Hybrid energy storage participated secondary frequency modulation capacity optimization configuration method based on IHHO-VMD

Abstract

The invention discloses an IHHO-VMD-based hybrid energy storage participation secondary frequency modulation capacity optimization configuration method, which comprises the steps of optimizing a parameter combination in a variational mode decomposition algorithm VMD by using an improved Harlisia eagle optimization algorithm HHO, and adaptively adjusting VMD parameters according to change characteristics and complexity of signals, a power signal needing to be responded by hybrid energy storage is decomposed into high and low frequency components by using an IHHO-VMD method; in consideration of output characteristics of flywheel and lithium battery energy storage, the decomposed high and low frequency components are respectively distributed to the flywheel and the lithium battery for response, and are used as reference power of flywheel energy storage and lithium battery energy storage; establishing a capacity optimization configuration model of hybrid energy storage participating in power grid frequency modulation; an IHHO-VMD method is adopted to decompose and reconstruct an input hybrid energy storage frequency modulation ACE signal, rated power and rated capacity of flywheel energy storage and lithium battery energy storage are taken as optimization variables, and a capacity optimization configuration model of hybrid energy storage participating in power grid frequency modulation is solved. The method can effectively improve the overall economy and safety of the hybrid energy storage system participating in frequency modulation.

Description

Technical Field

[0001] The present invention relates to the technical field of hybrid energy storage system participating in power grid frequency regulation, and in particular to a capacity optimization configuration method of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD. Background Art

[0002] With the increasing prominence of climate change and energy shortage, vigorously developing new energy sources such as wind and solar is an important measure to promote the clean transformation of the energy system in order to achieve the "dual carbon" goal. Against the background of the rapid development of new power systems, the grid connection of a large number of intermittent and highly volatile new energy sources has brought great challenges to the frequency stability of the power system. Traditional thermal power units are difficult to meet the frequency regulation requirements due to slow regulation speed, limited capacity, dead zone oscillation and other problems in the process of responding to automatic generation control (AGC). Compared with traditional units, electrochemical energy storage has the advantages of fast response speed and high regulation accuracy. It can efficiently assist thermal power units in frequency regulation and promote the consumption of clean energy.

[0003] The rationality of the power command signal decomposition and allocation method is a prerequisite for achieving economical and stable operation of the system. Commonly used power signal decomposition methods include variational mode decomposition (VMD), wavelet transform, empirical mode decomposition (EMD), etc. However, EMD is prone to modal aliasing, the decomposition effect is poor, and signals with similar frequencies cannot be accurately separated. The effect of wavelet transform mainly depends on the selection of fundamental value, and its accuracy depends on the fluctuation of the original signal. VMD has poor signal decomposition effect due to the subjectivity of parameter selection. Using a suitable command signal decomposition and allocation method is a prerequisite for ensuring the economic stability of the system. Moreover, the cost-benefit analysis of energy storage participating in frequency regulation is not in-depth enough. Summary of the invention

[0004] The present invention introduces the IHHO-VMD method into the capacity configuration process of hybrid energy storage systems participating in grid frequency regulation, quantitatively analyzes the economic efficiency of hybrid energy storage participating in grid frequency regulation, and proposes an IHHO-VMD-based hybrid energy storage participating in secondary frequency regulation capacity optimization configuration method, which can effectively improve the overall economy and safety of hybrid energy storage systems participating in frequency regulation.

[0005] The technical solution adopted by the present invention is:

[0006] The method for optimizing the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD includes the following steps:

[0007] Step 1: Use the improved Harris Hawk optimization algorithm HHO to optimize the parameter combination in the variational mode decomposition algorithm VMD, adaptively adjust the VMD parameters according to the changing characteristics and complexity of the signal, and use the optimized IHHO-VMD method to decompose the power signal that the hybrid energy storage needs to respond to into high- and low-frequency components;

[0008] Step 2: Considering the output characteristics of the flywheel and lithium battery energy storage, the high and low frequency components after decomposition in step 1 are respectively allocated to the flywheel and lithium battery for response, and used as the reference power of the flywheel energy storage and lithium battery energy storage;

[0009] Step 3: Based on the full life cycle theory, establish a capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation;

[0010] Step 4: The IHHO-VMD method is used to decompose and reconstruct the input hybrid energy storage frequency regulation ACE signal, with the rated power and rated capacity of the flywheel energy storage and lithium battery energy storage as optimization variables, and then the improved Harris Eagle optimization algorithm HHO is integrated to assist in solving the capacity optimization configuration model of hybrid energy storage participating in grid frequency regulation.

[0011] In the step 1, the improved Harris Hawk optimization algorithm HHO is used to solve the parameter combination preset modal decomposition number K and quadratic penalty factor α in the variational mode decomposition algorithm VMD, that is, (K, α);

[0012] Harris Hawk Optimization Algorithm HHO uses mathematical formulas to simulate and optimize the strategies of capturing prey under different mechanisms. Harris Hawk Optimization Algorithm HHO mainly includes three stages: search stage, transformation stage and round-up stage.

[0013] During the search phase, Harris's hawk searches for food according to two strategies. The position update strategy is as follows:

[0014]

[0015] Where: t is the number of population iterations; X(t) and X(t+1) represent the positions of the population at the current iteration t and the next iteration t+1 respectively; X rand (t) is the random position of Harris Hawk population; X rabbit (t) is the prey position; X m (t) is the average position of all individuals in the population; r 1 、r 2 、r 3 、r 4 , q are random numbers in [0,1], ub and lb are the upper and lower limits of the population search space; the HHO algorithm realizes the transition from search to capture phase according to the escape energy O of the prey. When |O|>1, the HHO algorithm enters the search phase, otherwise it enters the capture phase.

[0016]

[0017] Where: T represents the maximum number of iterations; O 0 ∈[-1,1] represents the random initial energy of the prey.

[0018] When |O|<1, the prey is not physically strong enough, and the hawk group will enter the hunting stage. The HHO algorithm uses four strategies to simulate the attack behavior in the hunting stage according to the prey escape and Harris hawk chasing. λ represents the probability of prey escape. When λ≥0.5, the prey is captured, otherwise the hunting fails.

[0019] Soft Siege:

[0020] When 1>|O|≥0.5 and λ≥0.5, the prey has enough energy to try to break out, and its position is updated by the following equation:

[0021]

[0022] Where: △X(t) represents the difference between the current position of the prey and the eagle; X rabbit (t) represents the position vector of the prey; O is the prey escape energy; J represents the random jump intensity of the prey escape process; r is a random number between 0 and 1.

[0023] Hard surround:

[0024] When |O|<0.5 and λ≥0.5, it means that the prey does not have sufficient energy and cannot escape the encirclement. The expression is as follows:

[0025] X(t+1)=X rabbit (t)-O|△X(t)|;

[0026] Progressive rapid dive soft surround:

[0027] When 1>|O|≥0.5 and λ<0.5, the prey has enough energy and may successfully escape capture. At this time, the Levy flight function needs to be combined to update the position of the prey:

[0028]

[0029] Where: Levy is the flight function; D and S are the problem dimension and random vector required to solve Levy flight respectively; Z represents the eagle based on Levy flight mode; Y represents the evaluation index of the eagle's next action;

[0030] Progressive fast dive hard encirclement:

[0031] When |O|<0.5 and λ<0.5, the prey has insufficient energy and cannot escape the capture. At this time, the position is updated as follows:

[0032]

[0033] Where: f(X(t)) represents the value of the Levy flight function when the position of the eagle group at the current iteration t is substituted.

[0034]

[0035] Where: Levy(D) represents the D-dimensional Levy process.

[0036] Combining Cauchy mutation and adaptive weight strategies to improve HHO algorithm:

[0037] Combined with the Cauchy operator, the Cauchy distribution function mutation effect is used to optimize the global object. The standard Cauchy distribution function formula is as follows:

[0038]

[0039] Where: x represents a random variable;

[0040] The Cauchy distribution function decreases gently from the peak to both sides. After updating the position through Cauchy mutation, the Harris Hawk is constrained by the local extreme point and can jump out of the local extreme point. The mathematical model of Cauchy mutation updates the optimal solution through the following formula to obtain the current global optimal solution X best :

[0041] X′ best =X best +X best ×Cauchy(0,1);

[0042] Where: X′ best is the updated population position; Cauchy(0,1) represents the standard Cauchy distribution;

[0043] In order to improve the local mining ability of the algorithm, the neighborhood of the prey position is updated to find a better solution. By integrating the adaptive weight factor ω, the algorithm has better local mining ability. The adaptive weight formula and prey position update are shown in the following formula:

[0044]

[0045] X′ rabbit =ω×X rabbit ;

[0046] Where: T 0 is the current iteration number, X rabbit represents the position of the prey after iteration;

[0047] The fitness function is constructed, and the improved Harris Hawk optimization IHHO algorithm is used to optimize the key decomposition parameters of VMD. Envelope Entropy (EE) is used as the fitness function of IHHO. Envelope Entropy reflects the sparse characteristics of the signal. The coefficient feature is strong and the envelope entropy is small, while the coefficient feature is weak and the envelope entropy is large. The envelope information entropy is expressed in the following formula:

[0048]

[0049] Where: p i,j Represents the modal component signal; IMF i,j represents the envelope amplitude of the jth sampling point of the i-th modal component after VMD decomposition, N is the length of the modal component signal, and j is the normalized modal component envelope;

[0050]

[0051] Of which: IMF EE (i) represents the envelope entropy of the i-th modal component.

[0052] In step 1, the fitness function value and the prey energy O are calculated, the state of the prey energy O and the probability λ of the prey escaping is determined, different strategies are selected according to different states to update the position of the Harris Hawk population, the Harris Hawk individuals and the global optimal solution are updated, and the solution is stopped when the number of iterations or iteration accuracy is reached, and the optimal parameter combination (K, α) is output; then, the set (K, α) is substituted into the VMD method for signal decomposition and envelope entropy analysis. The power signal that the hybrid energy storage needs to respond to is decomposed into high-frequency and low-frequency components by the IHHO-VMD method.

[0053] In step 2, based on the AGC frequency modulation command signal collected and calculated by the data collection center of the AGC system during the frequency modulation period and the real-time output statistical data of the thermal power unit, the superimposed ACE command signal ACE(t) and the unit response ACE output signal ACE in the frequency modulation process are obtained. G (t), thereby determining the signal ACE that the hybrid energy storage needs to respond to HE (t).

[0054] Original ACE signal ACE(t) and hybrid energy storage response signal ACE HE (t), unit response signal ACE G (t) The relationship between the three is:

[0055] ACE(t)=ACE HE (t)+ACE G (t);

[0056] According to the actual characteristics of ACE signal, in order to meet the dual needs of power-type energy storage (characterized by high power density and fast charging speed) and energy-type energy storage (known for high energy density and relatively fast charging and discharging speed) in the frequency modulation process, it is necessary to perform nonlinear decomposition of the hybrid energy storage frequency modulation power command signal, which includes high-frequency power components and medium- and low-frequency energy components.

[0057] In step 2, the modal component IMF can be obtained by decomposing it using the IHHO-VMD method i (t) and residual R n (t), using the improved HHO optimized VMD method, the high and low frequency components of the hybrid energy storage response part after decomposition are reconstructed. The reconstructed high and low frequency components are responded through the power-type energy storage flywheel energy storage system, and the low frequency components are responded through the energy-type energy storage lithium battery system. Taking full advantage of the characteristics of these two energy storage forms, by setting the reconstruction point m, the frequency modulation requirements of the AGC signal can be responded quickly and accurately, and the decomposed mode is reconstructed. The decomposed high frequency component and low frequency component can be expressed as:

[0058]

[0059] ACE HE (t) = ACE h (t)+ACE l (t);

[0060] Among them: ACE h (t) is the reconstructed high frequency component; ACE l (t) is the reconstructed low-frequency component; IMF i (t) is the i-th mode obtained by VMD decomposition; ACE HE (t) represents the signal that the hybrid energy storage needs to respond to; m represents the reconstruction point; K represents the modal number; i represents the i-th modal component.

[0061] In step 3, the capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation includes:

[0062] 3.1: Economic compensation model for hybrid energy storage system participating in AGC frequency regulation service:

[0063]

[0064] Where: C J represents the economic compensation benefits of the hybrid energy storage system participating in the AGC frequency regulation service; J p It is a comprehensive index of frequency modulation performance; D i is the i-th AGC regulation depth, which is the contribution of the unit to the grid frequency stability in the i-th AGC frequency regulation. When the specified power is reached before receiving the next AGC instruction, Di =|P A -P Si |,P A Indicates the power value required for secondary frequency modulation, P Si Indicates the original power when the frequency modulation command is received. If the specified power is not reached, D i =|P Ei -P Si |,P Ei Indicates the power value when the unit receives the next frequency modulation instruction, Y AGC is the unit price of AGC frequency modulation performance compensation; n is the number of replacements, n = T LCC / T life , T life Is the service life of lithium battery.

[0065] 3.2: Initial investment cost model of hybrid energy storage system:

[0066]

[0067] Where: C 1 represents the initial investment cost of the hybrid energy storage system; and are the unit power costs of lithium battery energy storage and flywheel energy storage respectively; and are the rated powers of lithium battery energy storage and flywheel energy storage respectively; and are the unit capacity costs of lithium battery energy storage and flywheel energy storage respectively; and are the rated capacities of lithium battery energy storage and flywheel energy storage respectively; r is the discount rate, n is the number of replacements, n=T LCC / T life , T life is the service life of the lithium battery; T LCC Represents the entire life cycle.

[0068] 3.3: Hybrid energy storage system operation and maintenance cost model:

[0069]

[0070] Where: C 2 represents the operation and maintenance cost of the hybrid energy storage system; and They are the unit power operation and maintenance costs of lithium battery energy storage and flywheel energy storage respectively; and are the unit capacity operation and maintenance costs of lithium battery energy storage and flywheel energy storage respectively; W L (t) and W F (t) are the annual charge and discharge capacity of lithium battery energy storage and flywheel energy storage, respectively.

[0071] 3..4: Daily revenue of hybrid energy storage participating in grid frequency regulation C total The objective function is as follows:

[0072] max(C total )=max(C J -C 1 -C 2 );

[0073] Where: max(C total ) represents the maximum daily profit of hybrid energy storage participating in grid frequency regulation.

[0074] The constraints of the capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation specifically include:

[0075] ①. Energy storage charging and discharging power constraints:

[0076]

[0077] ②. Constraints of charge state on energy storage:

[0078]

[0079] Among them: SOC L (t) and SOC F (t) are the charge states of the lithium battery and flywheel energy storage at time t respectively; and They are respectively the low warning value and the high warning value of the lithium battery charge state; and are the low warning value and high warning value of the flywheel energy storage charge state respectively; η is the charge and discharge efficiency of energy storage; η L and η F are the charging and discharging efficiencies of lithium batteries and flywheel energy storage respectively; P S (t) is the power at time t; △t is the sampling interval; E is the energy storage capacity;

[0080] Indicates the minimum charging power of the lithium battery at time t; represents the maximum charging power of the lithium battery at time t, Represents the charging power of the lithium battery at time t; Indicates the minimum discharge power of the lithium battery at time t; represents the maximum discharge power of the lithium battery at time t, Represents the discharge power of the lithium battery at time t;

[0081] SOC(t) represents the state of charge of the energy storage at time t; SOC(t-1) represents the state of charge of the energy storage at time (t-1);

[0082] ③.Power balance constraints:

[0083] The power demand of hybrid energy storage frequency regulation is equal to the output power of lithium battery and flywheel energy storage at time t.

[0084] P(t)=P L (t)+P F (t);

[0085] Where: P(t) represents the hybrid energy storage frequency regulation power demand at time t.

[0086] The step 4 comprises the following steps:

[0087] (1). Parameter initialization, reconstruction point m is 0;

[0088] (2) Decompose the hybrid energy storage frequency modulation power command signal into K IMF components through IHHO-VMD;

[0089] (3) Set the charge and discharge power instructions of the flywheel and lithium battery according to the reconstructed high and low frequency component formulas, and use the maximum charge and discharge power and cumulative capacity of the flywheel and lithium battery as reference values ​​for the rated power and rated capacity respectively;

[0090] (4). Initialize population parameters;

[0091] (5) Calculate the fitness value of individuals in the population based on the objective function;

[0092] (6). Calculate the prey energy O;

[0093] (7) If |O|<1, adopt four strategies to capture prey and update the population position, otherwise perform the position update in the search phase;

[0094] (8) Calculate the fitness value of the function in step (7). If a better fitness value is obtained, update the global optimal solution and the global optimal fitness value;

[0095] (9) Determine whether the number of iterations or iteration accuracy has been reached. If so, output the optimal energy storage benefit and the corresponding hybrid energy storage configuration plan, and the algorithm ends; otherwise, return to step (4);

[0096] (10) Determine whether the reconstruction point m is equal to K. If not, repeat steps (3) to (9) until m=K. At this time, output the hybrid energy storage configuration and benefit at each reconstruction point m, and the algorithm ends.

[0097] The present invention provides a method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD, and the technical effects are as follows:

[0098] 1) In step 1 of the present invention, the parameter combination in the variational mode decomposition algorithm VMD is optimized by using the improved Harris Hawk optimization algorithm HHO, and the VMD parameters are adaptively adjusted according to the changing characteristics and complexity of the signal, thereby avoiding relying on subjective judgment to set the number of decomposition modes, and can improve the problems of modal aliasing and redundant decomposition of co-frequency signals to a certain extent, thereby improving the accuracy and results of the decomposition.

[0099] 2) Step 2 of the present invention takes into account the output characteristics of flywheel and lithium battery energy storage, and distributes the high and low frequency components after decomposition in step 1 to the flywheel and lithium battery for response, and serves as the reference power of flywheel energy storage and lithium battery energy storage. Since the power storage elements and energy storage elements of hybrid energy storage can complement each other in terms of usage characteristics, when the frequency fluctuates, the high-frequency components in the power system can be borne by flywheel energy storage with long cycle time and fast charge and discharge speed, and the medium and low frequency components can be borne by energy storage lithium batteries with high energy density and long charge and discharge time. The organic combination of the two energy storage methods can effectively make up for the defects of a single energy storage method, achieve the purpose of extending the service life of the energy storage unit, and improve the stability and economy of power system operation.

[0100] 3) Step 3 of the present invention is to establish a capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation based on the full life cycle theory. This can deeply and quantitatively analyze the benefits of energy storage participating in frequency regulation and judge the economic feasibility of energy storage participating in frequency regulation.

[0101] 4) In step 4 of the present invention, the IHHO-VMD method is used to decompose and reconstruct the input hybrid energy storage frequency modulation ACE signal, and the rated power and rated capacity of the flywheel energy storage and lithium battery energy storage are used as optimization variables, and then the improved Harris Eagle optimization algorithm HHO is integrated to assist in solving the capacity optimization configuration model. The use of an appropriate command signal decomposition and allocation method is a prerequisite for achieving economic and stable operation of the system. The use of an improved optimization algorithm to solve the capacity optimization configuration model can more quickly and accurately obtain the optimal configuration. Overall, the use of this step for capacity optimization configuration reduces the configuration power and capacity, and improves the overall economy of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0103] Figure 1 A flowchart of a method for optimizing the configuration of capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD provided by the present invention.

[0104] Figure 2 This is the ACE power signal curve that the hybrid energy storage system needs to respond to.

[0105] Figure 3This is the decomposition result diagram of IHHO-VMD for the hybrid energy storage demand power.

[0106] Figure 4 This is the spectrum diagram of each component after decomposition by IHHO-VMD algorithm.

[0107] Figure 5 This is the frequency spectrum of each component after decomposition using the EMD method.

[0108] Figure 6 This is the spectrum diagram of each component after decomposition using the VMD method. DETAILED DESCRIPTION

[0109] The capacity optimization configuration method of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD first uses the hybrid energy storage system composed of flywheel and lithium battery to assist thermal power units to quickly respond to and adjust the frequency fluctuations caused by the access of new energy sources such as wind and solar to the power grid. The improved Harris Hawks optimization (IHHO) algorithm is used to optimize the parameter combination in the variational mode decomposition (VMD) algorithm. The VMD parameters can be adaptively adjusted according to the change characteristics and complexity of the signal to obtain the optimized power signal components, thereby dividing the power signal that the hybrid energy storage needs to respond into high and low frequencies. Then, based on the output characteristics of the flywheel and lithium battery energy storage, the internal frequency modulation signal of the hybrid energy storage is reconstructed and allocated, and the high-frequency component is allocated to the flywheel energy storage for response, and the low-frequency component is allocated to the lithium battery for response. The high and low frequency components after decomposition are used as the reference power of the flywheel energy storage and the lithium battery energy storage, respectively. Finally, based on the full life cycle theory, a capacity optimization configuration model for hybrid energy storage participating in power grid frequency regulation is established, and the IHHO algorithm is used to solve and optimize to obtain the final optimization configuration result. The configuration scheme obtained by the hybrid energy storage participating in the secondary frequency regulation capacity optimization configuration method based on IHHO-VMD proposed in the present invention can effectively improve the overall economy and safety of the hybrid energy storage system.

[0110] like Figure 1 As shown, a method for optimizing the configuration of hybrid energy storage participating in secondary frequency regulation capacity based on IHHO-VMD includes the following steps:

[0111] Step (1): Optimize the parameter combination in the variational mode decomposition algorithm using the improved Harris Eagle optimization algorithm, adaptively adjust the VMD parameters according to the changing characteristics and complexity of the signal, and use the optimized IHHO-VMD method to decompose the power signal that the hybrid energy storage needs to respond to into high and low frequency components;

[0112] Define the intrinsic mode function (IMF) as:

[0113]

[0114] Among them: A k (t) is u k (t) is the instantaneous amplitude function; t is time; is the instantaneous phase function; u k (t) is the instantaneous frequency

[0115] The constrained variational mathematical model established by VMD is expressed as:

[0116]

[0117] Where: {u k}={u 1 ,…,u k} represents the corresponding component set; {ω k}={ω 1 ,…,ω k} represents the center frequency set; σ(t) is the pulse function; the extended Lagrange multiplier function is introduced to solve the global optimal solution of the above problem, namely:

[0118]

[0119] Where: α is the quadratic penalty factor, σ is the Lagrange operator. The alternating multiplier algorithm ADMM is used to solve the above equation and continuously update each modal component u k , center frequency ω k And Lagrange operator σ to obtain the optimal solution of the extended Lagrange multiplier function. All components can be obtained from the frequency domain by the following formula:

[0120]

[0121] Where: ω is the frequency, and the ones with subscripts are their corresponding Fourier transforms. The algorithm updates the center frequency according to the center of gravity of each modal power spectrum, that is:

[0122]

[0123] The intelligent algorithm is used to improve the Harris Eagle optimization algorithm to solve the parameter combination (K, α) in the VMD algorithm:

[0124] Harris Hawk Optimization algorithm uses mathematical formulas to simulate and optimize the strategy of capturing prey under different mechanisms. The HHO algorithm mainly includes three stages: search stage, transformation stage and round-up stage.

[0125] During the search phase, Harris's hawk searches for food using two strategies. The position update strategy is as follows:

[0126]

[0127] Where: t is the number of population iterations; X(t) and X(t+1) represent the positions of the population at the current iteration t and the next iteration t+1 respectively; X rand (t) is the random position of Harris Hawk population; X rabbit (t) is the prey position; X m (t) is the average position of all individuals in the population; r 1 、r 2 、r 3 、r 4 , q are both random numbers in [0,1], ub and lb are the upper and lower limits of the population search space.

[0128] The HHO algorithm realizes the transition from the search phase to the hunting phase according to the prey's escape energy O. When |O|>1, the HHO algorithm enters the search phase, otherwise it enters the hunting phase.

[0129]

[0130] Where: T represents the maximum number of iterations; O 0 ∈[-1,1] represents the random initial energy of the prey.

[0131] When |O| is less than 1, the prey is not physically strong enough and the hawk flock will enter the hunting stage. The HHO algorithm uses four strategies to simulate the attack behavior in the hunting stage according to the prey escape and Harris hawk chasing. λ is used to represent the probability of prey escape. When λ≥0.5, the prey is captured, otherwise the hunting fails.

[0132] Soft Siege:

[0133] When 1>|O|≥0.5 and λ≥0.5, the prey has enough energy to try to break out, and its position is updated by the following equation:

[0134]

[0135] Where: ΔX(t) represents the difference between the current position of the prey and the eagle, X rabbit (t) represents the position vector of the prey, O is the prey escape energy, J represents the random jump intensity of the prey escape process, and r is a random number between 0 and 1.

[0136] Hard surround:

[0137] When |O|<0.5 and λ≥0.5, it means that the prey does not have sufficient energy and cannot escape the encirclement. The expression is as follows:

[0138] X(t+1)=X rabbit (t)-O|△X(t)|;

[0139] Progressive rapid dive soft surround:

[0140] When 1>|O|≥0.5 and λ<0.5, the prey has enough energy and may successfully escape capture. At this time, the Levy flight function needs to be combined to update the position of the prey:

[0141]

[0142] Where: Levy is the flight function, D and S are the problem dimension and random vector required to solve Levy flight, Z represents the eagle based on Levy flight mode, and Y represents the evaluation index of the eagle's next action.

[0143] Progressive fast dive hard encirclement:

[0144] When |O|<0.5 and λ<0.5, the prey has insufficient energy and cannot escape the capture. At this time, the position is updated as follows:

[0145]

[0146] Combining Cauchy mutation and adaptive weight strategies to improve HHO algorithm:

[0147] Combined with the Cauchy operator, the Cauchy distribution function mutation effect is used to optimize the global object. The standard Cauchy distribution function formula is as follows:

[0148]

[0149] The Cauchy function decreases gently from the peak to both sides. After updating the position through Cauchy mutation, the Harris eagle is constrained by the local extreme point and can jump out of the local extreme point. The mathematical model of Cauchy mutation updates the optimal solution through the following formula to obtain the current global optimal solution X best :

[0150] X′ best =X best +X best ×Cauchy(0,1);

[0151] Where: X′ best is the updated population position.

[0152] In order to improve the local mining ability of the algorithm, the neighborhood of the prey position is updated to find a better solution. By integrating the adaptive weight factor ω, the algorithm has better local mining ability. The adaptive weight formula and prey position update are shown in the following formula:

[0153]

[0154] X′ rabbit =ω×X rabbit;

[0155] Where: T 0 is the current iteration number.

[0156] The fitness function is constructed, and IHHO is used to optimize the key decomposition parameters of VMD. Envelope Entropy (EE) is used as the fitness function of IHHO. Envelope Entropy reflects the sparse characteristics of the signal. The coefficient feature is strong and the envelope entropy is small, while the coefficient feature is weak and the envelope entropy is large. The envelope information entropy is expressed in the following formula:

[0157]

[0158] Of which: IMF i,j represents the envelope amplitude of the jth sampling point of the i-th modal component after VMD decomposition, and N is the modal component signal p i The length of , j is the normalized modal component envelope:

[0159]

[0160] Of which: IMF EE (i) represents the envelope entropy of the i-th modal component.

[0161] Step (2): Calculate the fitness function value and prey energy O, determine the state of O and λ, select different strategies according to different states to update the position of the Harris Hawk population, update the Harris Hawk individual and the global optimal solution, stop solving when the number of iterations or iteration accuracy is reached, output the optimal parameter combination (K, α), and then substitute the set (K, α) into the VMD method for signal decomposition and envelope entropy analysis. IHHO-VMD decomposes the power signal that the hybrid energy storage needs to respond to into high and low frequency components.

[0162] Step (3): Considering the output characteristics of the flywheel and lithium battery energy storage, the decomposed high and low frequency components are respectively allocated to the flywheel and lithium battery for response and used as the reference power of the flywheel energy storage and lithium battery energy storage;

[0163] Based on the AGC frequency regulation command signal collected and calculated by the data collection center of the AGC system during the frequency regulation period and the real-time output statistical data of the thermal power unit, the superimposed ACE command signal ACE(t) and the unit response ACE output signal ACE in the frequency regulation process are obtained. G (t), thereby determining the signal ACE that the hybrid energy storage needs to respond to HE (t). Original ACE signal ACE(t) and hybrid energy storage response signal ACE HE (t), unit response signal ACE G (t) The relationship between the three is:

[0164] ACE(t)=ACE HE (t)+ACE G (t);

[0165] According to the actual characteristics of ACE signal, in order to meet the dual needs of power-type energy storage (characterized by high power density and fast charging speed) and energy-type energy storage (known for high energy density and relatively fast charging and discharging speed) in the frequency modulation process, it is necessary to perform nonlinear decomposition of the hybrid energy storage frequency modulation power command signal, including high-frequency power components and medium- and low-frequency energy components. The modal component IMF can be obtained through IHHO-VMD decomposition. i (t) and residual R n (t), using the improved HHO optimized VMD method, the high and low frequency components of the decomposed hybrid energy storage response are reconstructed. The reconstructed high frequency components are responded through the power-type energy storage flywheel energy storage system, and the low frequency components are responded through the energy-type energy storage lithium battery system. By making full use of the characteristic advantages of these two energy storage forms, by setting the reconstruction point m, the frequency modulation requirements of the AGC signal can be responded quickly and accurately, and the decomposed mode can be reconstructed.

[0166] The decomposed high-frequency and low-frequency components can be expressed as:

[0167]

[0168] ACE HE (t) = ACE h (t)+ACE l (t);

[0169] Among them: ACE h (t) is the reconstructed high frequency component, ACE l (t) is the reconstructed low-frequency component, IMF i (t) is the i-th mode obtained by VMD decomposition.

[0170] Step (4): Establish a capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation based on the full life cycle theory;

[0171] Establish an economic compensation model for hybrid energy storage systems participating in AGC frequency regulation services:

[0172]

[0173] Where: J p is the comprehensive index of frequency modulation performance, D j is the jth AGC regulation depth, which is the contribution of the unit to the grid frequency stability in the jth AGC frequency regulation. When the specified power is reached before receiving the next AGC instruction, D i =|P A -PSi |, when the specified power is not reached D i =|P Ei -P Si |, Y AGC Compensation unit price for AGC frequency modulation performance.

[0174] Frequency modulation performance comprehensive index J p Including adjustment speed indicator J 1 , Adjustment accuracy index J 2 and response time index J 3 Three parts, J p The indicator calculation formula is as follows:

[0175] J p =J 1 iJ 2 iJ 3 ;

[0176]

[0177] Where: V is the average adjustment speed of the unit in response to AGC instructions during the operation period; V n is the standard regulation rate of the unit, and its value varies according to the type of unit. ΔP is the total deviation between the unit and the target value after reaching the target power during the AGC frequency regulation service period. P n is the rated capacity of the unit. n is the response time reference value, which is determined by experience according to the unit type, ΔT i is the response time after receiving the AGC command for the i-th time;

[0178] Establishing the initial investment cost model of hybrid energy storage system:

[0179]

[0180] Where: and are the unit power costs of lithium battery energy storage and flywheel energy storage, and are the rated powers of lithium battery energy storage and flywheel energy storage, and are the unit capacity costs of lithium battery energy storage and flywheel energy storage, and are the rated capacities of lithium battery energy storage and flywheel energy storage, r is the discount rate, n is the number of replacements, n = T LCC / T life , T life Is the service life of lithium battery.

[0181] Establishing the operation and maintenance cost model of hybrid energy storage system:

[0182]

[0183] Where: and are the unit power operation and maintenance costs of lithium battery energy storage and flywheel energy storage, and are the unit capacity operation and maintenance costs of lithium battery energy storage and flywheel energy storage, W L (t) and W F (t) are the annual charge and discharge capacity of lithium battery energy storage and flywheel energy storage, respectively.

[0184] Daily revenue of hybrid energy storage participating in grid frequency regulation C total The objective function is as follows:

[0185] max(C total )=max(C J -C 1 -C 2 );

[0186] The model constraints are specifically:

[0187] Energy storage charging and discharging power constraints:

[0188]

[0189] Constraints of state of charge on energy storage:

[0190]

[0191] Among them: SOC L (t) and SOC F (t) are the charge states of the lithium battery and flywheel energy storage at time t, and They are the low warning value and high warning value of the lithium battery charge state, and are the low warning value and high warning value of the flywheel energy storage charge state, η is the charge and discharge efficiency of energy storage, η L and η F are the charging and discharging efficiency of lithium battery and flywheel energy storage, P S (t) is the power at time t, △t is the sampling interval, and E is the energy storage capacity.

[0192] Power balance constraints:

[0193] The power demand of hybrid energy storage frequency regulation is equal to the output power of lithium battery and flywheel energy storage at time t.

[0194] P(t)=PL (t)+P F (t).

[0195] Step (5): The improved Harris Eagle optimization algorithm is used to solve and optimize to obtain the final capacity optimization configuration result; the IHHO-VMD decomposition method is used to decompose and reconstruct the input hybrid energy storage frequency modulation ACE signal, and the rated power and rated capacity of the flywheel energy storage and lithium battery energy storage are used as optimization variables, and the hybrid energy storage charging and discharging control strategy is substituted, and then the improved Harris Eagle optimization algorithm is integrated to assist in solving the model. The specific solution steps are as follows:

[0196] (1) Parameter initialization, the reconstruction point m is 0;

[0197] (2) Decompose the hybrid energy storage frequency modulation power command signal into K IMF components through IHHO-VMD;

[0198] (3) setting the charge and discharge power instructions of the flywheel and the lithium battery according to the reconstructed high and low frequency component formulas, and taking the maximum charge and discharge power and cumulative capacity of the flywheel and the lithium battery as reference values ​​of the rated power and rated capacity, respectively;

[0199] (4) Initialize population parameters;

[0200] (5) Calculate the fitness value of the individuals in the population based on the objective function;

[0201] (6) Calculate prey energy O;

[0202] (7) If |O|<1, adopt four strategies to capture prey and update the population position, otherwise perform the position update in the search phase;

[0203] (8) Calculate the fitness value of the function in step (7), and if a better fitness value is obtained, update the global optimal solution and the global optimal fitness value;

[0204] (9) Determine whether the number of iterations or the iteration accuracy has been reached. If so, output the optimal energy storage benefit and the corresponding hybrid energy storage configuration scheme, and the algorithm ends; otherwise, return to step (4);

[0205] (10) Determine whether the reconstruction point m is equal to K. If not, repeat steps (3) to (9) until m=K. At this time, output the hybrid energy storage configuration and benefit at each reconstruction point m, and the algorithm ends.

[0206] Example:

[0207] Taking the actual data provided by the power grid company as an example, the effectiveness of the method proposed in the present invention is analyzed and verified.

[0208] The actual operation data of a power grid company for one day is selected to verify the effectiveness of the method provided by the present invention, in which the ACE command signal that the hybrid energy storage needs to respond to is as follows: Figure 2 As shown. IHHO-VMD is used to decompose the ACE command signal that the hybrid energy storage needs to respond to. The population size N is set to 5, the number of iterations T is set to 20, and the spectrum analysis is performed on the decomposed components. The optimal parameter combination [K, α] = [7, 1524] can be obtained by optimizing the calculation of the IHHO algorithm.

[0209] Figure 3 The time domain waveform after IHHO-VMD decomposition, the spectrum analysis diagram of each component is as follows Figure 4 shown. Figure 4 , Figure 5 and Figure 6 The spectra are obtained by decomposing the method of the present invention and the EMD and VMD methods respectively. Figure 4 , Figure 5 and Figure 6 It can be seen that the signal decomposed by EMD shows obvious aliasing, which may cause large errors in subsequent analysis. VMD has the problem of redundant decomposition of signals with the same frequency, while IHHO-optimized VMD accurately decomposes the signal without any problems of over-decomposition or incomplete decomposition, effectively characterizing the basic components of the original signal. Therefore, the result of decomposition using IHHO-VMD is significantly better than that of EMD and VMD methods, and can better complete the reasonable distribution of power in the hybrid energy storage system by distinguishing high and low frequency components.

[0210] The present invention limits the flywheel energy storage conversion efficiency to 0.9 and the lithium battery conversion efficiency to 0.8. By using the improved Harris Eagle optimization algorithm to calculate the capacity configuration model, setting the population size N to 30 and the number of iterations T to 300, the capacity configuration and economic results of the lithium battery and flywheel energy storage can be obtained, as shown in Table 1.

[0211] Table 1 Hybrid energy storage capacity configuration results

[0212]

[0213] It can be seen that when the reconstruction point is too small, the frequency regulation component borne by the flywheel energy storage is very small, and its excellent frequency regulation performance cannot be well exerted, resulting in poor frequency regulation effect and excessive lithium battery energy storage configuration, resulting in increased costs and reduced system benefits. When the reconstruction point is too large, the flywheel energy storage bears more frequency regulation components. Although it can exert its fast-response frequency regulation characteristics, it increases the frequency regulation burden of the flywheel energy storage energy, which is not conducive to charge state management. With the change of the reconstruction point, the daily frequency regulation benefit shows a trend of first rising and then falling. When the reconstruction point is 3, the frequency regulation effect of the energy storage system is optimal and the benefit is maximized.

[0214] The present invention can effectively improve the economy and safety of the hybrid energy storage system participating in the overall frequency regulation.

Claims

1. The capacity optimization configuration method of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD is characterized by The following steps are involved: Step 1: Use the improved Harris Hawk optimization algorithm HHO to optimize the parameter combination in the variational mode decomposition algorithm VMD, adaptively adjust the VMD parameters according to the changing characteristics and complexity of the signal, and use the optimized IHHO-VMD method to decompose the power signal that the hybrid energy storage needs to respond to into high- and low-frequency components; Step 2: Considering the output characteristics of the flywheel and lithium battery energy storage, the high and low frequency components after decomposition in step 1 are respectively allocated to the flywheel and lithium battery for response, and used as the reference power of the flywheel energy storage and lithium battery energy storage; Step 3: Establish a capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation; Step 4: The IHHO-VMD method is used to decompose and reconstruct the input hybrid energy storage frequency regulation ACE signal, with the rated power and rated capacity of the flywheel energy storage and lithium battery energy storage as optimization variables, and then the improved Harris Eagle optimization algorithm HHO is integrated to assist in solving the capacity optimization configuration model of hybrid energy storage participating in grid frequency regulation.

2. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 1 is characterized by: In the step 1, the improved Harris Hawk optimization algorithm HHO is used to solve the parameter combination preset mode decomposition number K and quadratic penalty factor α in the variational mode decomposition algorithm VMD; The Harris Hawk Optimization Algorithm (HHO) consists of three phases: search phase, transformation phase, and roundup phase. In the search phase, the Harris Hawk searches for food according to two strategies. The position update strategy is as follows: Where: t is the number of population iterations; X(t) and X(t+1) represent the positions of the population at the current iteration t and the next iteration t+1 respectively; X rand (t) is the random position of Harris Hawk population; X rabbit (t) is the prey position; X m (t) is the average position of all individuals in the population; r1, r2, r3, r4, q are all random numbers in [0,1], ub, lb are the upper and lower limits of the population search space; the HHO algorithm realizes the transition from search to capture stage according to the escape energy O of the prey; when |O|>1, the HHO algorithm enters the search stage, otherwise it enters the capture stage; Where: T represents the maximum number of iterations; O0∈[-1,1] represents the random initial energy of the prey; When |O|<1, the prey is not physically strong enough, and the hawk group will enter the hunting stage. The HHO algorithm uses four strategies to simulate the attack behavior in the hunting stage according to the prey escape and Harris hawk chasing. λ represents the probability of prey escape. When λ≥0.5, the prey is captured, otherwise the hunting fails. Soft Siege: When 1>|O|≥0.5 and λ≥0.5, the prey has enough energy to try to break out, and its position is updated by the following equation: Where: △X(t) represents the difference between the current position of the prey and the eagle; X rabbit (t) represents the position vector of the prey; O is the prey escape energy; J represents the random jump intensity of the prey escape process; r is a random number between 0 and 1; Hard surround: When |O|<0.5 and λ≥0.5, it means that the prey does not have sufficient energy and cannot escape the encirclement. The expression is as follows: X(t+1)=X rabbit (t)-O|△X(t)|; Progressive rapid dive soft surround: When 1>|O|≥0.5 and λ<0.5, the prey has enough energy and may successfully escape capture; at this time, the Levy flight function needs to be combined to update the position of the prey: Where: Levy is the flight function; D and S are the problem dimension and random vector required to solve Levy flight respectively; Z represents the eagle based on Levy flight mode; Y represents the evaluation index of the eagle's next action; Progressive fast dive hard encirclement: When |O|<0.5 and λ<0.5, the prey has insufficient energy and cannot escape the capture. At this time, the position is updated as follows: Where: f(X(t)) represents the value of the Levy flight function when the position of the eagle group at the current iteration t is substituted; Where: Levy(D) represents the D-dimensional Levy process; Combining Cauchy mutation and adaptive weight strategies to improve HHO algorithm: Combined with the Cauchy operator, the global object is optimized using the mutation effect of the Cauchy distribution function; the standard Cauchy distribution function formula is as follows: Where: x represents a random variable; The Cauchy distribution function decreases gently from the peak to both sides. After updating the position through Cauchy mutation, the Harris Hawk is constrained by the local extreme point and can jump out of the local extreme value. The mathematical model of Cauchy mutation updates the optimal solution through the following formula to obtain the current global optimal solution X best : X′ best =X best +X best ×Cauchy(0,1); Where: X′ best is the updated population position; Cauchy(0,1) represents the standard Cauchy distribution; In order to improve the local mining ability of the algorithm, the neighborhood of the prey position is updated to find a better solution. By integrating the adaptive weight factor ω, the algorithm has better local mining ability; the adaptive weight formula and prey position update are shown in the following formula: X′ rabbit =ω×X rabbit ; Where: T0 is the current iteration number, X rabbit represents the position of the prey after iteration; The fitness function is constructed, and the improved Harris Hawk optimization IHHO algorithm is used to optimize the key decomposition parameters of VMD. The envelope entropy (EE) is used as the fitness function of IHHO. The envelope entropy reflects the sparse characteristics of the signal. The coefficient feature is strong and the envelope entropy is small, while the coefficient feature is weak and the envelope entropy is large. The envelope information entropy is expressed in the following formula: Where: p i,j Represents the modal component signal; IMF i,j represents the envelope amplitude of the jth sampling point of the i-th modal component after VMD decomposition, N is the length of the modal component signal, and j is the normalized modal component envelope; Of which: IMF EE (i) represents the envelope entropy of the i-th modal component.

3. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 2 is characterized by: In the step 1, the fitness function value and the prey energy O are calculated, the states of the prey energy O and the probability λ of the prey escaping are determined, different strategies are selected according to different states to update the position of the Harris Hawk population, the Harris Hawk individuals and the global optimal solution are updated, the solution is stopped when the number of iterations or the iteration accuracy is reached, and the optimal parameter combination (K, α) is output; then, the set (K, α) is substituted into the VMD method for signal decomposition, and envelope entropy analysis is performed; the power signal that the hybrid energy storage needs to respond to is decomposed into high-frequency and low-frequency components by the IHHO-VMD method.

4. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 1 is characterized by: In step 2, based on the AGC frequency modulation command signal collected and calculated by the data collection center of the AGC system during the frequency modulation period and the real-time output statistical data of the thermal power unit, the superimposed ACE command signal ACE(t) and the unit response ACE output signal ACE in the frequency modulation process are obtained. G (t), thereby determining the signal ACE that the hybrid energy storage needs to respond to HE (t); original ACE signal ACE(t) and hybrid energy storage response signal ACE HE (t), unit response signal ACE G (t) The relationship between the three is: ACE(t)=ACE HE (t)+ACE G (t); The hybrid energy storage frequency modulation power command signal is nonlinearly decomposed, including high-frequency power components and medium- and low-frequency energy components.

5. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 4 is characterized by: The modal component IMF can be obtained by decomposing it using the IHHO-VMD method. i (t) and residual R n (t), using the improved HHO optimized VMD method, the high and low frequency components of the hybrid energy storage response part after decomposition are reconstructed; the reconstructed high and low frequency components are responded through the power type energy storage flywheel energy storage system, and the low frequency components are responded through the energy type energy storage lithium battery system; by setting the reconstruction point m, the frequency modulation requirements of the AGC signal can be responded quickly and accurately, and the decomposed mode is reconstructed. The decomposed high frequency component and low frequency component can be expressed as: ACE HE (t)=ACE h (t)+ACE l (t); Among them: ACE h (t) is the reconstructed high-frequency component; ACE l (t) is the reconstructed low-frequency component; IMF i (t) is the i-th mode obtained by VMD decomposition; ACE HE (t) represents the signal that the hybrid energy storage needs to respond to; m represents the reconstruction point; K represents the modal number; i represents the i-th modal component.

6. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 1 is characterized by: In step 3, the capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation includes: 3.1: Economic compensation model for hybrid energy storage system participating in AGC frequency regulation service: Where: C J represents the economic compensation benefits of the hybrid energy storage system participating in the AGC frequency regulation service; J p It is a comprehensive index of frequency modulation performance; D i is the i-th AGC regulation depth, which is the contribution of the unit to the grid frequency stability in the i-th AGC frequency regulation. When the specified power is reached before receiving the next AGC instruction, D i =|P A -P Si |,P A Indicates the power value required for secondary frequency modulation, P Si Indicates the original power when the frequency modulation command is received. If the specified power is not reached, D i =|P Ei -P Si |,P Ei Indicates the power value when the unit receives the next frequency modulation instruction, Y AGC is the unit price of AGC frequency modulation performance compensation; n is the number of replacements, n = T LCC / T life , T life is the service life of the lithium battery; 3.2: Initial investment cost model of hybrid energy storage system: Where: C1 represents the initial investment cost of the hybrid energy storage system; and are the unit power costs of lithium battery energy storage and flywheel energy storage respectively; and are the rated powers of lithium battery energy storage and flywheel energy storage respectively; and are the unit capacity costs of lithium battery energy storage and flywheel energy storage respectively; and are the rated capacities of lithium battery energy storage and flywheel energy storage respectively; r is the discount rate, n is the number of replacements, n=T LCC / T life , T life is the service life of the lithium battery; T LCC Represents the entire life cycle; 3.3: Hybrid energy storage system operation and maintenance cost model: Where: C2 represents the operation and maintenance cost of the hybrid energy storage system; and They are the unit power operation and maintenance costs of lithium battery energy storage and flywheel energy storage respectively; and are the unit capacity operation and maintenance costs of lithium battery energy storage and flywheel energy storage respectively; W L (t) and W F (t) are the annual charge and discharge capacity of lithium battery energy storage and flywheel energy storage respectively; 3.4: Daily revenue of hybrid energy storage participating in grid frequency regulation C total The objective function is as follows: max(C total )=max(C J -C1-C2); Where: max(C total ) represents the maximum daily profit of hybrid energy storage participating in grid frequency regulation.

7. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 6 is characterized by: The constraints of the capacity optimization configuration model for hybrid energy storage participating in grid frequency regulation specifically include: ①. Energy storage charging and discharging power constraints: ②. Constraints of charge state on energy storage: Among them: SOC L (t) and SOC F (t) are the charge states of the lithium battery and flywheel energy storage at time t respectively; and They are respectively the low warning value and the high warning value of the lithium battery charge state; and are the low warning value and high warning value of the flywheel energy storage charge state respectively; η is the charge and discharge efficiency of energy storage; η L and η F are the charging and discharging efficiencies of lithium batteries and flywheel energy storage respectively; P S (t) is the power at time t; △t is the sampling interval; E is the energy storage capacity; Indicates the minimum charging power of the lithium battery at time t; represents the maximum charging power of the lithium battery at time t, Represents the charging power of the lithium battery at time t; Indicates the minimum discharge power of the lithium battery at time t; represents the maximum discharge power of the lithium battery at time t, Represents the discharge power of the lithium battery at time t; SOC(t) represents the state of charge of the energy storage at time t; SOC(t-1) represents the state of charge of the energy storage at time (t-1); ③.Power balance constraints: The power required for frequency regulation of hybrid energy storage is equal to the output power of lithium battery and flywheel energy storage at time t; P(t)=P L (t)+P F (t); Where: P(t) represents the hybrid energy storage frequency regulation power demand at time t.

8. The method for optimizing the configuration of the capacity of hybrid energy storage participating in secondary frequency regulation based on IHHO-VMD according to claim 1 is characterized by: The step 4 comprises the following steps: (1). Parameter initialization, reconstruction point m is 0; (2) Decompose the hybrid energy storage frequency modulation power command signal into K IMF components through IHHO-VMD; (3) Set the charge and discharge power instructions of the flywheel and lithium battery according to the reconstructed high and low frequency component formulas, and use the maximum charge and discharge power and cumulative capacity of the flywheel and lithium battery as reference values ​​for the rated power and rated capacity respectively; (4). Initialize population parameters; (5) Calculate the fitness value of individuals in the population based on the objective function; (6). Calculate the prey energy O; (7) If |O|<1, adopt four strategies to capture prey and update the population position, otherwise perform the position update in the search phase; (8) Calculate the fitness value of the function in step (7). If a better fitness value is obtained, update the global optimal solution and the global optimal fitness value; (9) Determine whether the number of iterations or iteration accuracy has been reached. If so, output the optimal energy storage benefit and the corresponding hybrid energy storage configuration plan, and the algorithm ends; otherwise, return to step (4); (10) Determine whether the reconstruction point m is equal to K. If not, repeat steps (3) to (9) until m=K. At this time, output the hybrid energy storage configuration and benefit at each reconstruction point m, and the algorithm ends.

Images