extremely high power density, and cycle life, making them an ideal choice for coping with high-frequency, instantaneous power surges in the power grid. However, their inherent weakness of low energy density prevents them from providing long-term energy support.

[0006] Contradiction 2: Energy storage units are mainly represented by lithium-ion batteries, which have high energy density and are suitable for power fluctuation regulation on a minute-by-minute basis. However, their cycle life is limited. If they are frequently used for high-frequency, deep charge-discharge regulation, their performance will degrade rapidly, significantly increasing the replacement and maintenance costs throughout their entire life cycle.

[0007] Contradiction 3: Long-duration energy storage and energy conversion units, primarily represented by fuel cell technology, include two modes: solid oxide fuel cells (SOFC) and proton exchange membrane electrolyzers (PEM). SOFCs can efficiently convert chemical energy such as hydrogen into electrical energy with extremely high energy density, providing continuous power supply for hours or even days, which is crucial for ensuring the continuity of power supply during extreme weather conditions. However, their response speed is relatively slow, making them unsuitable for rapid adjustment tasks. When wind and solar power curtailment occurs, PEM electrolyzers can utilize surplus electricity to electrolyze water to produce hydrogen, efficiently converting electrical energy into chemical energy for storage. This not only alleviates the peak-shaving pressure on the power grid but also provides a fuel source for SOFCs, improving the flexibility and energy utilization rate of the entire power system.

[0008] Currently, most existing energy storage control technologies focus on energy management strategies within a single, centralized hybrid energy storage power station, generally lacking a higher-level collaborative scheduling framework for large-scale new energy clusters. For large clusters containing multiple independently operating hybrid energy storage power stations, how to optimally and hierarchically decompose and distribute the overall dispatch instructions issued by the power grid within the cluster, and on this basis guide the heterogeneous energy storage units (such as supercapacitors, lithium batteries, SOFC / PEM) within each station to work collaboratively according to their respective optimal operating characteristics, is a key technical challenge that has not yet been effectively solved. Summary of the Invention

[0009] Therefore, the purpose of this application is to provide a power coordinated regulation method and device based on heterogeneous energy storage clusters, which aims to achieve a systematic and refined response to the overall grid dispatch command. By optimizing the power distribution among different power stations and heterogeneous energy storage units, the technical advantages of various energy storage technologies can be fully utilized, ultimately achieving the comprehensive goal of improving the overall dispatchability of new energy bases, significantly extending the operating life of energy storage systems, and reducing their total life cycle cost.

[0010] In a first aspect, embodiments of the present invention provide a power coordinated regulation method based on a heterogeneous energy storage cluster. The regulation method is applied to a power coordinated regulation system based on a heterogeneous energy storage cluster. The regulation system includes multiple power stations and energy substitute subsystems. Each power station includes a set of renewable energy subsystems and a set of energy storage subsystems. Each energy storage subsystem includes a set of supercapacitor units and a set of battery units. Each supercapacitor unit includes multiple individual supercapacitors, and each battery unit includes multiple individual batteries. The regulation method includes: S102: determining an initial high-frequency signal and an initial low-frequency signal based on user requirements, the output power of the renewable energy subsystems, and a variational mode algorithm; S104: optimizing the initial high-frequency signal and the initial low-frequency signal based on the state of charge of the individual supercapacitors and the state of charge of the individual batteries, respectively. S106: Determine the charging and discharging power of each supercapacitor based on the target high-frequency signal, the health status and state of charge of the individual supercapacitors; S108: Determine the charging and discharging power of each battery based on the target low-frequency signal, the health status and state of charge of the individual batteries; S110: Determine whether the charging and discharging power of each supercapacitor and each battery meets the preset charging and discharging constraints, power balance constraints and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem; S112: If all constraints are met, the charging and discharging power of the supercapacitors determined in S106 and the battery determined in S108 are used as the final charging and discharging scheme.

[0011] Further, S102 includes: S102-2: determining the overall dispatch instruction P0 by subtracting the pre-acquired user demand power from the output power of the renewable energy subsystem; S102-4: if the overall dispatch instruction P0 is positive, then the energy storage subsystem is determined to be in a discharging state; S102-6: if the overall dispatch instruction P0 is negative, then the energy storage subsystem is determined to be in a charging state; S102-8: decomposing and reconstructing the overall dispatch instruction P0 according to the charging and discharging state of the energy storage subsystem and the variational mode algorithm to determine the initial high-frequency signal and the initial low-frequency signal.

[0012] Further, if the energy storage subsystem is in a discharging state, then S102-8 includes: S102-8-2: determining whether all current energy storage subsystems meet user requirements based on the overall scheduling instruction P0 and the energy storage capacity of individual supercapacitors and individual batteries in all current energy storage subsystems; S102-8-4: if user requirements are met, then decomposing and reconstructing the overall scheduling instruction P0 based on the variational mode algorithm to determine the initial high-frequency signal and the initial low-frequency signal; S102-8-6: if not met, then adjusting the operating state of the energy replenishment subsystem until user requirements are met.

[0013] Furthermore, if the energy storage subsystem is in a charging state, then S102-8-2 is replaced by: determining whether all the current energy storage subsystems meet the user's needs based on the scheduling general instruction P0, the energy storage capacity of the individual supercapacitors in all the current energy storage subsystems, the energy storage capacity of the individual batteries, the rated capacity of the individual supercapacitors, and the rated capacity of the individual batteries.

[0014] Further, S102-8-4 includes: S102-8-4-2: decomposing the total scheduling instruction P0 into multiple constrained solid-state model components based on pre-acquired variational mode decomposition parameters; S102-8-4-4: optimizing the solid-state model components based on pre-acquired quadratic penalty factor α and Lagrange multiplier operators to determine unconstrained frequency conversion model components; S102-8-4-6: dividing the unconstrained frequency conversion model components into high-frequency signal groups and low-frequency signal groups based on preset frequency thresholds; S102-8-4-8: summing the frequency conversion model components in the high-frequency signal group and the low-frequency signal group respectively to obtain initial high-frequency signals and initial low-frequency signals.

[0015] Further, S104 includes: S104-2: Optimizing the initial high-frequency signal based on the pre-acquired state of charge and charge threshold range of all current single supercapacitors to obtain a target high-frequency signal; S104-4: Optimizing the initial low-frequency signal based on the pre-acquired state of charge and charge threshold range of all current single battery cells to obtain a target low-frequency signal.

[0016] Further, the control system includes N stations, each supercapacitor unit includes X individual supercapacitors, and each battery unit includes Y individual batteries. Different supercapacitor units include different numbers of individual supercapacitors, and different battery units include different numbers of individual batteries. S108 includes: S108-2: Decompose the target low-frequency signal into N1 first battery charging and discharging powers according to the health state and state of charge of the battery units and allocate them to the corresponding battery units, where N1≤N; S108-4: For each battery unit, decompose the first battery charging and discharging power into Y1 second battery charging and discharging powers according to the health state and state of charge of the individual batteries and allocate them to the corresponding individual batteries, where Y1≤Y.

[0017] Further, S108-2 includes: S108-2-2: Constructing a battery cell sorting function based on the health state and state of charge of the battery cells, sorting N groups of battery cells, and obtaining a first battery sorting result; S108-2-4: Determining the battery cells participating in the charging and discharging action based on the first battery sorting result and a preset first battery sorting threshold, wherein the number of battery cells participating in the charging and discharging action is N1; S108-2-6: Constructing a power allocation function based on the health state and state of charge of the battery cells, decomposing the target low-frequency signal into N1 first battery charging and discharging powers and allocating them to the corresponding battery cells.

[0018] Further, for each group of battery cells, S108-4 is executed, which includes: S108-4-2: sorting the Y individual cells in the battery cell according to the health state and state of charge of the individual cells to obtain the second battery sorting result; S108-4-4: determining the individual cells participating in the charging and discharging action based on the second battery sorting result and the preset second battery sorting threshold, wherein the number of individual cells participating in the charging and discharging action is Y1; S108-4-6: constructing a power allocation function according to the health state and state of charge of the individual cells, decomposing the charging and discharging power of the first battery into Y1 second battery charging and discharging powers and allocating them to the corresponding individual cells.

[0019] Further, S106 includes: S106-2: Decomposing the target low-frequency signal into N2 first supercapacitor charging and discharging powers according to the health state and state of charge of the supercapacitor units and allocating them to the corresponding supercapacitor units, where N2≤N; S106-4: For each group of supercapacitor units, decomposing the first supercapacitor charging and discharging power into X1 second supercapacitor charging and discharging powers according to the health state and state of charge of the individual supercapacitors and allocating them to the corresponding individual supercapacitors, where X1≤X.

[0020] Further, S106-2 includes: S106-2-2: Constructing a supercapacitor unit sorting function based on the health state and state of charge of the supercapacitor units, sorting N groups of supercapacitor units, and obtaining a first supercapacitor sorting result; S106-2-4: Determining the supercapacitor units participating in the charging and discharging action based on the first supercapacitor sorting result and a preset first supercapacitor sorting threshold, wherein the number of supercapacitor units participating in the charging and discharging action is N2; S106-2-6: Constructing a power allocation function based on the health state and state of charge of the supercapacitor units, decomposing the target low-frequency signal into N2 first supercapacitor charging and discharging powers and allocating them to the corresponding supercapacitor units.

[0021] Further, for each group of supercapacitor units, S106-4 is executed, which includes: S106-4-2: sorting the X individual supercapacitors in the supercapacitor unit according to the health state and state of charge of the individual supercapacitors to obtain a second supercapacitor sorting result; S106-4-4: determining the individual supercapacitors participating in the charging and discharging action based on the second supercapacitor sorting result and a preset second supercapacitor sorting threshold, wherein the number of individual supercapacitors participating in the charging and discharging action is X1; S106-4-6: constructing a power allocation function according to the health state and state of charge of the individual supercapacitors, decomposing the charging and discharging power of the first supercapacitor into X1 second supercapacitor charging and discharging powers and allocating them to the corresponding individual supercapacitors.

[0022] Furthermore, S110 includes: determining whether the power balance constraint condition is met based on the operating status of the energy substitute subsystem, the charging and discharging power of each of the individual supercapacitors, and the charging and discharging power of each of the individual batteries.

[0023] Furthermore, S110 also includes: determining the investment cost based on the charging and discharging power of each of the individual supercapacitors and the charging and discharging power of each of the individual batteries; constructing a cost function based on the investment cost, the predetermined operation and maintenance cost, and the predetermined scrapping cost; and determining whether the cost constraint conditions are met based on a preset cost threshold.

[0024] Furthermore, S112 also includes: if the charging and discharging power of each supercapacitor and the charging and discharging power of each battery do not meet any of the preset charging and discharging constraints, power balance constraints, and cost constraints, then adjust the operating state of the energy substitute subsystem and / or the renewable energy subsystem, and return to S102.

[0025] Secondly, embodiments of the present invention provide a power coordinated regulation device based on a heterogeneous energy storage cluster. The regulation device includes: a first regulation module, configured to determine an initial high-frequency signal and an initial low-frequency signal based on user demand, the output power of the renewable energy subsystem, and a variational mode algorithm; a second regulation module, configured to optimize the initial high-frequency signal and the initial low-frequency signal based on the state of charge of individual supercapacitors and the state of charge of individual batteries, respectively, to obtain a target high-frequency signal and a target low-frequency signal; and a third regulation module, configured to determine the charging state of each individual supercapacitor based on the target high-frequency signal, the health state of the individual supercapacitor, and the state of charge. The system includes: a fourth control module for determining the charging and discharging power of each individual battery based on the target low-frequency signal, the health status of the individual battery, and the state of charge; a fifth control module for determining whether the charging and discharging power of each supercapacitor and each individual battery meets preset charging and discharging constraints, power balance constraints, and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem; and a sixth control module for using the charging and discharging power of the supercapacitor determined in S106 and the individual battery determined in S108 as the final charging and discharging scheme if all constraints are met.

[0026] The beneficial effects of the embodiments of the present invention are as follows:

[0027] This application discloses a power coordinated regulation method and device based on heterogeneous energy storage clusters. The method includes determining and optimizing target high-frequency and low-frequency signals based on user needs, the output power of the renewable energy subsystem, and variational mode algorithms; determining the charging and discharging power of each supercapacitor based on the target high-frequency signal, the health state and state of charge of individual supercapacitors; determining the charging and discharging power of each battery based on the target low-frequency signal, the health state and state of charge of individual batteries; determining whether the charging and discharging power of each supercapacitor and each battery satisfies preset charging and discharging constraints, power balance constraints, and cost constraints; if all constraints are met, the above scheme is adopted as the final charging and discharging scheme. This application achieves a systematic and refined response to the overall grid dispatch command by optimizing the power distribution among different power stations and heterogeneous energy storage units.

[0028] Other features and advantages of this application will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the above-described techniques of this application.

[0029] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the specific embodiments of this application or 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 this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0031] Figure 1 A flowchart of a power coordinated control method based on heterogeneous energy storage clusters provided for this application;

[0032] Figure 2 A flowchart of another power coordinated regulation method based on heterogeneous energy storage clusters provided in this application;

[0033] Figure 3 A schematic diagram of a power coordinated control system based on heterogeneous energy storage clusters provided for this application;

[0034] Figure 4 A schematic diagram of another power coordinated control system based on heterogeneous energy storage clusters provided for this application;

[0035] Figure 5 A schematic diagram illustrating the SOC (State of Charge) of the supercapacitor and lithium battery provided in this application;

[0036] Figure 6 A schematic diagram of an AGC signal provided in this application;

[0037] Figure 7 This is a schematic diagram of a power coordinated control device based on a heterogeneous energy storage cluster, provided for this application. Detailed Implementation

[0038] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] This application is applied to power grid regulation scenarios.

[0040] Example 1

[0041] This application provides a power coordinated regulation method based on heterogeneous energy storage clusters. See the flowchart below. Figure 1 and Figure 2 This method is applied to Figure 3 , Figure 4 The regulatory system.

[0042] The regulation method is applied to a power coordinated regulation system based on heterogeneous energy storage clusters. The regulation system includes multiple power stations and energy substitute subsystems. Each power station includes a set of renewable energy subsystems and a set of energy storage subsystems. Each set of energy storage subsystems includes a set of supercapacitor units and a set of battery units. Each set of supercapacitor units includes multiple individual supercapacitors, and each set of battery units includes multiple individual batteries.

[0043] The control method includes:

[0044] S102: Determine the initial high-frequency signal and the initial low-frequency signal (corresponding to) based on user requirements, the output power of the renewable energy subsystem, and the variational mode algorithm. Figure 2 (S1-S4).

[0045] S102 includes:

[0046] S102-2: The difference between the pre-acquired user demand power and the output power of the renewable energy subsystem is used to determine the general dispatch instruction P0 (i.e., ... Figure 2 S1).

[0047] Specifically, S102-2 includes: a cluster dispatch center (see...) Figure 3 Based on the load required by the user side, and the current total actual wind and solar power output (i.e. Figure 3 The difference between the two is the active power absorbed or released by the energy storage power station from the renewable energy system. This difference is also the general dispatch instruction issued by the dispatch center to each hybrid energy storage station. Here, the instruction issued by the dispatch center is defined as P0, which is an AGC instruction.

[0048] S102-4: If the overall scheduling instruction P0 is positive, the energy storage subsystem is determined to be in a discharging state. S102-6: If the overall scheduling instruction P0 is negative, the energy storage subsystem is determined to be in a charging state.

[0049] Specifically, each independent energy storage power station will be allocated to the site (i.e. Figure 3The active power command (P0) of a power grid cluster is compared with the actual wind and solar power output of that cluster. When the actual wind and solar power output exceeds the active power command, indicating wind and solar power curtailment, the difference is positive. The energy storage power station absorbs active power from the renewable energy system and stores the electrical energy in energy storage media such as battery packs and supercapacitors (charging process). Energy storage station charging / discharging: When the actual wind and solar power output is less than the active power command, indicating excessive user load, the difference is negative. The energy storage power station releases the stored electrical energy back to the grid (discharging process). The cluster dispatch center comprehensively considers the real-time status, available capacity, and technology type of each energy storage power station within the cluster to allocate dispatch commands, completing cluster-level allocation and achieving coordinated response across the entire cluster.

[0050] S102-8: Based on the charging and discharging state of the energy storage subsystem and the variational mode algorithm, the scheduling command P0 is decomposed and reconstructed to determine the initial high-frequency signal and the initial low-frequency signal.

[0051] If the energy storage subsystem is in a discharge state, then S102-8 includes:

[0052] S102-8-2: Based on the overall scheduling instruction P0 and the energy storage capacity of the individual supercapacitors and individual batteries in all current energy storage subsystems, determine whether all current energy storage subsystems meet user needs (i.e., Figure 2 S2).

[0053] Specifically, first determine all supercapacitors in the current control system. And the energy storage capacity of all current lithium batteries Can it meet the existing scheduling instruction P0 requirements, given the capacity of the supercapacitor? and the energy storage capacity of lithium batteries If the requirements of scheduling instruction P0 are met, power allocation will be performed using the following subsequent methods. If not, the supercapacitor and lithium battery will operate at full capacity, with assistance from the energy replenishment module (i.e., S102-8-6). Figure 2 (S3).

[0054] 1) If the energy storage subsystem is in a discharge state, the determination formula is as follows:

[0055] Judgment during discharge: Formula 1;

[0056] T1 refers to the time it takes for all supercapacitors and all lithium batteries to discharge their current energy.

[0057] Specifically, T1 = MAX{Tsc, Tli}, where Tsc refers to the time required for the supercapacitor with the largest remaining capacity to completely discharge, and Tli refers to the time required for the lithium battery with the largest remaining capacity to completely discharge.

[0058] 2) If the energy storage subsystem is in a charging state, then S102-8-2 is replaced with:

[0059] Based on the general scheduling instruction P0, the energy storage capacity of each supercapacitor and battery in all current energy storage subsystems, the system determines whether all current energy storage subsystems meet user requirements.

[0060] |P0|×T2≤∑(Escrate,k-Esc,k)+∑(Elirate,k-Eli,k)Formula 2;

[0061] Where T2 refers to the time required for all supercapacitors and all lithium batteries to fully charge to their rated capacity from their current capacity; Escrate,k is the rated capacity of the supercapacitors; Elirate,k is the rated capacity of the lithium batteries; and Esc,k is the current capacity of a single supercapacitor. The sum of Esc,k is... Eli,k is the current capacity of a single lithium battery; the sum of Eli,k is... .

[0062] S102-8-4: If the user's needs are met, the total scheduling instruction P0 is decomposed and reconstructed based on the variational mode algorithm to determine the initial high-frequency signal and the initial low-frequency signal.

[0063] S102-8-6: If not satisfied, adjust the operating state of the energy replenishment subsystem until the user's needs are met (i.e., ...). Figure 2 (S3).

[0064] Specifically, 1) During the discharge process, if the demand cannot be met, the supercapacitor and lithium battery operate at full capacity while maintaining a healthy state (discharge process). The excess is powered by a solid oxide fuel cell (SOFC), and the large amount of waste heat generated is stored in molten salt. 2) During the charging process, if the demand cannot be met, the supercapacitor and lithium battery operate at full capacity while maintaining a healthy state (charging process). The excess is used to electrolyze water and store hydrogen in a PEM electrolyzer hydrogen production unit.

[0065] More specifically, all capacitors and batteries in the non-overcharged and non-over-discharged SOC region are operated at full power (i.e., at maximum charge and discharge power). SOFC (Solid Oxide Fuel Cell) uses hydrogen and oxygen as fuel, producing electricity and heat through a chemical reaction; the electricity is used to supply the insufficient portion of the power supply to the user side. PEM (Proton Exchange Membrane Electrolyte) and SOFC (Solid Oxide Fuel Cell) are the proton exchange membrane electrolyzer and solid oxide electrolyzer, respectively. The reaction is the opposite of the fuel cell described above; water is electrolyzed to produce hydrogen and oxygen, and the hydrogen can be stored in a hydrogen storage tank for energy storage. Figure 2The S3 section contains all of these features, and this part is used to supplement the parts that lithium batteries and supercapacitors cannot fully address.

[0066] S102-8-4 includes:

[0067] S102-8-4-2: Based on the pre-acquired variational mode decomposition parameters, the total scheduling instruction P0 is decomposed into multiple solid-state model components under constraint states.

[0068] Specifically, S102-8-4-2 includes: the dispatch controller of the field group dispatch center first uses VMD (variable mode decomposition) to decompose and reconstruct the total dispatch command signal P0 (i.e., an AGC signal) to roughly obtain a high-frequency signal. With a low-frequency signal .

[0069] Concept Explanation: VMD (Variational Mode Decomposition): VMD is an innovative multi-component signal decomposition algorithm that uses Wiener filtering, Hilbert transform, and hybrid techniques to decompose complex signals into intrinsic mode functions (IMFs) with different center frequencies. VMD is based on the principle of variational analysis and achieves signal decomposition by minimizing the objective function.

[0070] S102-8-4-2 includes: allocating the total active power command P0 (i.e., in the formula) to the field group according to the pre-acquired VMD. The model is decomposed into k solid-state model components (IMFs) to obtain the variational model formula:

[0071] Formula 3;

[0072] in, This represents the Dirac delta function, which is an idealized mathematical model of a spike or impact.

[0073] It is the set of IMF (Intrinsic Mode Function) components;

[0074] A set of center frequencies;

[0075] The original automatic gain control (AGC) signal before decomposition is P0;

[0076] It is a Lagrange multiplier operator.

[0077] The above formula decomposes the active power command (i.e., the dispatch command signal, a type of AGC signal) into K IMF components. Each IMF corresponds to a sub-signal with different frequency characteristics. Based on the frequency characteristics, the components are divided into two types of signals (high-frequency signals and low-frequency signals). "Active power command" refers to the original AGC signal before decomposition (the total active power command of the dispatch center), which is the input signal of VMD. The sum of all IMF components must be equal to this original signal.

[0078] S102-8-4-4: Optimize the solid-state model components based on the pre-acquired quadratic penalty factor α and Lagrange multiplier operators to determine the unconstrained frequency conversion model components.

[0079] S102-8-4-4 includes: transforming the above-mentioned constrained variational problem into the following unconstrained problem (in order to solve for K modal components), and using a quadratic penalty factor and Lagrange multiplier operators to construct an enhanced Lagrange function, the specific formulas of which are as follows:

[0080]

[0081] Formula 4;

[0082] After obtaining the solution to the above variational problem, the AGC (Automatic Generation Control) signal is divided into several modal signals with unique frequency characteristics. The optimal solution to Equation 3 is solved by introducing a quadratic penalty factor α (to ensure reconstruction accuracy) and Lagrange multipliers (to maintain the strictness of the constraints). This is achieved by applying the alternating direction multiplier algorithm combined with Fourier domain analysis to decompose the signal P0, optimizing the intrinsic modal components through an iterative process, testing the decomposition results for different K values, and selecting the K value with no mode aliasing and clear spectral separation. The unique frequencies, i.e., the center frequencies of the IMF components, are different; high-frequency subsequences are reconstructed, and low-frequency subsequences are recombined.

[0083] S102-8-4-6: Based on a preset frequency threshold, the components of the unconstrained frequency conversion model are divided into a high-frequency signal group and a low-frequency signal group.

[0084] S102-8-4-8: Summate the frequency conversion model components in the high-frequency signal group and the low-frequency signal group respectively to obtain the initial high-frequency signal and the initial low-frequency signal.

[0085] S102-8-4-6 and S102-8-4-8 include: dividing the decomposed signal into two groups based on frequency: group l (low-frequency signal) and group h (high-frequency signal). The boundary between the high-frequency and low-frequency signals can be defined according to requirements, allowing for flexible adjustment of the reconstruction scheme. The signals within each group are summed to reconstruct a frequency-adjusted signal suitable for various frequency adjustment resources. After being divided into k groups, a portion is reconstructed into a low-frequency signal. Partially reconstructed into a high-frequency signal That is, Formula 4 can yield k sets of uk, where l+h=k.

[0086] Formula 5;

[0087] Formula 6;

[0088] It is the low-frequency signal (i.e., the initial high-frequency signal) after the decomposition of the scheduling command signal. It is the high-frequency signal (i.e. the initial low-frequency signal) after the decomposition of the scheduling command signal.

[0089] like Figure 6 The diagram shows the VMD decomposition of the AGC signal. The unit of these signals on the vertical axis is "AGC signal / MW", which is the amplitude of the automatic generation control signal, and the unit is megawatts (MW). The original AGC signal is decomposed into all the AGC components below, corresponding to the amplitudes of u1, u2...uk in Formulas 5 and 6.

[0090] Specifically, N power stations correspond to one high-frequency signal and one low-frequency signal. At this stage, the signal has not yet been distributed to each power station. After the first power allocation, the high-frequency signal is distributed among all supercapacitors, and the low-frequency signal is distributed among all lithium batteries. At this point, each energy storage system will receive the following signal: the high-frequency signal received by the supercapacitors in the energy storage system plus the low-frequency signal received by the lithium batteries. The combined signal is the active power signal distributed by the dispatch center to each power station. In the third power allocation, the signal distributed to the power stations will be further refined and allocated to the supercapacitors and lithium batteries.

[0091] S104: Optimize the initial high-frequency signal and the initial low-frequency signal based on the state of charge of the individual supercapacitor and the individual battery, respectively, to obtain the target high-frequency signal and the target low-frequency signal (equivalent to a second power allocation, corresponding to...). Figure 2 (S5 and S6).

[0092] S104 includes:

[0093] S104-2: Optimize the initial high-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual supercapacitors to obtain the target high-frequency signal.

[0094] S104-4: Optimize the initial low-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual cells to obtain the target low-frequency signal.

[0095] If we consider all supercapacitor banks as a whole and lithium battery banks as a whole, then each supercapacitor bank at each power station can be considered a supercapacitor unit, and each lithium battery bank at each power station can be considered a lithium battery unit. The overall SOC of the supercapacitor bank and the overall SOC of the lithium battery bank are obtained by weighted averaging of these energy storage units. Taking lithium batteries as an example, the weighted calculation formula is as follows:

[0096] Formula 7;

[0097] in, It is the weighted average SOC of lithium battery packs from all sites. It is the SOC of the i-th lithium battery pack (lithium battery cell) at the power station. is the capacity of the lithium battery pack at the i-th power station, and N is the number of power stations.

[0098] Based on the same principle as Formula 7, the weighted average SOC of the supercapacitor banks at all stations was calculated. .

[0099] The charging power of supercapacitors and lithium batteries is defined as negative, and the discharging power as positive. When a supercapacitor or lithium battery is in the overcharge region, its maximum charging power is defined as follows: and This indicates that all values ​​are 0. The ideal operating point is set at 0.5 times the rated capacity.

[0100] When a supercapacitor or lithium battery is in the over-discharge region, its maximum charging power is equal to the difference between the ideal operating point and the current state of charge (SOC). When the supercapacitor or lithium battery is in other SOC regions, its maximum charging power is determined by the difference between the maximum SOC and the current SOC.

[0101] Energy storage devices cannot charge and discharge simultaneously, so the following constraints must be met:

[0102] Formula 8;

[0103] The allocation method applies when only supercapacitors and lithium batteries can respond to AGC signals. Here, all supercapacitors are considered as a whole, and all lithium batteries are considered as a whole.

[0104] During the charging process, the active power of the supercapacitor during discharge. With the active power of lithium battery discharge All are 0. The discharge process is the reverse of the charging process.

[0105] S104-2: Optimize the initial high-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual supercapacitors to obtain the target high-frequency signal (mainly based on formula 9).

[0106] Maximum charging power of supercapacitor as follows:

[0107] Formula 9;

[0108] in, This represents the current SOC value of the supercapacitor. This refers to the rated capacitance of the supercapacitor. This represents the maximum charge value of the capacitor. This represents the maximum charging power of the supercapacitor.

[0109] Specifically, in Formula 9, That is, the optimized target high-frequency signal. The state of charge is determined based on the initial high-frequency signal. Formula 9 is the process of optimizing the initial high-frequency signal to obtain the target low-frequency signal, which belongs to the second allocation process.

[0110] Eall refers to the sum of the remaining capacities of all current supercapacitors. Escrate is the rated capacity of the entire supercapacitor. Based on the SOC region division for supercapacitors, SOCmin is 0.1 and SOCmax is 0.9. For lithium batteries, SOCmin is 0.2 and SOCmax is 0.8.

[0111] Formula 7 is for combining Figure 5 Determine the SOC region of the overall supercapacitor and the overall lithium battery, and then use Formula 7 to constrain the maximum charging power of the overall energy storage group. Minimize the objective function to obtain the detailed power command values ​​Plt and Pht.

[0112] Figure 2 The contents of S4, S5, and S6 are as follows: first divide the frequency → limit the power by SOC → jointly solve the scheduling under the condition of minimizing cost.

[0113] S104-4: Optimize the initial low-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual cells to obtain the target low-frequency signal (mainly based on formula 10).

[0114] Maximum charging power of lithium battery :

[0115] Formula 10;

[0116] in, This represents the current SOC value of the lithium battery. The rated capacity of the lithium battery, This is the maximum charge value of the lithium battery. This refers to the maximum charging power of the lithium battery.

[0117] Specifically, in formula 10, That is, the optimized target low-frequency signal. The state of charge of the lithium battery is determined based on the initial low-frequency signal combined with Formula 7. Formula 10 is the process of optimizing the initial low-frequency signal to obtain the target low-frequency signal, which belongs to the second allocation process.

[0118] If it is a discharge process, then Formula 9 becomes:

[0119] Formula 11;

[0120] Equation 10 becomes:

[0121] Formula 12.

[0122] S106: Determine the charging and discharging power (corresponding to) of each individual supercapacitor based on the target high-frequency signal, the health state and state of charge of the individual supercapacitor. Figure 2 The high-frequency processing of S7 and S8, merging the third and fourth allocation processes.

[0123] like Figure 3 As shown, the control system includes N stations, each supercapacitor unit includes X individual supercapacitors, and each battery unit includes Y individual batteries. Different supercapacitor units have different numbers of individual supercapacitors, and different battery units have different numbers of individual batteries. Specifically, refer to... Figure 3 In the first station, X is x11, in the second station, X is x21, in the first station, Y is x12, and in the second station, Y is x22. The quantities can all be different.

[0124] S106 is the process of distributing the high-frequency signals, which are finely allocated by the field cluster dispatch center, to the supercapacitor energy storage units of each field station (third allocation), and then to the individual supercapacitors (fourth allocation). S108 is the process of processing low-frequency signals. Based on the current SOC and SOH of each supercapacitor group and each lithium battery group in all fields, the high-frequency signals are distributed among all supercapacitors, and the low-frequency signals are distributed among all lithium batteries.

[0125] Overview of the third allocation: After receiving the dispatch instruction, the hybrid energy storage power station will perform three fine allocations of the instruction based on the current SOC status of its configured power-type energy storage units (supercapacitors) and energy-type energy storage units (lithium batteries and solid oxide fuel cells), fully tapping the total regulation potential of multiple types of energy storage and completing the allocation of the dispatch instruction within the station.

[0126] S106 includes:

[0127] S106-2: Based on the health status and state of charge of the supercapacitor cells, the target low-frequency signal is decomposed into N2 first supercapacitor charging and discharging powers and distributed to the corresponding supercapacitor cells, where N2≤N (i.e., the third distribution).

[0128] S106-4: For each group of supercapacitor units, the charging and discharging power of the first supercapacitor is decomposed into X1 charging and discharging powers of the second supercapacitors according to the health state and state of charge of the individual supercapacitors and then allocated to the corresponding individual supercapacitors, where X1≤X (i.e., the fourth allocation).

[0129] S106-2 includes:

[0130] S106-2-2: Construct a supercapacitor cell sorting function based on the health state and charge state of the supercapacitor cells, sort the N groups of supercapacitor cells, and obtain the first supercapacitor sorting result.

[0131] S106-2-4: Based on the first supercapacitor sorting result and the preset first supercapacitor sorting threshold, determine the supercapacitor units participating in the charging and discharging action, wherein the number of supercapacitor units participating in the charging and discharging action is N2.

[0132] S106-2-6: Construct a power allocation function based on the health state and state of charge of the supercapacitor unit, decompose the target low-frequency signal into N2 first supercapacitor charging and discharging powers, and allocate them to the corresponding supercapacitor units.

[0133] For each group of supercapacitor cells, execute S106-4, which includes:

[0134] S106-4-2: Sort the X individual supercapacitors in the supercapacitor unit according to their health status and state of charge, and obtain the second supercapacitor sorting result.

[0135] S106-4-4: Based on the second supercapacitor sorting result and the preset second supercapacitor sorting threshold, determine the individual supercapacitors participating in the charging and discharging action, wherein the number of individual supercapacitors participating in the charging and discharging action is X1.

[0136] S106-4-6: Construct a power allocation function based on the health state and state of charge of a single supercapacitor, decompose the charging and discharging power of the first supercapacitor into X1 charging and discharging powers of the second supercapacitor, and allocate them to the corresponding single supercapacitors.

[0137] S108: Determine the charging and discharging power (corresponding to) of each individual battery based on the target low-frequency signal, the health status of the individual battery, and the state of charge. Figure 2 The low-frequency processing of S7 and S8).

[0138] S108 includes a high-frequency signal The process of allocating power to supercapacitor units follows the same method as the allocation of low-frequency signals among all lithium batteries, primarily based on the supercapacitor's current SOC (State of Charge) and SOH (State of Health). At this point, all supercapacitors and lithium batteries in each energy storage station have received their corresponding power commands. The sum of these commands, after power allocation, constitutes the overall dispatch command issued by the cluster dispatch center to the energy storage station.

[0139] S108 includes:

[0140] S108-2: Decompose the target low-frequency signal into N1 first battery charging and discharging powers according to the health status and state of charge of the battery cells and allocate them to the corresponding battery cells, where N1≤N.

[0141] S108-4: For each group of battery cells, the charging and discharging power of the first battery is decomposed into Y1 charging and discharging powers of the second battery according to the health status and state of charge of the individual cells and distributed to the corresponding individual cells, where Y1≤Y.

[0142] S108-2 includes:

[0143] S108-2-2: Construct a battery cell sorting function based on the health status and state of charge of the battery cells, sort the N groups of battery cells, and obtain the first battery sorting result.

[0144] S108-2-4: Based on the first battery sorting result and the preset first battery sorting threshold, determine the battery cells participating in the charging and discharging action, wherein the number of battery cells participating in the charging and discharging action is N1.

[0145] S108-2-6: Construct a power allocation function based on the health status and state of charge of the battery cells, decompose the target low-frequency signal into N1 first battery charging and discharging powers, and allocate them to the corresponding battery cells.

[0146] Specifically, the discharge direction of the energy storage system is assumed to be positive, and the charging direction is assumed to be negative.

[0147] S108-2-2 only includes the process of sorting N groups of lithium battery cells in the charging state. S108-2-2 specifically includes the following 1)-5):

[0148] 1) During charging, for Figure 3 The SOC optimization formula for N groups of lithium battery cells is as follows:

[0149] Formula 13;

[0150] in, The SOC value collected by any group of lithium battery cells at the initial moment. This refers to the current output power of any group of lithium battery cells, collected in real time. For any group of lithium battery cells, the transmission efficiency is... This represents the optimized SOC value for any set of lithium battery cells.

[0151] Applying Formula 13 to each group of lithium battery cells yields N units. .

[0152] The significance of Formula 13: Due to the short-board effect, this part uses the minimum SOC value of the energy storage unit in the power station energy storage group as the SOC value of the power station energy storage group (i.e., lithium battery unit).

[0153] 2) Constraining the SOC and transmission power P of each individual lithium battery cell in the lithium battery unit:

[0154] Formula 14;

[0155] Formula 15;

[0156] Where i represents the i-th lithium battery cell, and the maximum value of i is N. The above formula is obtained by optimization. , The output power of the current lithium battery cell is collected in real time (referring to the total low-frequency signal; a value greater than 0 indicates that the lithium battery is discharging; a value less than 0 indicates that the lithium battery is charging). For the first Rated power of each lithium battery energy storage unit.

[0157] Smax and Smin are the maximum and minimum SOC that a pre-obtained energy storage unit can achieve during operation, respectively, which are constraint values. This is the maximum SOC that a single lithium battery energy storage unit can achieve during operation; This is the minimum SOC that a single lithium battery energy storage unit can achieve during operation.

[0158] If the constraints are met, then execute step 3).

[0159] 3) Calculation ( Considering only energy storage units (sorting function)

[0160] Formula 16;

[0161] in, For lithium battery cells in such Figure 5 The points in the domain of SOC shown are The result obtained from Formula 13 , A negative value indicates that the lithium battery is currently charging.

[0162] More specifically, Generally, 0.5 is chosen because the lithium battery cell has the strongest charging and discharging capability at a value around 0.5.

[0163] 0.15 is the scaling factor for the SOC ranking function f1 of the energy storage units. It adapts to the operating range of SOC and controls the output value of the inverse hyperbolic tangent function within a reasonable range. If the factor is too small, f1 is not sensitive enough to SOC deviations from the midpoint of 0.5, and cannot effectively avoid unit actions in extreme SOC states. If the factor is too large, f1 will overly dominate the priority, causing units with good SOH but slightly deviating SOC to be overly restricted, reducing the energy storage regulation capability. 0.15 is an optimal value that balances SOC protection and regulation capabilities. 0.15 can be adjusted according to actual conditions.

[0164] 4) Calculation ( To consider the health status within the lithium battery cell Range and (Sorting judgment function for inconsistent cases)

[0165] Formula 17;

[0166] Where i represents the i-th group of lithium batteries, and the maximum value of i is N;

[0167] For the first Within each lithium battery cell Extremely poor;

[0168] For the first Within each lithium battery cell The lowest single cell and cell The highest single cell difference;

[0169] z is a preset threshold.

[0170] 5) Construct a sorting function for the capacitor units based on f1 and f2, and sort the N groups of supercapacitor units.

[0171] Formula 18;

[0172] Each lithium battery pack has its own F1 algorithm calculated for sorting, and N values ​​are obtained based on Equation 18. . Ranking Criterion Objective Function By function and lithium batteries The weighted average of lithium batteries is obtained, where b / c is the weight value, and is set as follows: The higher the lithium battery value, the lower the priority and the later it is ranked.

[0173] S108-2-4: Based on the first battery sorting result and the preset first battery sorting threshold, determine the battery cells participating in the charging and discharging action, wherein the number of battery cells participating in the charging and discharging action is N1.

[0174] S108-2-4, during the charging state, filters the sorted N groups of lithium battery cells to determine the number of cells participating in the action, defined as N1, and the number of cells not participating in the action, where N1 + y = N. Specifically, this includes:

[0175] Formula 18;

[0176] in, For the first Each lithium battery cell At all times The efficiency during charging and discharging. This represents the current target low-frequency signal (i.e., the processing result of Formula 10).

[0177] S108-2-6: Construct a power allocation function based on the health status and state of charge of the battery cells, decompose the target low-frequency signal into N1 first battery charging and discharging powers, and allocate them to the corresponding battery cells.

[0178] S108-2-6 includes the process of distributing charging power to the N1 group of lithium battery cells:

[0179] After sorting and selecting the participating lithium battery cells, the objective function F2 for power allocation is set as follows:

[0180] Formula 19;

[0181] Formula 20;

[0182] in, This is a flag bit for the lithium battery cell to participate in the action. It is 1 when the lithium battery cell is selected to participate in the action, and 0 otherwise. The maximum value of n is N.

[0183] The difference in SOC between the single cell with the lowest SOH and the single cell with the highest SOH in the first group of lithium battery cells;

[0184] Let M be the difference between the SOH of the single cell with the lowest SOH and the single cell with the highest SOH in the i-th lithium battery cell group. The maximum value of i is M.

[0185] This represents the current output power of the i-th lithium battery cell group. <0 indicates that the lithium battery cell is currently charging;

[0186] This represents the current charge level of the i-th lithium battery cell.

[0187] For the first Each lithium battery cell At all times The efficiency during charging;

[0188] In Formula 19, n represents n power stations (the control system has a total of N power stations), and in Formula 20, i represents the i-th lithium battery unit in any power station (any power station has M sets of lithium battery units). A power station can also have multiple lithium battery units. In this embodiment... Figure 3 One lithium battery unit is installed in each station, i.e., M=1.

[0189] Finally, the AMPSO adaptive hybrid particle swarm optimization algorithm is used to optimize the above-mentioned objective function and allocate the charging / discharging power P of each (N1 in total) lithium battery cell, thereby initially allocating the low-frequency signal to the lithium battery packs of the N1 energy storage power stations.

[0190] S108-4 includes:

[0191] S108-4-2: Sort the Y individual cells in the battery cell according to the health status and state of charge of the individual cells to obtain the second battery sorting result.

[0192] S108-4-4: Based on the second battery sorting result and the preset second battery sorting threshold, determine the individual battery cells participating in the charging and discharging action, wherein the number of individual battery cells participating in the charging and discharging action is Y1.

[0193] S108-4-6: Construct a power allocation function based on the health status and state of charge of individual cells, decompose the charging and discharging power of the first cell into Y1 charging and discharging powers of the second cell, and allocate them to the corresponding individual cells.

[0194] For each group of battery cells, S108-4 is executed. S108-4 is the same as S108-2 and will not be described again. S108-4 is the fourth allocation process, which is the process of allocating power to specific individual supercapacitors. Formulas 13-20 describe in detail the process of power allocation to N groups of lithium battery cells in the charging state (the third allocation process). For the power allocation to N groups of lithium battery cells in the discharging state and the power allocation to N groups of supercapacitor cells in the charging and discharging states, please refer to the following content.

[0195] Formula 101;

[0196] Constrained by the short-terminal battery within the energy storage unit, the short-terminal battery's... As the first The SOC value of each energy storage unit group. t is time, and t=1 is the SOC value at the first sampling time, meaning that the SOC value of the next energy storage unit is obtained based on the SOC value at the previous time after a charging or discharging process. Direct sampling is also calculated using basic formulas, which are given here. Formula 13 is for lithium batteries. Formula 101 is for capacitors, and it also includes charging and discharging conditions.

[0197] Time of the first SOC and transmission power of each energy storage unit group The following formula must be satisfied:

[0198] (9) Formula 102

[0199] (10) Formula 103

[0200] Smax and Smin are the maximum and minimum SOC that a single energy storage unit can achieve during operation, respectively, which are constraint values. No. Rated power of each lithium battery energy storage unit; No. The transmission power of each lithium battery energy storage unit. Figure 2 The S7 includes:

[0201] S7-1: (1) Divide a low-frequency signal Plt (the precise Plt obtained after steps S5 and S6) into N groups and distribute them to lithium battery packs in N different sites.

[0202] a. Sort the N groups of lithium battery packs and determine the number of them participating in the action, defined as x, and the number of them not participating in the action as y, where x + y = N.

[0203] The lithium battery clusters involved in the operation should meet the following requirements:

[0204] ,Formula 104

[0205] in, For the first Each lithium battery energy storage unit in At all times with power By calculating the charging and discharging efficiency, the optimal power distribution for the current operating cycle can be obtained. The ranking criterion function F1 for the lithium battery energy storage groups is as follows:

[0206] ;

[0207] Formula 105;

[0208] Formula 106;

[0209] The maximum SOC that a single lithium battery energy storage unit can achieve during operation;

[0210] The minimum SOC that a single lithium battery energy storage unit can achieve during operation;

[0211] No. Within each lithium battery cell group Extremely poor;

[0212] For the first Within each lithium battery energy storage unit The lowest single cell and group The highest single cell difference.

[0213] Ranking Criterion Objective Function By function and Weighted average and Set two weight values. The larger the value, the lower the priority.

[0214] This section only provides the sorting criterion function, which is used to sort the energy storage groups (between supercapacitor groups and between lithium battery groups) at the power station. The participating energy storage groups must meet certain criteria. Formula 107 can respond to the signal (the selected supercapacitor group responds to the signal). ).

[0215] Discharge process: Equation 105 is changed to Equation 108:

[0216] Formula 108;

[0217] Formula 106 is changed to Formula 109:

[0218] Formula 109

[0219] b. S7-2 allocates charging power to x groups of lithium batteries.

[0220] After sorting and selecting the participating lithium battery packs, the objective function for power allocation is set as follows:

[0221] Formula 110

[0222] It can be derived from the following formula:

[0223] Formula 111;

[0224] The flag bit for the energy storage unit to participate in the action is 1 when the unit is selected to participate in the action, and 0 otherwise.

[0225] Finally, the AMPSO adaptive hybrid particle swarm optimization algorithm is used to optimize the above-mentioned objective function and allocate the charging / discharging power P of each (N in total) lithium battery pack. In this way, the low-frequency signal is initially allocated to the lithium battery packs of the N energy storage power stations.

[0226] No. Within each lithium battery cell group Range

[0227] For the first Within each lithium battery energy storage unit The lowest single cell and group The highest single cell difference

[0228] Formula 110 is the set objective function for power distribution. For this function, when the lithium battery pack... When the value is small, the function value is small; when the value is small, the function value is small. As the value gradually increases, the function value increases rapidly, limiting the height. The lithium battery pack participates in power regulation to mitigate further degradation. After setting the objective function, the AMPSO adaptive hybrid particle swarm optimization algorithm is used to optimize power allocation.

[0229] In summary: F1 sorts the lithium battery packs and selects the participating lithium battery packs, while F2 is the weighting function for the lithium battery packs' participation in power regulation (that is, the total power command Plt is allocated to the lithium battery packs according to this weight). It does not directly represent Plt. Using the AMPSO adaptive hybrid particle swarm optimization algorithm to optimize F2 will yield the power allocation value of the lithium battery packs.

[0230] Discharge process: Equation 111 is changed to Equation 112:

[0231] Formula 112.

[0232] Discharge process of energy storage power station

[0233] Similar to the above process, first determine whether the current energy storage capacity of supercapacitors and lithium batteries can meet the existing dispatch command requirements;

[0234] 1. If the energy storage capacity of the supercapacitor and lithium battery can meet the dispatch command requirements, power allocation commands for the lithium battery and supercapacitor will be issued similarly to those in sections (S4, S5, S6, S7, S8). This will not be repeated here, but the following changes are required:

[0235] Equation (8) becomes:

[0236] (twenty two)

[0237] Formula at S5:

[0238] When the energy storage system is in the overcharge region, the optimal power allocation commands for the supercapacitor and lithium-ion battery are respectively used... and This means that the value is equal to the difference between the ideal operating point and the current SOC, with the ideal operating point set at 0.5 times the rated capacity. When the energy storage system is in the over-discharge region, the maximum discharge power of both the supercapacitor and the lithium-ion battery is 0. When the energy storage system is in other SOC regions, the maximum discharge power of the supercapacitor and the lithium-ion battery is determined by the difference between the maximum SOC and the current SOC. The formula is modified as follows:

[0239] Equation (6) becomes:

[0240] (twenty three)

[0241] Equation (7) becomes:

[0242] (twenty four)

[0243] Formula at S7-1:

[0244] Formula (11) is changed to formula (25):

[0245] (25)

[0246] Formula (12) is changed to formula (26):

[0247] (26)

[0248] Formula at S7-2:

[0249] Formula (14) is changed to formula (27):

[0250] (27).

[0251] S110: Determine whether the charging and discharging power of each supercapacitor and the charging and discharging power of each battery cell meet the preset charging and discharging constraints, power balance constraints, and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem.

[0252] S110 includes:

[0253] The power balance constraint is determined based on the operating status of the energy substitute subsystem, the charging and discharging power of each supercapacitor, and the charging and discharging power of each battery cell.

[0254] S110 also includes:

[0255] The investment cost is determined based on the charging and discharging power of each individual supercapacitor and each individual battery. A cost function is constructed based on the investment cost, a predetermined operation and maintenance cost, and a predetermined scrapping cost. A preset cost threshold is used to determine whether the cost constraints are met.

[0256] S110 includes:

[0257] S110-2: Charge and discharge power constraints of the energy storage subsystem:

[0258] Formula 21

[0259] Formula 22;

[0260] in, This refers to the rated power of the supercapacitor. This refers to the rated power of a single battery cell. This refers to the output power of the capacitor when it is charging. This represents the output power of the capacitor in its discharged state.

[0261] S110-4: Power balance constraints:

[0262] Formula 23;

[0263] in, These are the total command that the energy storage system needs to complete (i.e., P0), the current output power of the SOFC fuel cell, the current output power of the supercapacitor (i.e., the result of S106), the current output power of the lithium battery (i.e., the result of S108), and the current output power of the PEM electrolyzer. The SOFC and PEM constitute the energy substitute subsystem.

[0264] S110-6: Cost Constraints

[0265] 1) Investment Costs:

[0266] Initial investment cost of energy storage (ES) equipment This includes various types of capacity costs and electricity costs, which are considered one-time investment costs.

[0267] Formula 24;

[0268] in, The investment cost per unit power of lithium batteries. The investment cost per unit power of a supercapacitor. The investment cost per unit capacity of lithium batteries. The investment cost per unit capacity of a supercapacitor. This is the power signal in response to the lithium battery. This is the power signal in response to the supercapacitor. The capacity value required for the lithium battery to respond to the signal. This is the capacitance value required for the supercapacitor to respond to the signal.

[0269] in, =The subdivided Plt (i.e., the result of S108). Pht (i.e., the result of S106) is obtained by minimizing the cost objective function Fc.

[0270] 2) Operating and maintenance costs = Formula 25;

[0271] The operation and maintenance phase includes maintenance costs for various types of energy storage. and replacement costs Replacement costs primarily refer to lithium-ion batteries, which are related to their lifespan. The specific calculation is shown in the following formula:

[0272] Formula 26;

[0273] in, and The unit price for capacity maintenance is for lithium-ion batteries and supercapacitors, respectively. This is the discount rate, typically 0.08. For conversion factors, For the entire life cycle, The number of battery replacements throughout its entire lifespan.

[0274] 3) Costs of scrapping

[0275] Supercapacitors have a fast discharge rate, a high number of discharge cycles, a long cycle life, and slow performance degradation. Supercapacitors typically have a lifespan of >10-20 years and may not need replacement during the system's lifespan. Battery energy storage, on the other hand, has a limited lifespan and requires multiple replacements throughout its lifespan. The disposal and recycling of end-of-life energy storage devices incurs certain costs. The calculation of end-of-life battery energy storage disposal costs is as follows:

[0276] Formula 27;

[0277] and These represent the unit price for processing lithium-ion batteries based on their power and capacity, respectively.

[0278] 4) The objective function mainly includes the relevant costs of the system, which include the total lifecycle costs of various types of energy storage (the power command values ​​Plt and Pht are obtained by minimizing the cost objective function).

[0279] Formula 28;

[0280] Indicates the total investment cost of energy storage, This represents the total operation and maintenance cost of energy storage. This represents the total cost of disposing of energy storage devices when they are decommissioned.

[0281] If Fc is greater than the preset cost threshold, it is considered that the cost constraint condition is not met.

[0282] S112: If all constraints are met, the charging and discharging power of the single supercapacitor determined in S106 and the charging and discharging power of the single battery determined in S108 shall be used as the final charging and discharging scheme.

[0283] S112 also includes:

[0284] If the charging and discharging power of each supercapacitor and the charging and discharging power of each battery do not meet any of the preset charging and discharging constraints, power balance constraints, and cost constraints, then the operating status of the energy substitute subsystem and / or the renewable energy subsystem is adjusted, and the process returns to S102.

[0285] Specifically, S112 includes adjusting the output power of wind and solar power in the renewable energy subsystem, or adjusting the output power of SOFC and / or PEM in the energy substitute subsystem, if the constraint conditions are not met, and then returning to S102 to repeat the entire regulation process.

[0286] Appendix: Variable Description

[0287] The set of IMF (Intrinsic Mode Function) components

[0288] The set of center frequencies

[0289] This represents the original automatic gain control (AGC) signal before decomposition.

[0290] Lagrange multiplier operators

[0291] Low-frequency signal after decomposition of dispatch command signal

[0292] High-frequency signal after decomposition of scheduling command signal

[0293] Maximum charging power of supercapacitors

[0294] Maximum discharge power of supercapacitor

[0295] Maximum charging power of lithium batteries

[0296] Maximum charging power of lithium batteries

[0297] Lithium battery capacity

[0298] Capacitance of supercapacitors

[0299] No. The transmission power of each lithium battery energy storage unit

[0300] No. Rated capacity of each lithium battery energy storage unit

[0301] Sampling time interval

[0302] No. Each lithium battery energy storage unit in At all times with power Efficiency during charging and discharging

[0303] Lithium battery SOH

[0304] No. SOC value of a lithium battery energy storage unit

[0305] Maximum SOC achievable by a single lithium battery energy storage unit during operation

[0306] Minimum SOC achievable by a single lithium battery energy storage unit during operation

[0307] No. Rated power of each lithium battery energy storage unit

[0308] The midpoint of SOC is around 0.5.

[0309] No. Within each lithium battery cell group Range

[0310] For the first Within each lithium battery energy storage unit The lowest single cell and group The highest single cell difference

[0311] Imbalance threshold

[0312] weight values ​​of b and c

[0313] The flag bit for energy storage unit participation in the operation is set to 1 when the unit is selected to participate in the operation, and 0 otherwise.

[0314] Within the scheduling period maximum value

[0315] lithium battery rated capacity

[0316] The current maximum usable energy of a supercapacitor (which can be calculated using capacitance and voltage).

[0317] The nominal maximum energy of a supercapacitor (energy at the maximum rated voltage).

[0318] Investment costs

[0319] Operation and maintenance costs

[0320] Disposal costs

[0321] Rated power of supercapacitor

[0322] Supercapacitor charging power

[0323] Supercapacitor discharge power

[0324] Rated power of lithium batteries

[0325] Lithium battery charging power

[0326] lithium battery discharge power

[0327] Charging efficiency of energy storage systems

[0328] Discharge efficiency of energy storage system

[0329] The general instructions that the energy storage system needs to complete

[0330] Instructions required by SOFC

[0331] Instructions required for supercapacitors

[0332] Instructions required for lithium batteries

[0333] The beneficial effects of the embodiments of this application are as follows:

[0334] 1. This application provides a two-level collaborative scheduling method applicable to large-scale new energy power plant clusters containing multiple independent hybrid energy storage power stations, aiming to fill the gap in the existing technology of collaborative scheduling of large-scale new energy power plant clusters; it aims to achieve a systematic and refined response to the overall grid dispatching command, and by optimizing the power distribution among different power plants and heterogeneous energy storage units, give full play to the technical advantages of various energy storage technologies, and ultimately achieve the comprehensive goal of improving the overall dispatchability of new energy bases, significantly extending the operating life of energy storage systems, and reducing their total life cycle cost.

[0335] 2. The present invention, "A Heterogeneous Energy Storage Cluster Collaborative Control Method for Power Grid Stability Support," involves four power allocation processes: First allocation: roughly determining one high-frequency signal Pht' and one low-frequency signal Plt'. Second allocation: refining the results of the first allocation to obtain one high-frequency signal Pht and one low-frequency signal Plt. Third allocation: splitting Pht into N parts and distributing them to supercapacitors in different power stations, resulting in P11 / P21 / PN1; splitting Plt into N parts and distributing them to batteries in different power stations, resulting in P12 / P22 / PN2. Fourth allocation: specifically determining... Figure 3 The power allocation to each of the x11 supercapacitor energy storage units is determined by dividing P11 into x11 parts (other details are omitted). This four-stage allocation yields a more accurate operating power.

[0336] 3. After receiving the dispatch command, the hybrid energy storage power station will perform four fine-grained allocations of the command based on the current SOC status of its configured power-type energy storage units (supercapacitors) and energy-type energy storage units (lithium batteries and solid oxide fuel cells), fully tapping the total regulation potential of multiple types of energy storage and completing the allocation of the dispatch command within the station.

[0337] 4. Based on this method, the energy storage power station can efficiently complete the charging process, ensuring that the supercapacitor and lithium battery maintain good health and a long service life while meeting dispatch instructions.

[0338] 5. This application fully proposes a collaborative scheduling method for heterogeneous energy storage clusters to support grid stability, aiming to resolve the inherent contradictions in performance and cost among different energy storage technologies. This invention can optimally allocate the overall dispatch commands issued by the grid within the energy storage cluster and provide refined guidance for the collaborative operation of heterogeneous energy storage units (such as supercapacitors (SC), solid oxide fuel cells (SOFC), lithium-ion batteries, etc.) within each site, thus completing a collaborative scheduling architecture at the heterogeneous energy storage cluster level.

[0339] 6. The innovation of this application is mainly reflected in the following aspects: The first allocation pre-allocates the AGC signal into high-frequency and low-frequency signals. Through the second allocation, based on the state of charge (SOC) partitioning strategy, the high-frequency and low-frequency signals are optimized and corrected. The third and fourth power allocations consider the limitation of the lowest SOC of individual cells within the group and the range of SOH within the group. Subsequently, considering the system safety margin, the optimal values ​​of the super lithium battery and supercapacitor energy storage groups in the power station are obtained based on the AMPSO adaptive hybrid particle swarm optimization algorithm. The fourth power allocation is when the energy storage cells receive the commands that require a response. Through the above-mentioned collaborative scheduling mechanism, this invention significantly improves the overall dispatchability and economy of large-scale heterogeneous energy storage systems, and optimizes the whole life cycle cost while ensuring the stable operation of the power grid.

[0340] 7. This application proposes a collaborative scheduling method for heterogeneous energy storage clusters to support grid stability, aiming to systematically resolve the inherent contradictions in performance and cost among different energy storage technologies. Existing technologies are mostly limited to the internal control of a single hybrid energy storage power station, generally lacking a cluster-level collaborative scheduling architecture. This invention, by constructing a two-level scheduling framework, can optimally allocate the overall scheduling commands issued by the grid within the energy storage cluster, and provide refined guidance for the collaborative operation of heterogeneous energy storage units (such as supercapacitors (SC), solid oxide fuel cells (SOFC), lithium-ion batteries, etc.) within each station.

[0341] 8. In the first allocation of this application, the AGC signal is pre-allocated into high-frequency and low-frequency signals. The second allocation, based on a State of Charge (SOC) partitioning strategy, optimizes and corrects the high-frequency and low-frequency signals. The third and fourth power allocations consider the minimum SOC limit of individual cells within the group and the SOH range within the group. Subsequently, considering system safety margins, the optimal values ​​for the super lithium-ion battery and supercapacitor energy storage groups at the site are obtained based on the AMPSO adaptive hybrid particle swarm optimization algorithm. The fourth power allocation is when the energy storage cells receive the required response command. Through the above-mentioned collaborative scheduling mechanism, this invention significantly improves the overall dispatchability and economy of large-scale heterogeneous energy storage systems, achieving optimization of the entire lifecycle cost while ensuring stable grid operation.

[0342] Example 2

[0343] like Figure 7 As shown, this embodiment of the invention provides a power coordinated regulation device based on a heterogeneous energy storage cluster, the regulation device comprising:

[0344] The first control module is used to determine the initial high-frequency signal and the initial low-frequency signal based on user needs, the output power of the renewable energy subsystem, and the variational mode algorithm.

[0345] The second control module is used to optimize the initial high-frequency signal and the initial low-frequency signal based on the state of charge of the individual supercapacitor and the state of charge of the individual battery, respectively, to obtain the target high-frequency signal and the target low-frequency signal.

[0346] The third control module is used to determine the charging and discharging power of each individual supercapacitor based on the target high-frequency signal, the health status of the individual supercapacitor, and the state of charge.

[0347] The fourth control module is used to determine the charging and discharging power of each individual battery cell based on the target low-frequency signal, the health status of the individual battery cells, and the state of charge.

[0348] The fifth control module is used to determine whether the charging and discharging power of each supercapacitor and the charging and discharging power of each battery meet the preset charging and discharging constraints, power balance constraints, and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem.

[0349] The sixth control module is used to take the charging and discharging power of the single supercapacitor determined in S106 and the charging and discharging power of the single battery determined in S108 as the final charging and discharging scheme if all constraints are met.

[0350] The power coordinated regulation device based on heterogeneous energy storage clusters provided in this application has the same implementation principle and technical effect as the aforementioned power coordinated regulation method based on heterogeneous energy storage clusters. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0351] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A power coordinated regulation method based on heterogeneous energy storage clusters, characterized in that, The regulation method is applied to a power coordinated regulation system based on heterogeneous energy storage clusters. The regulation system includes multiple power stations and energy substitute subsystems. Each power station includes a set of renewable energy subsystems and a set of energy storage subsystems. Each set of energy storage subsystems includes a set of supercapacitor units and a set of battery units. Each set of supercapacitor units includes multiple individual supercapacitors, and each set of battery units includes multiple individual batteries. The control method includes: S102: Determine the initial high-frequency signal and the initial low-frequency signal based on user requirements, the output power of the renewable energy subsystem, and the variational mode algorithm; S104: Optimize the initial high-frequency signal and the initial low-frequency signal based on the state of charge of the individual supercapacitor and the state of charge of the individual battery, respectively, to obtain the target high-frequency signal and the target low-frequency signal; S106: Determine the charging and discharging power of each individual supercapacitor based on the target high-frequency signal, the health state and state of charge of the individual supercapacitor; S108: Determine the charging and discharging power of each individual battery cell based on the target low-frequency signal, the health status of the individual battery cell, and the state of charge. S110: Determine whether the charging and discharging power of each supercapacitor and the charging and discharging power of each battery meet the preset charging and discharging constraints, power balance constraints, and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem. S112: If all constraints are met, the charging and discharging power of the single supercapacitor determined in S106 and the charging and discharging power of the single battery determined in S108 shall be used as the final charging and discharging scheme.

2. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 1, characterized in that, S102 includes: S102-2: Determine the general dispatch instruction P0 by subtracting the pre-acquired user demand power from the output power of the renewable energy subsystem; S102-4: If the general scheduling instruction P0 is positive, then the energy storage subsystem is determined to be in a discharge state; S102-6: If the total scheduling instruction P0 is negative, then the energy storage subsystem is determined to be in a charging state; S102-8: Based on the charging and discharging state of the energy storage subsystem and the variational mode algorithm, the scheduling command P0 is decomposed and reconstructed to determine the initial high-frequency signal and the initial low-frequency signal.

3. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 2, characterized in that, If the energy storage subsystem is in a discharge state, then S102-8 includes: S102-8-2: Determine whether all current energy storage subsystems meet user needs based on the general scheduling instruction P0 and the energy storage capacity of individual supercapacitors and individual batteries in all current energy storage subsystems; S102-8-4: If the user's needs are met, the total scheduling instruction P0 is decomposed and reconstructed based on the variational mode algorithm to determine the initial high-frequency signal and the initial low-frequency signal; S102-8-6: If not satisfied, adjust the operating status of the energy replenishment subsystem until the user's needs are met.

4. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 3, characterized in that, If the energy storage subsystem is in a charging state, then S102-8-2 is replaced by: Based on the general scheduling instruction P0, the energy storage capacity of each supercapacitor and battery in all current energy storage subsystems, the system determines whether all current energy storage subsystems meet user requirements.

5. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 3, characterized in that, S102-8-4 includes: S102-8-4-2: Based on the pre-acquired variational mode decomposition parameters, the total scheduling instruction P0 is decomposed into multiple solid-state model components under constraint states; S102-8-4-4: Optimize the solid-state model components based on the pre-acquired quadratic penalty factor α and Lagrange multiplier operators to determine the unconstrained frequency conversion model components; S102-8-4-6: Based on a preset frequency threshold, the components of the unconstrained frequency conversion model are divided into a high-frequency signal group and a low-frequency signal group; S102-8-4-8: Summate the frequency conversion model components in the high-frequency signal group and the low-frequency signal group respectively to obtain the initial high-frequency signal and the initial low-frequency signal.

6. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 5, characterized in that, S104 includes: S104-2: Optimize the initial high-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual supercapacitors to obtain the target high-frequency signal; S104-4: Optimize the initial low-frequency signal based on the pre-acquired state of charge and charge threshold range of all current individual cells to obtain the target low-frequency signal.

7. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 6, characterized in that, The control system includes N stations, each supercapacitor unit includes X individual supercapacitors, and each battery unit includes Y individual batteries. Different supercapacitor units include different numbers of individual supercapacitors, and different battery units include different numbers of individual batteries. S108 includes: S108-2: Decompose the target low-frequency signal into N1 first battery charging and discharging powers according to the health status and state of charge of the battery cells and allocate them to the corresponding battery cells, where N1≤N; S108-4: For each group of battery cells, the charging and discharging power of the first battery is decomposed into Y1 charging and discharging powers of the second battery according to the health status and state of charge of the individual cells and distributed to the corresponding individual cells, where Y1≤Y.

8. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 7, characterized in that, S108-2 includes: S108-2-2: Construct a battery cell sorting function based on the health status and state of charge of the battery cells, sort the N groups of battery cells, and obtain the first battery sorting result; S108-2-4: Based on the first battery sorting result and the preset first battery sorting threshold, determine the battery cells participating in the charging and discharging action, wherein the number of battery cells participating in the charging and discharging action is N1; S108-2-6: Construct a power allocation function based on the health status and state of charge of the battery cells, decompose the target low-frequency signal into N1 first battery charging and discharging powers, and allocate them to the corresponding battery cells.

9. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 8, characterized in that, For each group of battery cells, execute S108-4, which includes: S108-4-2: Sort the Y individual cells in the battery cell according to the health status and state of charge of the individual cells to obtain the second battery sorting result; S108-4-4: Based on the second battery sorting result and the preset second battery sorting threshold, determine the individual battery cells participating in the charging and discharging action, wherein the number of individual battery cells participating in the charging and discharging action is Y1; S108-4-6: Construct a power allocation function based on the health status and state of charge of individual cells, decompose the charging and discharging power of the first cell into Y1 charging and discharging powers of the second cell, and allocate them to the corresponding individual cells.

10. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 9, characterized in that, S106 includes: S106-2: Based on the health status and state of charge of the supercapacitor cells, the target low-frequency signal is decomposed into N2 first supercapacitor charging and discharging powers and distributed to the corresponding supercapacitor cells, where N2≤N; S106-4: For each group of supercapacitor units, the charging and discharging power of the first supercapacitor is decomposed into X1 charging and discharging powers of the second supercapacitors according to the health state and state of charge of the individual supercapacitors and distributed to the corresponding individual supercapacitors, where X1≤X.

11. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 10, characterized in that, S106-2 includes: S106-2-2: Construct a supercapacitor cell sorting function based on the health state and charge state of the supercapacitor cells, sort N groups of supercapacitor cells, and obtain the first supercapacitor sorting result. S106-2-4: Based on the first supercapacitor sorting result and the preset first supercapacitor sorting threshold, determine the supercapacitor units participating in the charging and discharging action, wherein the number of supercapacitor units participating in the charging and discharging action is N2. S106-2-6: Construct a power allocation function based on the health state and state of charge of the supercapacitor unit, decompose the target low-frequency signal into N2 first supercapacitor charging and discharging powers, and allocate them to the corresponding supercapacitor units.

12. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 11, characterized in that, For each group of supercapacitor cells, execute S106-4, which includes: S106-4-2: Sort the X individual supercapacitors in the supercapacitor unit according to the health state and state of charge of the individual supercapacitors to obtain the second supercapacitor sorting result. S106-4-4: Based on the second supercapacitor sorting result and the preset second supercapacitor sorting threshold, determine the individual supercapacitors participating in the charging and discharging action, wherein the number of individual supercapacitors participating in the charging and discharging action is X1. S106-4-6: Construct a power allocation function based on the health state and state of charge of a single supercapacitor, decompose the charging and discharging power of the first supercapacitor into X1 charging and discharging powers of the second supercapacitor, and allocate them to the corresponding single supercapacitors.

13. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 1, characterized in that, S110 includes: The power balance constraint is determined based on the operating status of the energy substitute subsystem, the charging and discharging power of each supercapacitor, and the charging and discharging power of each battery cell.

14. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 13, characterized in that, S110 also includes: The investment cost is determined based on the charging and discharging power of each of the individual supercapacitors and the charging and discharging power of each of the individual batteries. Construct a cost function based on investment costs, predetermined operation and maintenance costs, and predetermined scrapping costs; The preset cost threshold is used to determine whether the cost constraints are met.

15. The power coordinated regulation method based on heterogeneous energy storage clusters according to claim 14, characterized in that, S112 also includes: If the charging and discharging power of each supercapacitor and the charging and discharging power of each battery do not meet any of the preset charging and discharging constraints, power balance constraints, and cost constraints, then the operating status of the energy substitute subsystem and / or the renewable energy subsystem is adjusted, and the process returns to S102.

16. A power coordinated regulation device based on heterogeneous energy storage clusters, characterized in that, The control device includes: The first control module is used to determine the initial high-frequency signal and the initial low-frequency signal based on user needs, the output power of the renewable energy subsystem and the variational mode algorithm. The second control module is used to optimize the initial high-frequency signal and the initial low-frequency signal based on the state of charge of the individual supercapacitor and the state of charge of the individual battery, respectively, to obtain the target high-frequency signal and the target low-frequency signal. The third control module is used to determine the charging and discharging power of each individual supercapacitor based on the target high-frequency signal, the health status and state of charge of the individual supercapacitor. The fourth control module is used to determine the charging and discharging power of each individual battery cell based on the target low-frequency signal, the health status of the individual battery cells, and the state of charge. The fifth control module is used to determine whether the charging and discharging power of each supercapacitor and the charging and discharging power of each battery meet the preset charging and discharging constraints, power balance constraints, and cost constraints based on the operating status of the energy substitute subsystem and the renewable energy subsystem. The sixth control module is used to take the charging and discharging power of the single supercapacitor determined in S106 and the charging and discharging power of the single battery determined in S108 as the final charging and discharging scheme if all constraints are met.

Images