Hybrid energy storage optimal configuration method meeting energy storage frequency dynamic response limitation and related device

Abstract

The invention discloses a hybrid energy storage optimal configuration method meeting energy storage frequency dynamic response limitation and a related device. The method comprises the following steps: establishing a full-system frequency response model comprising a synchronous generator, multi-type energy storage and renewable energy sources; establishing frequency indexes including the frequency change rate, the frequency lowest point, the steady-state frequency, the hybrid energy storage power capacity and the hybrid energy storage energy capacity; based on the full-system frequency response model, establishing an energy storage system optimization configuration model taking energy storage controller control parameters as optimization variables under the constraint of frequency indexes, so as to minimize energy storage investment; and evaluating the explicit gradient of the frequency index to the controller parameter, and carrying out nonlinear solution on the energy storage system optimal configuration model based on the display gradient to obtain the hybrid energy storage optimal configuration. According to the invention, the advantages of high-rate energy storage and low-rate energy storage are combined, so that the power and energy requirements during the frequency support period are met, and the energy storage investment is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage configuration, and in particular, to an optimal configuration method and related device for hybrid energy storage that meets the dynamic response limit of energy storage frequency. Background Art

[0002] With a large number of converter-interfaced renewable energies replacing synchronous generators, the frequency stability problem of power systems has attracted increasing attention. Different from synchronous generators, in renewable energies, there is no rotor or governor directly connected to the power grid, resulting in low inertia and low damping characteristics of the system. The fluctuations in the output power of renewable energies lead to more and more frequent frequency events.

[0003] Energy storage is a flexible resource that can be adjusted bidirectionally, and configuring energy storage is an effective method to maintain frequency stability. What power grid operation units are more concerned about is how much energy storage must be configured to meet the dynamic security constraints of system frequency. Rich results have been achieved in the research on the energy storage configuration method for high-proportion new energy power systems, but there is still room for improvement both in physical models and theoretical methods. First, existing energy storage configuration methods usually ignore the spatial distribution characteristics of frequency, assuming that the global frequency is consistent, which results in the energy storage layout plan may not provide effective frequency support for the energy storage access points. Then, when configuring existing energy storage, the response time limit of energy storage is usually ignored, and the response time characteristics of different energy storages such as energy-type and power-type will significantly affect the size of the required configured energy storage capacity. Finally, when establishing an optimal configuration model, the quantitative evaluation of the frequency response characteristics of energy storage is based on given frequency control parameters. When the actual control parameters change, the adjustment ability that the energy storage can provide will change significantly, thereby affecting the frequency support effect and further affecting the energy storage capacity configuration. Summary of the Invention

[0004] Aiming at the above-mentioned defects of the prior art, the present invention provides an optimal configuration method and related device for hybrid energy storage that meets the dynamic response limit of energy storage frequency. A node frequency response model including different types of energy storage is established based on the state space, a numerical calculation method for frequency indicators is provided, the energy storage control parameters are optimized according to the economic principle, and a configuration method for allocating frequency modulation power among different types of energy storage is proposed. A numerical calculation method based on the explicit gradient of frequency and energy storage indicators is proposed, which improves the calculation efficiency.

[0005] An optimal configuration method for hybrid energy storage that meets the dynamic response limit of energy storage frequency includes the following steps:

[0006] Establish a full-system frequency response model including synchronous generators, multiple types of energy storage, and renewable energies;

[0007] Establish frequency indicators including the rate of change of frequency, the lowest point of frequency, the steady-state frequency, the power capacity of hybrid energy storage, and the energy capacity of hybrid energy storage;

[0008] Based on the above-mentioned full-system frequency response model, an energy storage system optimization configuration model is established with the control parameters of the energy storage controller as the optimization variables under the constraint of frequency indicators, minimizing the energy storage investment;

[0009] Evaluate the explicit gradient of the frequency indicators with respect to the controller parameters, and nonlinearly solve the energy storage system optimization configuration model based on the explicit gradient to obtain the optimal configuration of the hybrid energy storage.

[0010] Furthermore, the steps for establishing the full-system frequency response model including synchronous generators, multiple types of energy storage, and renewable energy are as follows:

[0011] First, separately derive the frequency response models of synchronous generators, energy storage, and renewable energy, and then establish the full-system frequency response model. On this basis, further obtain the full-system frequency response model for step disturbances;

[0012] The motion equation of the synchronous generator is:

[0013]

[0014] where m i is the inertia; d i is the damping coefficient; Δu i is the external disturbance; Δω i is the rotor speed deviation; Δp mi is the mechanical power deviation; Δp ei is the electromagnetic power deviation;

[0015] The model of the reheating steam turbine and its governing system is:

[0016]

[0017] where T Gi , T CHi and T RHi are the time constants of the governor, the main inlet volume and chamber, and the reheater respectively; F HPi is the proportion of the power generated by the high-pressure cylinder in the total steam turbine power; R i is the adjustment factor of the controller; Δμ i is the valve position change; Δp HPi is the change in the output power of the high-pressure cylinder; Δp RHi is the change in the output power of the reheater;

[0018] The frequency response transfer function of the energy storage system actively participating in frequency support is

[0019]

[0020] where pesi is the output power of the energy storage system; m esi and d esi are the virtual inertia and droop control parameters respectively; T esi is the response time constant, and the high-speed energy storage and low-rate energy storage are represented by subscripts H and L respectively, Δp esHi and Δp esLi represent the output powers of the high-rate and low-rate energy storage systems respectively;

[0021] The grid-forming renewable energy is modeled in a way similar to a synchronous generator, and the grid-following renewable energy is modeled in a way similar to an energy storage. The transfer functions are

[0022]

[0023] where is the auxiliary state variable;

[0024] Combining the motion equation of the synchronous generator and the transfer function of the renewable energy, the generator frequency considering the support of the hybrid energy storage is obtained:

[0025]

[0026] The electromagnetic power fluctuation Δp ei is derived from the linearized model as:

[0027]

[0028] where L ij is the element of the grid Laplacian matrix L:

[0029]

[0030] where V i is the voltage amplitude; b ij is the line susceptance; θ i is the angle of the generator at the balanced power flow;

[0031] Combining the frequency response transfer functions of the synchronous generator, multi-type energy storage and renewable energy, the full-system frequency response model is obtained as:

[0032]

[0033] B = [1 / m 0 0 0 0 0] T (10)

[0034]

[0035] y = [Δω Δp esH Δp esL T ​(12)

[0036] where Δω, Δμ, Δp HP , Δp RP , Δp e , Δp esH , Δp esL are column vectors of the corresponding variables; m, m esH , m esL are diagonal matrices composed of the corresponding variables. The output vector y is the frequency and output power of each bus hybrid energy storage system, and the input vector u is the perturbation vector obtained by Kron reduction:

[0037] The explicit expression of the system's response to the step perturbation u is:

[0038]

[0039] where λ i is the eigenvalue of A; w i and q i are the normalized left and right eigenvectors corresponding to λ i respectively; the residual r i = Cq i w i T B; the left eigenvector matrix w = [w 1 T w 2 T … w 7n T T ; the right eigenvector matrix q = [q 1 q 2 … q 7n ; the relationship between the two matrices is wAq = Λ = diag{λ i}.

[0040] Furthermore, the establishment of frequency indexes including the rate of change of frequency, the lowest frequency point, the steady-state frequency, the power capacity of the hybrid energy storage, and the energy capacity of the hybrid energy storage specifically includes the following steps:

[0041] Rate of change of frequency RoCoF: Using 500 ms after the disturbance as the time interval for RoCoF measurement, the frequency step response can be derived from the response expression of the system to the step perturbation u as:

[0042] Δω(t) = D ω y(t) (14)

[0043] where D ω = [I n 00];

[0044] ​Substitute \(t = 0.5s\) into (4.14), and the RoCoF of each node is obtained as follows:

[0045]

[0046] where RoCoF = [RoCoF 1 RoCoF 2 …RoCoF n T .RoCoF i is the RoCoF of the \(i\)-th generator;

[0047] The lowest point of frequency \(t\) nadir : The corresponding moment of the lowest point of frequency \(t\) nadir can be calculated by the Newton-Raphson method as follows:

[0048]

[0049] where \(t\) nadir (k)=[t nadir1 (k)t nadir2 (k)…t nadirn (k)] T is the time vector of the lowest point of frequency at the \(k\)-th iteration of all generators, and are the first-order and second-order derivatives of frequency respectively, and are expressed as follows:

[0050]

[0051] where \(\odot\) is the Hadamard product. When the iteration converges, \(t\) nadir (k)→t nadir , and the lowest point of frequency is obtained as:

[0052] Δf nadir =Δω(t nadir ) / 2π (19)

[0053] where Δf nadir =[Δf nadir1 ,Δf nadir1 …Δf nadirn T ;

[0054] Steady-state frequency: Taking the steady-state frequency deviation after primary frequency control as the frequency security index, considering the dynamic process of secondary frequency modulation, the actual frequency deviation will be less than this index and meet the conservation requirements. The steady-state frequency depends on the damping of the rotor, the regulation coefficient of the governor, and the droop parameters of the energy storage. The expression of the steady-state frequency deviation Δf ss is as follows:

[0055] ​​

[0056] Hybrid energy storage power capacity: The output power of the hybrid energy storage is calculated from the response expression of the system to the step disturbance u as follows:

[0057] Δp(t) = D p y(t) (21)

[0058] where D p = [0I n 0; 00I n . The output power is Δp = [Δp esH Δp esL T .

[0059] The power capacity of the hybrid energy storage must be greater than the limit power response. The extreme value time t of Δp esH and Δp esL is calculated by the Newton iteration method: p :

[0060]

[0061] where t p = [t p1 t p2 …t p2n T , when the iteration converges to t p (k) → t p , the extreme value of the output power of the hybrid energy storage is obtained. The power capacity of the energy storage system should satisfy the following constraint conditions:

[0062] -P ES ≤ Δp(t p ) ≤ P ES (23)

[0063] where P ES = [P ESH P ESL T is the power capacity of the hybrid energy storage system;

[0064] Hybrid energy storage energy capacity: The energy change of the hybrid energy storage during frequency support can be calculated by integrating the output power of the hybrid energy storage:

[0065]

[0066] The extreme value time t of the energy change e can be given by the Newton iteration method:

[0067]

[0068] where t​​​e = [t e1 t e2 …t e2n T . The energy capacity of the energy storage system must satisfy the following constraints:

[0069] -E ES ≤ ΔE(t e ) ≤ E ES (26)

[0070] where E ES = [E ESH E ESL T is the energy capacity of the hybrid energy storage system.

[0071] Furthermore, based on the full-system frequency response model, an optimal configuration model of the energy storage system is established with the control parameters of the energy storage controller as the optimization variables under the constraint of frequency indexes, so as to minimize the energy storage investment. The specific steps include:

[0072] Considering the constraint of frequency indexes, optimize the configuration of the energy storage system to minimize the energy storage investment. Taking the controller parameters as the optimization variable κ, the optimization problem is summarized as:

[0073]

[0074] s.t. -1 n RoCoF max ≤ RoCoF ≤ 1 n RoCoF max (28)

[0075] -1 n Δf nadir,max ≤ Δf nadir ≤ 1 n Δf nadir,max (29)

[0076] -Δf ss,max ≤ Δf ss ≤ Δf ss,max , (4.23), (4.26), and (30)

[0077] where κ = (m esH , d esH , m esL , d esL ) is the optimization variable, 1 n is an n-dimensional column vector with all elements being 1, C ph and C pl are the unit power costs of the high-rate energy storage and the low-rate energy storage respectively; C eh and C​​el are the unit energy costs of high-rate energy storage and low-rate energy storage respectively; where RoCoF max , Δf nadir,max , and Δf ss,max are the maximum allowable RoCoF, the maximum allowable frequency deviation, and the maximum allowable steady-state frequency deviation respectively.

[0078] Furthermore, the explicit gradient of the evaluation frequency index with respect to the controller parameters is used to nonlinearly solve the optimal configuration model of the energy storage system based on the explicit gradient, and the optimal configuration of the hybrid energy storage is obtained. The specific steps include:

[0079] Evaluate the explicit gradient of the evaluation frequency index with respect to the control parameters m esi and d esi ;

[0080] The gradient of the eigenvalue with respect to the control parameter is:

[0081]

[0082] where

[0083]

[0084] where

[0085] Calculate the gradient of the residual r i as:

[0086]

[0087] Determine the RoCoF gradient as:

[0088]

[0089] Δf nadir , Δp(t p ), ΔE(t e ) are expressed in relation to t nadir , t p , t e respectively. The gradient of the peak time is approximated as:

[0090]

[0091] where the gradient of the frequency derivative can be expressed as:

[0092]

[0093] For the gradients of other peak times, t p and t e , calculate them in the same way. Δf nadir, Δp(t p ), and the gradients of ΔE(t e ) can be expressed as:

[0094]

[0095] Based on RoCoF, for Δf nadir , Δp(t p ), and the gradients of ΔE(t e ), a non - linear solver is used to solve the optimal configuration model of the energy storage system, and the optimal configuration of the hybrid energy storage that meets the energy storage frequency dynamic response limit is obtained.

[0096] An optimal configuration device of a hybrid energy storage that meets the energy storage frequency dynamic response limit includes:

[0097] A full - system frequency response model establishment module for establishing a full - system frequency response model including synchronous generators, multiple types of energy storage, and renewable energy;

[0098] A frequency index establishment module for establishing frequency indexes including the rate of change of frequency, the lowest frequency point, the steady - state frequency, the power capacity of the hybrid energy storage, and the energy capacity of the hybrid energy storage;

[0099] An energy storage system optimal configuration model establishment module for establishing an energy storage system optimal configuration model with the control parameters of the energy storage controller as optimization variables under the constraint of frequency indexes based on the full - system frequency response model to minimize energy storage investment;

[0100] A hybrid energy storage optimal configuration solving module for evaluating the explicit gradients of frequency indexes with respect to controller parameters and non - linearly solving the energy storage system optimal configuration model based on the explicit gradients to obtain the optimal configuration of the hybrid energy storage.

[0101] An optimal configuration system of a hybrid energy storage that meets the energy storage frequency dynamic response limit includes: a computer - readable storage medium and a processor;

[0102] The computer - readable storage medium is used to store executable instructions;

[0103] The processor is used to read the executable instructions stored in the computer - readable storage medium and execute the optimal configuration method of the hybrid energy storage that meets the energy storage frequency dynamic response limit.

[0104] A non - transitory computer - readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the optimal configuration method of the hybrid energy storage that meets the energy storage frequency dynamic response limit.

[0105] The present invention has the following beneficial effects compared with the prior art:

[0106] The present invention transforms the original optimal layout and capacity problem of energy storage into an optimal control parameter problem of energy storage to ensure sufficient frequency support. Aiming at the limitations of existing methods that do not consider the spatial characteristics of frequency response and the influence of the response speed of different types of energy storage on energy storage configuration, the present invention establishes a node frequency response model including different types of energy storage based on state space, and proposes a numerical calculation method for frequency indicators, which can calculate frequency dynamics more accurately. A configuration method is provided to optimize the control parameters of energy storage according to economic principles and allocate frequency modulation power among different types of energy storage. The calculation efficiency is improved by a numerical calculation method based on the explicit gradient of frequency and energy storage indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] Figure 1 FIG. is a schematic flow chart of an optimal configuration method for a hybrid energy storage that meets the limitations of energy storage frequency dynamic response provided by an embodiment of the present invention;

[0108] Figure 2 FIG. is a single-line schematic diagram of a power system in southeastern Australia provided by an embodiment of the present invention;

[0109] Figure 3 FIG. is the distribution result of the optimal inertia and droop coefficients in the optimal configuration of a hybrid energy storage provided by an embodiment of the present invention;

[0110] Figure 4 FIG. is the power result in the optimal configuration of a hybrid energy storage provided by an embodiment of the present invention;

[0111] Figure 5 FIG. is the energy result in the optimal configuration of a hybrid energy storage provided by an embodiment of the present invention;

[0112] Figure 6 FIG. is the maximum frequency deviation under 9 disturbance conditions after configuring a hybrid energy storage provided by an embodiment of the present invention;

[0113] Figure 7 FIG. is the RoCoF under 9 disturbance conditions after configuring a hybrid energy storage provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0114] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0115] Please refer to Figure 1, the first aspect of the present invention provides an optimal configuration method for hybrid energy storage to meet the dynamic response limit of energy storage frequency, including the following steps:

[0116] Step 1: Establish a full-system frequency response model including synchronous generators, multiple types of energy storage, and renewable energy;

[0117] Step 2: Establish frequency indicators including the rate of change of frequency, the lowest point of frequency, the steady-state frequency, the power capacity of the hybrid energy storage, and the energy capacity of the hybrid energy storage;

[0118] Step 3: Based on the full-system frequency response model, establish an optimal configuration model of the energy storage system with the control parameters of the energy storage controller as the optimization variables under the constraint of frequency indicators to minimize the energy storage investment;

[0119] Step 4: Evaluate the explicit gradient of the frequency indicators with respect to the controller parameters, and nonlinearly solve the optimal configuration model of the energy storage system based on the explicit gradient to obtain the optimal configuration of the hybrid energy storage.

[0120] The specific steps for establishing the full-system frequency response model including synchronous generators, multiple types of energy storage, and renewable energy in Step 1 are as follows:

[0121] Step 1.1: The motion equation of the synchronous generator is:

[0122]

[0123] where m i is the inertia; d i is the damping coefficient; Δu i is the external disturbance; Δω i is the rotor speed deviation; Δp mi is the mechanical power deviation; Δp ei is the electromagnetic power deviation.

[0124] The model of the reheating steam turbine and its governing system is:

[0125]

[0126] where T Gi , T CHi and T RHi are the time constants of the governor, the main inlet volume and the air chamber, and the reheater respectively; F HPi is the proportion of the power generated by the high-pressure cylinder in the total steam turbine power; R i is the adjustment factor of the controller; Δμ i is the valve position change; Δp HPi is the change in the output power of the high-pressure cylinder; Δp RHi is the change in the output power of the reheater.

[0127] Step 1.2: The frequency response transfer function for the energy storage system to actively participate in frequency support is

[0128]

[0129] where p esi is the output power of the energy storage system; m esi and d esi are the virtual inertia and droop control parameters respectively; T esi is the response time constant, and the response time varies with different energy storage types. High-speed energy storage (such as supercapacitors) has a shorter response time, which is in the order of 1 - 20 ms, and each single cell has a large power capacity and a small energy capacity; low-rate energy storage (such as lithium batteries), this type of energy storage has a longer response time, and the response time constant is in the order of 20 ms - s. These two types of energy storage are represented by subscripts H and L respectively. Δp esHi and Δp esLi represent the output powers of the high-rate and low-rate energy storage systems respectively.

[0130] Step 1.3: The renewable energy with grid-forming control is modeled in a way similar to a synchronous generator, and the renewable energy with grid-following control is modeled in a way similar to energy storage. The transfer functions are

[0131]

[0132] where is the auxiliary state variable.

[0133] Step 1.4: Combining the motion equation of the synchronous generator and the transfer function of the renewable energy, the generator frequency considering the support of hybrid energy storage is obtained as:

[0134]

[0135] The electromagnetic power fluctuation Δp ei is derived from the linearized model as:

[0136]

[0137] where L ij is the element of the grid Laplacian matrix L.

[0138]

[0139] where V i is the voltage amplitude; b ij is the line susceptance; θ i is the angle of the generator at the balanced power flow.

[0140] Combining the frequency response transfer functions of the synchronous generator, multiple types of energy storage, and renewable energy, the simplified frequency response model of the entire system is

[0141]

[0142] B = [1 / m 000000] T (10)

[0143]

[0144] y = [Δω Δp esH Δp esL T (12)

[0145] where Δω, Δμ, Δp HP , Δp RP , Δp e , Δp esH , Δp esL are column vectors of the corresponding variables; m, m esH , m esL are diagonal matrices composed of the corresponding variables. The output vector y is the frequency and output power of the hybrid energy storage system at each bus. The input vector u is the perturbation vector obtained by Kron reduction.

[0146] The explicit expression for the system's response to a step perturbation u is:[[]]

[0147]

[0148] where λ i is the eigenvalue of A; w i and q i are the normalized left and right eigenvectors corresponding to λ i respectively; the residual r i = Cq i w i T B; the left eigenvector matrix w = [w 1 T w 2 T … w 7n T T ; the right eigenvector matrix q = [q 1 q 2 … q 7n ; the relationship between the two matrices is wAq = Λ = diag{λ i}.

[0149] ​​Step 2 establishes frequency indexes including rate of change of frequency, lowest frequency point, steady-state frequency, hybrid energy storage power capacity, and hybrid energy storage energy capacity. The specific steps are as follows:

[0150] Step 2.1: Rate of change of frequency RoCoF: The 500 ms after the disturbance is used as the time interval for RoCoF measurement. The frequency step response can be derived from the response expression of the system to the step disturbance u as follows:

[0151] Δω(t) = D ω y(t) (14)

[0152] where D ω = [I n 00].

[0153] Substituting t = 0.5 s into (4.14), the RoCoF of each node is obtained as follows:

[0154]

[0155] where RoCoF = [RoCoF 1 RoCoF 2 …RoCoF n T .RoCoF i is the RoCoF of the i-th generator.

[0156] Step 2.2: Lowest frequency point t nadir : The corresponding moment of the lowest frequency point t nadir can be calculated by the Newton iteration method as follows:

[0157]

[0158] where t nadir (k) = [t nadir1 (k)t nadir2 (k)…t nadirn (k)] T is the time vector of the lowest frequency point of all generators at the k-th iteration, and are the first-order and second-order derivatives of the frequency, respectively, which are expressed as follows.

[0159]

[0160] where ⊙ is the Hadamard product. When the iteration converges, t nadir (k) → t nadir , and the lowest frequency point is obtained as follows:

[0161] Δf nadir = Δω(t​nadir ) / 2π (19)

[0162] where Δf nadir =[Δf nadir1 , Δf nadir1 … Δf nadirn T .

[0163] Step 2.3: Steady-state frequency: The steady-state frequency deviation after primary frequency control is used as the frequency safety index. Considering the secondary frequency modulation dynamic process, the actual frequency deviation will be less than this index, meeting the conservation requirements. The frequency at steady state depends on the damping of the rotor, the regulation coefficient of the governor, and the droop parameter of the energy storage. The expression for the steady-state frequency deviation Δf ss is as follows:

[0164]

[0165] Hybrid energy storage power capacity: The output power of the hybrid energy storage is calculated from the response expression of the system to the step disturbance u as:

[0166] Δp(t) = D p y(t) (21)

[0167] where D p =[0 I n 0; 0 0 I n . The output power is Δp = [Δp esH Δp esL T .

[0168] Step 2.4: The power capacity of the hybrid energy storage must be greater than the limit power response. Use the Newton iteration method to calculate the extreme value time t esH of Δp esL and Δp p .

[0169]

[0170] where t p =[t p1 t p2 … t p2n T . When the iteration converges to t p (k) → t p , the extreme value of the hybrid energy storage output power can be obtained. The power capacity of the energy storage system should satisfy the following constraint conditions:

[0171] -P ES ≤ Δp(t p ) ≤ P ES (23) ​​​

[0172] where P ES =[P ESH P ESL T is the power capacity of the hybrid energy storage system.

[0173] Step 2.5: Hybrid energy storage energy capacity: The energy change of the hybrid energy storage during frequency support can be calculated by integrating the output power of the hybrid energy storage.

[0174]

[0175] The extreme value time t of the energy change e can be given by the Newton iteration method:

[0176]

[0177] where t e =[t e1 t e2 …t e2n T . The energy capacity of the energy storage system must satisfy the following constraint conditions:

[0178] -E ES ≤ΔE(t e )≤E ES (26)

[0179] In the formula, E ES =[E ESH E ESL T is the energy capacity of the hybrid energy storage system.

[0180] The third step is to establish an optimized configuration model of the energy storage system with the control parameters of the energy storage controller as the optimization variables under the frequency index constraints based on the full-system frequency response model, so as to minimize the energy storage investment. The specific steps include:

[0181] Considering the frequency index constraints, optimize the configuration of the energy storage system to minimize the energy storage investment. Taking the controller parameters as the optimization variable κ, the optimization problem can be summarized as:

[0182]

[0183] s.t. -1 n RoCoF max ≤RoCoF≤1 n RoCoF max (28)

[0184] -1 n Δf nadir,max ≤Δf nadir ​​​≤ 1 n Δf nadir,max (29)

[0185] -Δf ss,max ≤ Δf ss ≤ Δf ss,max , (4.23), (4.26), and (30)

[0186] where κ = (m esH , d esH , m esL , d esL ) are optimization variables. 1 n is an n-dimensional column vector with all elements equal to 1. C ph and C pl are the unit power costs of high-rate energy storage and low-rate energy storage, respectively; C eh and C el are the unit energy costs of high-rate energy storage and low-rate energy storage, respectively; where RoCoF max , Δf nadir,max , and Δf ss,max are the maximum allowable RoCoF, the maximum allowable frequency deviation, and the maximum allowable steady-state frequency deviation, respectively.

[0187] Step 4 evaluates the explicit gradient of the frequency index with respect to the controller parameters, and nonlinearly solves the optimal configuration model of the energy storage system based on the display gradient to obtain the optimal configuration of the hybrid energy storage. The specific steps include:

[0188] Evaluate the explicit gradient of the frequency index with respect to the control parameters m esi and d esi .

[0189] The gradient of the eigenvalue with respect to the control parameter is:

[0190]

[0191] where

[0192]

[0193] where

[0194] Calculate the gradient of the residual r i as

[0195]

[0196] Determine the RoCoF gradient as

[0197]

[0198] Δfnadir , Δp(t p ), ΔE(t e ) are respectively related to t nadir , t p , t e . Since the explicit expression of the peak time cannot be obtained, the gradient of the peak time is approximated as

[0199]

[0200] where the gradient of the frequency derivative can be expressed as:

[0201]

[0202] The gradients of other peak times, t p and t e , are calculated in the same way. The gradients of Δf nadir , Δp(t p ), and ΔE(t e ) can be expressed as:

[0203]

[0204] Based on RoCoF, for the gradients of Δf nadir , Δp(t p ), and ΔE(t e ), a commercially available nonlinear solver (such as Interior Point Optimizer (IPOPT)) is used to solve the optimal configuration model of the energy storage system, and the optimal configuration of the hybrid energy storage that meets the dynamic response limit of the energy storage frequency is obtained.

[0205] The embodiment of the present invention adopts a modified case of the southeastern power system in Australia. The modified line graph is as shown in Figure 2 . This system consists of five parts. The coupling between regions under this topological structure is very weak. The original 9 generators are replaced by wind power plants, and the wind power plants provide the same power to simulate a power system with a high new energy penetration rate. The energy storage configuration goal of this case is to maintain the system frequency within the constraint range when one of the wind farms is disconnected from the grid. Therefore, there will be a disturbance scenario where 9 wind farms are disconnected from the grid. The hybrid energy storage configuration results obtained by the present invention are as shown in Figure 3 , Figure 4 , Figure 5 . The power demand during the inertial support process is mainly met by the high-rate energy storage system. The energy demand during the droop support process is mainly met by the low-rate energy storage. Due to the limitation of the energy storage inertia coefficient in each region, the low-rate energy storage system in Region 2 can still provide a large inertial support. After configuring the hybrid energy storage by the method in this article, Δf under 9 disturbance scenarios nadirand RoCoF are respectively as Figure 6 and Figure 7 shown. It can be seen from the histogram that under different interferences, the frequency of each area is within the constraint range. The simulation results show that this method can adapt to the hybrid energy storage configuration of large power systems under various disturbances.

[0206] On the other hand, the present invention provides a hybrid energy storage optimal configuration device that meets the dynamic response limit of energy storage frequency, including:

[0207] A full-system frequency response model establishment module, which is used to establish a full-system frequency response model including synchronous generators, multiple types of energy storage, and renewable energy;

[0208] A frequency index establishment module, which is used to establish frequency indexes including rate of change of frequency, lowest frequency point, steady-state frequency, hybrid energy storage power capacity, and hybrid energy storage energy capacity;

[0209] An energy storage system optimal configuration model establishment module, which is used to establish an energy storage system optimal configuration model with the control parameters of the energy storage controller as optimization variables under the constraint of frequency indexes based on the full-system frequency response model, so as to minimize the energy storage investment;

[0210] A hybrid energy storage optimal configuration solving module, which is used to evaluate the explicit gradient of the frequency indexes with respect to the controller parameters, and non-linearly solve the energy storage system optimal configuration model based on the explicit gradient to obtain the optimal configuration of the hybrid energy storage.

[0211] On the other hand, the present invention provides a hybrid energy storage optimal configuration system that meets the dynamic response limit of energy storage frequency, including: a computer-readable storage medium and a processor;

[0212] The computer-readable storage medium is used to store executable instructions;

[0213] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the hybrid energy storage optimal configuration method described in the first aspect that meets the dynamic response limit of energy storage frequency.

[0214] On the other hand, the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the hybrid energy storage optimal configuration method described in the first aspect that meets the dynamic response limit of energy storage frequency.

[0215] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0216] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or a combination of multiple flows and / or blocks

[0217] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one or more of the flows Figure 1 or a combination of multiple flows and / or blocks

[0218] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or a combination of multiple flows and / or blocks

[0219] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A hybrid energy storage optimal configuration method that meets the dynamic response constraints of energy storage frequency, characterized in that: The steps include: Establish a full system frequency response model including synchronous generators, multiple types of energy storage and renewable energy; Establish frequency indicators including frequency change rate, frequency minimum point, steady-state frequency, hybrid energy storage power capacity, and hybrid energy storage energy capacity; Based on the full system frequency response model, an energy storage system optimization configuration model is established under frequency index constraints with energy storage controller control parameters as optimization variables to minimize energy storage investment; The explicit gradient of the frequency index to the controller parameters is evaluated, and the nonlinear solution of the energy storage system optimization configuration model is obtained based on the explicit gradient to obtain the optimal configuration of the hybrid energy storage.

2. The hybrid energy storage optimal configuration method that meets the energy storage frequency dynamic response constraints as claimed in claim 1 is characterized in that: The specific steps of establishing a full system frequency response model including synchronous generators, multi-type energy storage and renewable energy include: First, the frequency response models of synchronous generators, energy storage and renewable energy are derived respectively, and then the frequency response model of the whole system is established. On this basis, the frequency response model of the whole system for step disturbance is further obtained. The equation of motion for a synchronous generator is: Where m i is the inertia; d i is the damping coefficient; Δu i is the external disturbance; Δω i is the rotor speed deviation; Δp mi is the mechanical power deviation; Δp ei is the electromagnetic power deviation; The model of the reheat steam turbine and its speed control system is: Where T Gi , T CHi and T RHi are the time constants of the governor, main intake volume and air chamber, and reheater respectively; F HPi It is the proportion of the power generated by the high-pressure cylinder in the total turbine power; R i is the adjustment factor of the controller; Δμ i is the valve position change; Δp HPi is the change in output power of the high-pressure cylinder; Δp RHi is the change in reheater output power; The frequency response transfer function of the energy storage system actively participating in frequency support is: where p esi is the output power of the energy storage system; m esi and d esi are virtual inertia and droop control parameters respectively; T esi is the response time constant. High-speed energy storage and low-speed energy storage are represented by subscripts H and L, respectively. esHi and Δp esLi Represent the output power of high-rate and low-rate energy storage systems respectively; The renewable energy controlled by the grid is modeled in a similar way to the synchronous generator, and the renewable energy controlled by the grid is modeled in a similar way to the energy storage. The transfer functions are in is the auxiliary state variable; Combining the equation of motion of the synchronous generator and the transfer function of renewable energy, the generator frequency supported by the hybrid energy storage must be considered: Electromagnetic power fluctuation Δp ei It can be derived from the linearized model as follows: Where L ij is the elements of the grid Laplacian matrix L: Where V i is the voltage amplitude; b ij is the line susceptance; θ i is the angle of the generator when balancing the power flow; Combining the frequency response transfer functions of synchronous generators, multi-type energy storage and renewable energy, the frequency response model of the whole system is obtained as follows: B=[1 / m 000000] T (10) y=[ΔωΔp esH Δp esL ] T (12) where Δω, Δμ, Δp HP ,Δp RP ,Δp e , Δp esH ,Δp esL is the column vector of the corresponding variable; m,m esH ,m esL is a diagonal matrix composed of corresponding variables, the output vector y is the frequency and output power of each bus hybrid energy storage system, and the input vector u is the disturbance vector obtained by Kron simplification: The system response to the step disturbance u is explicitly expressed as: where λ i is the eigenvalue of A; w i and q i They are λ i The corresponding normalized left and right eigenvectors; the residual r i =Cq i w i T B; left eigenvector matrix w = [w1 T w2 T …w 7n T ] T ; Right eigenvector matrix q = [q1 q2…q 7n ]; the relationship between the two matrices is wAq = Λ = diag{λ i }.

3. The hybrid energy storage optimal configuration method that meets the energy storage frequency dynamic response constraints as claimed in claim 2 is characterized in that: The specific steps of establishing frequency indicators including frequency change rate, frequency minimum point, steady-state frequency, hybrid energy storage power capacity, and hybrid energy storage energy capacity include: Frequency change rate RoCoF: 500ms after the disturbance is used as the time interval for RoCoF measurement. The frequency step response can be derived from the system response expression to the step disturbance u: Δω(t)=D ω y(t) (14) Where D ω =[I n 00]; Substituting t=0.5s into (4.14), we get the RoCoF of each node: Where RoCoF = [RoCoF1 RoCoF2…RoCoF n ] T .RoCoF i is the RoCoF of the i-th generator; Frequency minimum point t nadir : Frequency minimum point t nadir The corresponding time can be calculated using Newton's iteration method as: Where t nadir (k) = [t nadir1 (k)t nadir2 (k)…t nadirn (k)] T is the time vector of the lowest frequency point of all generators at the kth iteration, and are the first and second order derivatives of the frequency, respectively, expressed as follows: Where ⊙ is the Hadamard product. When the iteration converges, t nadir (k)→t nadir , the lowest frequency point is: Δf nadir =Dω(t nadir ) / 2π (19) where Δf nadir =[Δf nadir1 , Δf nadir1 … Δf nadirn T ;​ Steady-state frequency: The steady-state frequency deviation after primary frequency control is used as the frequency safety index. Considering the dynamic process of secondary frequency modulation, the actual frequency deviation will be smaller than this index, meeting the conservation requirement. The frequency at steady state depends on the rotor damping, the speed regulator adjustment coefficient and the energy storage droop parameter. The steady-state frequency deviation Δf ss The expression is: Hybrid energy storage power capacity: The output power of the hybrid energy storage is calculated from the system's response expression to the step disturbance u: Δp(t)=D p y(t) (21) Where D p =[0I n 0;00I n ]. The output power is Δp = [Δp esH Δp esL ] T . The power capacity of the hybrid energy storage must be greater than the limit power response, and the Newton iteration method is used to calculate Δp esH and Δp esL The extreme value time t p : where t p =[t p1 t p2 …t p2n ] T , when the iteration converges to t p (k)→t p , the output power of the hybrid energy storage is obtained, and the power capacity of the energy storage system should meet the following constraints: -P ES ≤Δp(t p )≤P ES (23) where P ES =[P ESH P ESL ] T is the power capacity of the hybrid energy storage system; Hybrid energy storage capacity: The energy change of the hybrid energy storage during the frequency support period can be calculated by integrating the output power of the hybrid energy storage: The extreme time of energy change t e It can be given by Newton's iteration method: where t e =[t e1 t e2 …t e2n ] T The energy capacity of the energy storage system must meet the following constraints: -E ES ≤ΔE(t e )≤E ES (26) where E ES =[E ESH E ESL ] T is the energy capacity of the hybrid energy storage system.

4. The hybrid energy storage optimal configuration method that meets the energy storage frequency dynamic response constraints as claimed in claim 3 is characterized by: The step of establishing an energy storage system optimization configuration model based on the full system frequency response model under frequency index constraints and taking energy storage controller control parameters as optimization variables to minimize energy storage investment includes: Considering the frequency index constraint, the energy storage system configuration is optimized to minimize the energy storage investment. The controller parameters are taken as the optimization variable κ. The optimization problem is summarized as: s.t.-1 n RoCoF max ≤RoCoF≤1 n RoCoF max (28) -1 n Δf nadir,max ≤Δf nadir ≤1 n Δf nadir,max (29) -Δf ss,max ≤Δf ss ≤Δf ss,max , (4.23), (4.26), and (30) In the formula, κ=(m esH ,d esH ,m esL ,d esL ) is the optimization variable, 1 n is an n-dimensional column vector with all elements equal to 1, C ph and C pl are the unit power costs of high-rate energy storage and low-rate energy storage respectively; C eh and C el are the unit energy costs of high-rate energy storage and low-rate energy storage, respectively; among them, RoCoF max ,Δf nadir,max , and Δf ss,max They are the maximum allowable RoCoF, the maximum allowable frequency deviation and the maximum allowable steady-state frequency deviation respectively.

5. The hybrid energy storage optimal configuration method that meets the energy storage frequency dynamic response constraints as claimed in claim 4 is characterized in that: The explicit gradient of the evaluation frequency index to the controller parameter is used to nonlinearly solve the energy storage system optimization configuration model based on the explicit gradient to obtain the optimal configuration of the hybrid energy storage. The specific steps include: Evaluation frequency index on control parameter m esi and d esi The explicit gradient of The gradient of the eigenvalue with respect to the control parameter is: in in Calculate the residual r i The gradient of is: The RoCoF gradient is determined as: Δf nadir ,Δp(t p )、ΔE(t e ) are expressed as t nadir ,t p ,t e Related, the gradient of the peak time is approximated as: The gradient of the frequency derivative can be expressed as: Gradients at other peak times, t p and t e , calculate in the same way. nadir ,Δp(t p ), and ΔE(t e ) can be expressed as: Based on RoCoF, Δf nadir ,Δp(t p ), and ΔE(t e ) is calculated and the nonlinear solver is used to solve the optimal configuration model of the energy storage system to obtain the optimal configuration of the hybrid energy storage that meets the dynamic response constraints of the energy storage frequency.

6. A hybrid energy storage optimal configuration device that meets the dynamic response constraints of energy storage frequency, characterized in that: include: The whole system frequency response model establishment module is used to establish the whole system frequency response model including synchronous generators, multi-type energy storage and renewable energy; A frequency index establishment module is used to establish frequency indicators including frequency change rate, frequency minimum point, steady-state frequency, hybrid energy storage power capacity, and hybrid energy storage energy capacity; An energy storage system optimization configuration model establishment module is used to establish an energy storage system optimization configuration model based on the full system frequency response model under frequency index constraints, with energy storage controller control parameters as optimization variables, so as to minimize energy storage investment; The hybrid energy storage optimal configuration solving module is used to evaluate the explicit gradient of the frequency index to the controller parameters, and to obtain the optimal configuration of the hybrid energy storage by nonlinearly solving the energy storage system optimization configuration model based on the explicit gradient.

7. A hybrid energy storage optimal configuration system that meets the dynamic response constraints of energy storage frequency, comprising: A computer readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the hybrid energy storage optimal configuration method that meets the dynamic response constraints of energy storage frequency as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the hybrid energy storage optimal configuration method that meets the energy storage frequency dynamic response constraints as described in any one of claims 1 to 5.

Images