A method for frequency modulation power double-layer optimization of fire-multiple storage system

Abstract

A fire-multiple storage system frequency modulation power double-layer optimization method belongs to the technical field of energy storage frequency modulation. The present application is a fire-multiple storage system frequency modulation power double-layer optimization method based on ensemble empirical mode decomposition and multi-objective genetic algorithm. In view of the problems of high overall operation cost and poor frequency modulation performance caused by the participation of multiple energy storage systems in frequency modulation, the fire-multiple storage system frequency modulation power double-layer optimization method optimizes the fire-multiple storage system frequency modulation power layer based on the double-layer strategy framework of regional power grid, constructs a time-space filter based on ensemble empirical mode decomposition to determine the frequency modulation power instructions of the fire-multiple storage system, and optimizes the multiple energy storage system frequency modulation power layer to determine the frequency modulation cost and SOC two optimization objectives to construct a model, and the model is solved based on the multi-objective genetic algorithm to obtain the output instructions of each energy storage system. The simulation analysis is carried out based on the above method, and the traditional strategy scene is compared and analyzed, and the orderliness and scientificity of the present application are shown.

Description

Technical Field

[0001] This invention belongs to the field of energy storage frequency regulation technology, and in particular relates to a two-layer optimization method for frequency regulation power of fire-multiple storage systems based on ensemble empirical mode decomposition and multi-objective genetic algorithm. Background Technology

[0002] Amidst nationwide calls for green and low-carbon development, enterprises are vigorously promoting the transformation and upgrading of the energy industry, leading to a rapid increase in the installed capacity of new energy sources such as wind and solar power. However, the intermittent and random nature of new energy power generation devices weakens the stability of power system frequency operation and increases the difficulty of frequency regulation due to their large-scale grid connection. The development of energy storage technology and national support have made it an effective means of auxiliary frequency regulation. However, the high cost and requirements for energy storage capacity and power make coordinating the power allocation among all energy storage systems within a region crucial for achieving optimal regional economic efficiency and frequency regulation performance. Currently, in regional power grids containing multiple energy storage systems, there are performance and cost differences between thermal power and energy storage frequency regulation resources. Furthermore, multiple energy storage systems may prematurely lose their frequency regulation capability due to excessively low or high SOC (State of Charge) for a particular energy storage system, resulting in higher overall operating costs and poorer frequency regulation performance when multiple energy storage systems participate in frequency regulation.

[0003] Therefore, there is an urgent need for a new technical solution to address this problem. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a two-layer optimization method for frequency regulation power of a thermal-multi-energy storage system. This method can improve the frequency regulation effect of the regional power grid and reduce the frequency regulation cost, and balance the SOC of multiple energy storage systems. It is used to solve the technical problems of high overall operating cost and poor frequency regulation performance caused by the participation of multiple energy storage systems in frequency regulation in the current regional power grid containing multiple energy storage systems.

[0005] A two-layer optimization method for frequency regulation power in a fire-multiple-storage system includes the following steps, which are performed sequentially:

[0006] Step 1: Establish a two-tiered control center in the fire-multiple storage system and determine the relationship between the two control centers.

[0007] The dual-layer control center includes an upper-layer thermal-storage frequency regulation system control center and a lower-layer multi-energy storage frequency regulation system control center. The thermal-storage frequency regulation system control center receives the overall automatic generation control (AGC) command, generates thermal-storage frequency regulation power commands through Empirical Mode Decomposition (EEMD), and distributes them between the thermal power units and the energy storage systems. The thermal-storage frequency regulation power commands include those for the thermal power units and those for the energy storage systems. The multi-energy storage frequency regulation system control center receives the frequency regulation power commands for the energy storage systems from the upper-layer thermal-storage frequency regulation system control center and distributes them among the various energy storage systems.

[0008] Step 2: Construct a two-level optimization model for the frequency regulation power of the thermal power-multiple energy storage system.

[0009] The construction of the dual-layer optimization model for the frequency regulation power of the thermal-multi-storage system includes the construction of a thermal-storage frequency regulation power optimization layer and a multi-storage system frequency regulation power optimization layer.

[0010] 1) Construct a thermal-storage frequency modulation power optimization layer;

[0011] ① Construct a spatiotemporal filter based on ensemble empirical mode decomposition (EEMD);

[0012] ② Determine the optimal order k of the spatiotemporal filter;

[0013] ③ Based on the determined spatiotemporal filter of optimal order k, the frequency regulation power command allocation between thermal power unit and energy storage system is carried out to obtain the frequency regulation power command undertaken by the thermal power unit as a whole and the frequency regulation power command undertaken by the energy storage system as a whole, and the frequency regulation power command is adjusted according to different situations;

[0014] 2) Construct a frequency regulation power optimization layer for multi-energy storage systems;

[0015] ①To address the two optimization objectives of the total frequency regulation cost of the multi-energy storage system and the overall SOC state of the multi-energy storage system, a mathematical formula for the multi-objective optimization problem is established;

[0016] ② Construct a frequency regulation power optimization model for multiple energy storage systems and obtain the objective function for the frequency regulation cost of each energy storage system;

[0017] ③ Construct a comprehensive control model of the state of charge (SOC) of the energy storage system and obtain the objective function of the comprehensive SOC deviation;

[0018] ④ Apply conditional constraints to each objective function;

[0019] 3) Use genetic algorithms to solve the mathematical formulas of multi-objective optimization problems;

[0020] ① Use fuzzy set theory to find the optimal solution set of the mathematical formula for multi-objective optimization problems, express the satisfaction corresponding to each objective function in each solution in the optimal solution set as a function, and obtain the objective function satisfaction matrix of each set of solutions in the optimal solution set;

[0021] ② Construct an adaptive weighting coefficient matrix for the system frequency regulation demand and the overall deviation of SOC, and then obtain the overall satisfaction matrix;

[0022] ③The solution corresponding to the maximum value of the comprehensive satisfaction matrix is ​​the optimal output of the fire-multiple storage system;

[0023] Step 3: Determine whether the remaining frequency regulation capacity of the energy storage is sufficient;

[0024] If the remaining operable frequency regulation capacity of all energy storage systems is greater than or equal to the frequency regulation power command value allocated to the upper-level multi-energy storage system as a whole, it is determined that the remaining frequency regulation capacity of energy storage is sufficient. Then, the dual-layer optimization model of the fire-multi-energy storage system frequency regulation power outputs the optimal output as the frequency regulation power command of each energy storage system.

[0025] If the remaining operable frequency regulation capacity of all energy storage systems is less than the frequency regulation power command value allocated to the upper-level multi-energy storage system as a whole, it is determined that the remaining frequency regulation capacity of energy storage is insufficient. In this case, the dual-layer optimization model of the fire-multi-energy storage system frequency regulation power outputs the maximum operable frequency regulation power output as the frequency regulation power command of each energy storage system.

[0026] The relationship between the control centers of each frequency modulation system in step 1 is as follows:

[0027] P AGC (t)=P G (t)+P B,total (t) (1)

[0028]

[0029] In the formula, P AGC (t) represents the total frequency modulation power demand at time t, P G (t) represents the frequency regulation power allocated to the entire thermal power unit at time t, P B,total (t) represents the frequency regulation power allocated to the entire energy storage system at time t, P B,i (t) represents the frequency regulation power allocated to energy storage system i at time t.

[0030] The specific method for constructing the spatiotemporal filter based on ensemble empirical mode decomposition in step 2 is as follows:

[0031] The spatiotemporal filter based on ensemble empirical mode decomposition is expressed as:

[0032]

[0033] In the formula, X(t) is the original signal, and imf i (t) represents the intrinsic mode components of the original signal after ensemble empirical mode decomposition, r n (t) represents the decomposition remainder, and n represents the number of intrinsic modal components;

[0034] Construct spatiotemporal filters based on different feature scales of the original signal. The low-pass filter is represented as:

[0035]

[0036] A high-pass filter is represented as:

[0037]

[0038] In equations (4) and (5), X lf (t) represents the low-frequency signal of the original signal, X hf (t) represents the high-frequency signal of the original signal, and k represents the order of the spatiotemporal filter.

[0039] The specific method for determining the optimal order k of the spatiotemporal filter in step 2 is as follows:

[0040] The method for determining the order k with the overall stable maximum output of thermal power units as the objective is as follows:

[0041] X lf,k (t)=X(t)-X hf (t), k=1,2,3,...,n(6)

[0042]

[0043] In the formula, X lf,k (t) represents the low-frequency signal when the time-space filter order is k, X hf,k (t) represents the high-frequency signal when the time-space filter order is k;

[0044] Record the maximum value of the frequency regulation power command borne by the thermal power unit as a whole under each filtering result. The specific expression is as follows:

[0045] X lf,max (t)=[X lf,max,1 (t)X lf,max,2 (t)...X lf,max,k (t)](8),

[0046] In the formula, X lf,max (t) is the matrix representing the maximum frequency regulation power command undertaken by the entire thermal power unit, X lf,max,k (t) represents the maximum frequency modulation power command undertaken by the thermal power unit as a whole when the order of the spatiotemporal filter is k;

[0047] Calculate and obtain the root mean of the maximum frequency regulation power command and the maximum ramp rate of the thermal power unit under each filtering result. Select the order corresponding to the minimum root mean as the order k of the spatiotemporal filter for this application. The expression for the root mean is:

[0048]

[0049] In the formula, ST(k) is the root mean, which also represents the filter order selection function; P G,climb,max This represents the maximum gradient rate for the entire thermal power unit.

[0050] The frequency regulation power commands for the thermal power unit as a whole and the frequency regulation power commands for the energy storage system as a whole obtained in step 2, and the specific steps for adjusting the frequency regulation power commands according to different situations are as follows:

[0051] The frequency regulation power command undertaken by the thermal power unit as a whole is:

[0052]

[0053] The frequency regulation power command undertaken by the energy storage system as a whole is:

[0054]

[0055] When the overall frequency regulation power command of the thermal power unit exceeds the limit, but the overall frequency regulation power command of the energy storage system does not exceed the limit, the frequency regulation power command is adjusted as follows:

[0056]

[0057]

[0058] In the formula, For the revised overall output command of thermal power units, P G,max (t) represents the maximum power output command that the thermal power unit can deliver at time t. For the revised overall output command of the energy storage system, P B,total,max (t) represents the maximum power command that the energy storage system can output at time t;

[0059] When the overall frequency regulation power command of the thermal power unit does not exceed the limit, but the overall frequency regulation command of the energy storage system exceeds the limit, the frequency regulation power command is adjusted as follows:

[0060]

[0061]

[0062] When both the overall frequency regulation power command of the thermal power unit and the overall energy storage system exceed the limit, the frequency regulation power command is adjusted as follows:

[0063]

[0064] The mathematical formula for the multi-objective optimization problem in step 2 is:

[0065]

[0066] In the formula, F1(t) is objective function 1, representing the total frequency regulation cost of the multi-energy storage system at time t; F2(t) is objective function 2, representing the overall SOC state of the multi-energy storage system at time t; C(P B,i (t) is the cost objective function, which is related to the frequency modulation power at time t; SOC(P) B,i (t) is the SOC equalization control objective function, which is related to the frequency modulation power at time t; g j (X) is the inequality constraint function; h k (X) represents the equality function constraint.

[0067] The objective functions for frequency regulation costs of each energy storage system in step 2 are as follows:

[0068]

[0069]

[0070]

[0071] In the formula, Let be the capacity investment cost of the i-th energy storage system. The energy loss cost of the i-th energy storage system, Let the lifetime depreciation cost be that of the i-th energy storage system. Let s be the unit capacity cost of the i-th energy storage system. e Let c be the feed-in tariff for the i-th energy storage system. P,i Let be the unit power cost of the i-th energy storage system. The charging power of the i-th energy storage system. For the discharge power of the i-th energy storage system, Let i be the rated power of the i-th energy storage system. Let be the charging efficiency of the i-th energy storage system. Let be the discharge efficiency of the i-th energy storage system. Let T be the rated capacity of the i-th energy storage system, r be the annual interest rate (r = 8%), and T be the annual interest rate. f,i Let N be the float charge lifetime of the i-th energy storage system. o,i Let t be the number of cycles for the i-th energy storage system under full charge and full discharge conditions. oc Let Δt be the number of scheduling attempts, and Δt be the optimization period.

[0072] The objective function for the frequency regulation cost of each frequency regulation power station is shown in the following formula:

[0073]

[0074] In the formula, F 1,i This represents the sum of frequency regulation costs of energy storage system i at time t;

[0075] When a regional power grid contains i energy storage systems, the objective function for the frequency regulation cost of all energy storage systems at time t is as follows:

[0076]

[0077] The specific method for obtaining the SOC comprehensive deviation objective function in step 2 is as follows:

[0078] The SOC integrated control model of the energy storage system is constructed as shown in the following equation:

[0079] SOC bias,i (t)=(SOC i (t-1)+△SOC i (t)-SOC mean (t-1)) 2 (twenty one)

[0080]

[0081] In the formula, SOC bias,i (t) represents the overall SOC deviation of energy storage system i, where SOC i (t-1) represents the SOC of energy storage system i at the previous moment, and ΔSOC i (t) represents the change in SOC of energy storage system i at that moment, where SOC mean (t-1) represents the average SOC of the entire energy storage system at the previous moment;

[0082] The SOC (Solution of Cost) overall deviation objective function is as follows:

[0083]

[0084] In the formula, F2 represents the sum of the SOC equilibrium deviations of all energy storage systems at time t.

[0085] In step 2, the objective functions are subject to the following constraints:

[0086]

[0087] P B,total (t) represents the frequency regulation power command undertaken by the energy storage system as a whole; P B,i (t) represents the frequency regulation power allocated to energy storage system i at time t;

[0088] The output of each energy storage system during each frequency regulation optimization cycle should be less than the maximum output limit, as shown in the following formula:

[0089]

[0090] In the formula, This is the minimum output power limit for energy storage system i; This is the maximum output power limit for energy storage system i;

[0091] The capacity of each energy storage system must be maintained within the threshold range during each frequency regulation optimization cycle to reduce lifespan reduction, as shown in the following formula:

[0092]

[0093] In the formula, SOC i (t) represents the SOC value of energy storage system i at time t; Let i be the minimum SOC limit for energy storage system i; This represents the maximum SOC limit for energy storage system i.

[0094] The overall satisfaction matrix in step 2 is as follows:

[0095]

[0096] In the formula, W ij (t) is the overall satisfaction matrix at time t; U ij v1(t) is the objective function satisfaction matrix of each solution in the Pareto optimal solution set at time t; v2(t) is the weight coefficient of the frequency modulation cost objective function at time t; and v2(t) is the weight coefficient of the SOC comprehensive deviation objective function at time t.

[0097] Through the above design scheme, the present invention can bring the following beneficial effects:

[0098] This invention presents a two-layer optimization method for frequency regulation power of a thermal power unit-multi-energy storage system based on ensemble empirical mode decomposition and multi-objective genetic algorithm. Addressing the issues of high overall operating costs and poor frequency regulation performance resulting from the participation of multiple energy storage systems in frequency regulation, this invention optimizes the thermal power unit-energy storage system frequency regulation power layer within a regional power grid two-layer strategy framework. It constructs a spatiotemporal filter based on ensemble empirical mode decomposition to determine the frequency regulation power commands undertaken by the thermal power unit and the multiple energy storage system. Further optimization is performed at the multiple energy storage system frequency regulation power layer, establishing a model with two optimization objectives: frequency regulation cost and State of Charge (SOC). This model is then solved using a multi-objective genetic algorithm to derive the output commands for each energy storage system. Simulation analysis based on this method and comparison with traditional strategy scenarios demonstrate the method's systematic and scientific nature. Attached Figure Description

[0099] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0100] Figure 1 This is a flowchart of a two-layer optimization method for frequency regulation power in a fire-multiple-storage system according to the present invention;

[0101] Figure 2 This is a schematic diagram of the working principle of the dual-layer control center in the dual-layer optimization method for frequency regulation power of a fire-multiple energy storage system according to the present invention.

[0102] Figure 3 This is a flowchart of the two-layer control center in the two-layer optimization method for frequency regulation power of a fire-multiple storage system according to the present invention;

[0103] Figure 4 This is a schematic diagram of a combined frequency regulation system for a thermal-multi-storage system, as described in an embodiment of a dual-layer optimization method for frequency regulation power in a thermal-multi-storage system according to the present invention.

[0104] Figure 5 This is a diagram showing the command following performance curves of thermal power units under different strategies in an embodiment of a dual-layer optimization method for frequency regulation power of a thermal-multi-storage system according to the present invention.

[0105] Figure 6 This is a diagram showing the overall energy storage output curves under different strategies in an embodiment of the dual-layer optimization method for frequency regulation power of a fire-multiple-storage system according to the present invention.

[0106] Figure 7 This is a comparison diagram of the power output of each energy storage under different strategies in an embodiment of the dual-layer optimization method for frequency regulation power of a fire-multiple energy storage system according to the present invention.

[0107] Figure 8 This is a comparison of the SOC variation curves of various energy storage coefficients under different strategies in an embodiment of the dual-layer optimization method for frequency regulation power of a fire-multiple energy storage system according to the present invention.

[0108] Figure 9 This is a diagram showing the system frequency deviation under different strategies in an embodiment of a two-layer optimization method for frequency regulation power in a fire-multiple-storage system according to the present invention. Detailed Implementation

[0109] The present invention will be further described in detail below with reference to specific embodiments. The following examples are used to illustrate the present invention, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0110] like Figures 1 to 4As shown, this invention discloses a two-layer optimization method for frequency regulation power of a thermal power-multi-storage system. This method employs ensemble empirical mode decomposition (EEMD) and multi-objective genetic algorithm (MOGA), including: first, establishing the relationship between control centers of different frequency regulation systems; second, constructing a two-layer optimization model for frequency regulation power of the thermal power-multi-storage system; then, designing the optimization layers for both the thermal power-storage system and the multi-storage system; solving the constructed multi-objective function; and finally obtaining the two-layer optimization model for frequency regulation power of the thermal power-multi-storage system. The frequency regulation power command of this model guides the thermal power-multi-storage system to output power. The specific steps are as follows:

[0111] Step 1: Establish relationships between control centers of different frequency modulation systems.

[0112] A two-tiered control center is established for the thermal power-multi-energy storage joint frequency regulation system: the upper layer is the thermal power-energy storage frequency regulation system control center, whose purpose is to receive the overall automatic generation control (AGC) commands, and through Empirical Mode Decomposition (EEMD), generate thermal power-energy storage frequency regulation power commands, which are then distributed between the thermal power units and the energy storage system as a whole. These thermal power-energy storage frequency regulation power commands include those for the thermal power units as a whole and those for the energy storage system as a whole. The lower layer is the multi-energy storage frequency regulation system control center, whose purpose is to receive the frequency regulation power commands transmitted from the upper-level control center for the energy storage system as a whole, and distribute them among the various energy storage systems. The specific relationships are as follows:

[0113] P AGC (t)=P G (t)+P B,total (t) (1)

[0114]

[0115] In the formula, P AGC (t) represents the total frequency modulation power demand at time t, P G (t) represents the frequency regulation power allocated to the thermal power unit at time t, P B,total (t) represents the frequency regulation power allocated to the entire energy storage system at time t, P B,i (t) represents the frequency regulation power allocated to energy storage system i at time t;

[0116] The relationship between the different frequency modulation system control centers mentioned in step 1 is as follows: Figure 1 ;

[0117] Step 2: Construct a two-level optimization model for the frequency regulation power of the thermal power-multiple energy storage system.

[0118] First, the design of the thermal power-storage frequency regulation power optimization layer was carried out.

[0119] The first step involves constructing a spatiotemporal filter based on the ensemble empirical mode decomposition method. The result of decomposing the original signal using the ensemble empirical mode decomposition method is shown below:

[0120]

[0121] In the formula, X(t) is the original signal, and imf i (t) represents the intrinsic mode components of the original signal after ensemble empirical mode decomposition, r n (t) represents the decomposition remainder, and n represents the number of intrinsic mode components; a spatiotemporal filter based on different feature scales of the original signal is constructed, consisting of a low-pass filter and a high-pass filter, with the specific expressions as follows:

[0122]

[0123]

[0124] In the formula, X lf (t) represents the low-frequency signal of the original signal, X hf (t) represents the high-frequency signal of the original signal, and k represents the order of the spatiotemporal filter;

[0125] The second step is to determine the optimal order k of the spatiotemporal filter. The method for determining the order k is established with the goal of achieving stable maximum output of the thermal power unit, as shown below:

[0126] X lf,k (t)=X(t)-X hf (t), k=1,2,3,...,n(6)

[0127]

[0128] In the formula, X lf,k (t) represents the low-frequency signal when the time-space filter order is k, X hf,k (t) represents the high-frequency signal when the time-space filter order is k; the maximum value of the frequency modulation signal carried by the thermal power unit under each filtering result is recorded, and the specific expression is as follows:

[0129] X lf,max (t)=[X lf,max,1 (t)X lf,max,2 (t)...X lf,max,k (t)](8),

[0130] In the formula, X lf,max (t) is the matrix representing the maximum value of the frequency regulation command undertaken by the thermal power unit, X lf,max,k (t) represents the maximum value of the frequency modulation command undertaken by the thermal power unit when the order of the spatiotemporal filter is k; based on this, the root mean of the maximum value of the frequency modulation signal undertaken by the thermal power unit and the maximum ramp rate are calculated for each filtering result. The order corresponding to the minimum root mean is selected as the order k of the spatiotemporal filter for this application. The expression for the root mean is:

[0131] In the formula, ST(k) is the filter order selection function, and P G,climb,max This represents the maximum gradeability of the thermal power unit.

[0132] The third step is the allocation of frequency regulation commands between thermal power units and energy storage units based on spatiotemporal filters. The specific expressions are as follows:

[0133]

[0134]

[0135] The fourth step is to adjust the frequency regulation power command as follows when the frequency regulation command of the thermal power unit exceeds the limit, but the overall frequency regulation command of the energy storage does not exceed the limit:

[0136]

[0137]

[0138] In the formula, For the revised output command of thermal power units, P G,max (t) represents the maximum power output command of the thermal power unit at time t. For the revised overall energy storage output command, P B,total,max (t) represents the maximum power output command that the entire energy storage system can deliver at time t;

[0139] When the frequency regulation command of the thermal power unit does not exceed the limit, but the overall frequency regulation command of the energy storage exceeds the limit, the frequency regulation power command is adjusted as follows:

[0140]

[0141]

[0142] When both the frequency regulation commands for thermal power units and energy storage exceed the limits, the frequency regulation power command is adjusted as follows:

[0143]

[0144] Secondly, the design of the frequency regulation power optimization layer for multi-energy storage systems is discussed.

[0145] The first step is to determine the multi-objective optimization problem to be studied. For the two optimization objectives of frequency regulation cost and State of Charge (SOC), the mathematical description of the multi-objective optimization problem is as follows:

[0146]

[0147] In the formula, F1(t) is objective function 1, representing the total frequency regulation cost of the multi-energy storage system at time t; F2(t) is objective function 2, representing the overall SOC state of the multi-energy storage system at time t; and C(P) is the total frequency regulation cost of the multi-energy storage system at time t. B,i (t) is the cost objective function, which is related to the frequency regulation power of energy storage station i at time t, SOC(P) B,i (t) is the SOC equalization control objective function, which is related to the frequency regulation power of energy storage station i at time t, g j (X) is the inequality constraint function, h k (X) represents the equality function constraint;

[0148] The second step is to construct a frequency regulation power optimization model for multiple energy storage systems. The frequency regulation cost function for each energy storage system is shown below:

[0149]

[0150]

[0151]

[0152] In the formula, Let be the capacity investment cost of the i-th energy storage system. The energy loss cost of the i-th energy storage system, Let the lifetime depreciation cost be that of the i-th energy storage system. Let s be the unit capacity cost of the i-th energy storage system. e Let c be the feed-in tariff for the i-th energy storage system. P,i Let be the unit power cost of the i-th energy storage system. The charging power of the i-th energy storage system. For the discharge power of the i-th energy storage system, Let i be the rated power of the i-th energy storage system. Let be the charging efficiency of the i-th energy storage system. Let be the discharge efficiency of the i-th energy storage system. Let T be the rated capacity of the i-th energy storage system, r be the annual interest rate (r = 8%), and T be the annual interest rate. f,i Let N be the float charge lifetime of the i-th energy storage system. o,i Let t be the number of cycles for the i-th energy storage system under full charge and full discharge conditions. oc Let Δt be the number of scheduling attempts, and Δt be the optimization period.

[0153] The objective function for frequency regulation cost of different frequency regulation power plants is shown in the following formula:

[0154]

[0155] In the formula, F 1,iLet represent the sum of frequency regulation costs of energy storage system i at time t; if the regional power grid contains i energy storage systems, then the objective function for the frequency regulation costs of all energy storage systems at time t is as follows:

[0156]

[0157] The third step is to construct the SOC integrated control model of energy storage system i, as shown in the following equation:

[0158] SOC bias,i (t)=(SOC i (t-1)+△SOC i (t)-SOC mean (t-1)) 2 (twenty one)

[0159]

[0160] In the formula, SOC bias,i (t) represents the overall SOC deviation of energy storage system i, where SOC i (t-1) represents the SOC of energy storage system i at the previous moment, and ΔSOC i (t) represents the change in SOC of energy storage system i at that moment, where SOC mean (t-1) represents the average SOC of the energy storage system at the previous moment. The objective function for the comprehensive SOC deviation of the energy storage power station is as follows:

[0161]

[0162] In the formula, F2 represents the sum of the SOC equilibrium deviations of all energy storage systems at time t;

[0163] The fourth step is to design constraints on the objective function. The sum of the frequency regulation outputs of each energy storage system in each frequency regulation optimization cycle should equal the total frequency regulation demand of the entire energy storage system, as shown in the following formula:

[0164]

[0165] The output of each energy storage system during each frequency regulation optimization cycle should be less than the maximum output limit, as shown in the following formula:

[0166]

[0167] In the formula, These are the minimum and maximum output power limits for energy storage system i, respectively; the capacity of each energy storage system must be maintained within a reasonable range during each frequency regulation optimization cycle to reduce lifespan loss, as shown in the following formula:

[0168] In the formula, These are the minimum and maximum SOC limits for energy storage system i, respectively.

[0169] Then, the multi-objective problem of the genetic algorithm is solved:

[0170] The first step is to determine the optimal compromise solution based on fuzzy set theory. The satisfaction level corresponding to each objective function in each solution of the optimal solution set is represented by a fuzzy membership function as follows:

[0171]

[0172] In the formula, u ij (t) represents the satisfaction value of the j-th objective function in the i-th solution at time t, f ij (t) represents the value of the j-th objective function in the i-th solution at time t, f ij,max (t) represents the maximum value of the j-th objective function in the i-th solution at time t, f ij,min (t) represents the minimum value of the j-th objective function in the i-th solution at time t;

[0173] Based on the above formula, the objective function satisfaction matrix of each solution in the Pareto optimal solution set at time t can be obtained:

[0174]

[0175] The second step involves constructing an adaptive weighting coefficient matrix for the overall system state, taking into account both the urgency of power system frequency adjustment needs and the degree of overall SOC deviation.

[0176] V(t) = [v1(t)v2(t)](29),

[0177] In the formula, V(t) is the adaptive weight coefficient matrix of system frequency regulation demand and SOC comprehensive deviation at time t, v1(t) is the weight coefficient of frequency regulation cost objective function at time t, and v2(t) is the weight coefficient of SOC comprehensive deviation objective function at time t.

[0178] The relationship between the value of v1(t) and the frequency modulation requirement at each time point:

[0179]

[0180] The sum of the SOC deviations of each energy storage unit and the overall energy storage system is defined as SOC. total As shown in the following formula:

[0181]

[0182] In the formula, SOC i (t-1) represents the SOC of each energy storage at the previous moment. avg(t-1) represents the average SOC of the entire energy storage system at the previous moment; v2(t) is determined by the sum of the SOC deviations at each moment.

[0183]

[0184] In the formula, SOC total,max This is the maximum value of the sum of the deviations between the SOC of each energy storage unit and the overall SOC of the energy storage system;

[0185] The third step is to obtain the overall satisfaction matrix at time t. The method for obtaining the overall satisfaction matrix is ​​shown in the following formula:

[0186]

[0187] The two-layer optimization control strategy block described in step 2 is as follows: Figure 2 As shown, the solution process is as follows: Figure 4 ;

[0188] Step 3, evaluate the control effect

[0189] The first step is to calculate the number of deviations from the traditional unit schedule, T. dev This is used to evaluate the ability of thermal power units to follow the original AGC commands under different strategies, as shown in the following formula:

[0190]

[0191] In the formula, T dev (t) represents the ability of the thermal power unit to follow the original AGC command at time t. This is true when the frequency modulation command of the thermal power unit is opposite in direction to the original AGC command, the frequency modulation command is greater than the original AGC command (AGC command greater than 0), or the frequency modulation command is less than the original AGC command (AGC command less than 0). dev (t) is denoted as 1; T dev Used to evaluate the ability of thermal power units to follow the original AGC command under different strategies, T dev The smaller the size, the stronger the ability to follow.

[0192] The second step is to calculate the unit energy consumption cost, which represents the overall unit energy consumption cost of multiple energy storage systems within the total scheduling cycle, as shown in the following formula:

[0193]

[0194] In the formula, C B,i E represents the frequency regulation cost of energy storage system i within the scheduling period t; η This represents the unit energy consumption cost of the multi-energy storage system within the total scheduling cycle, expressed in yuan / kWh. It is used to evaluate the overall energy utilization efficiency of the multi-energy storage system; the lower the value, the better the utilization efficiency.

[0195] The third step is to determine the SOC balance of multiple energy storage systems, which is used to evaluate the degree of deviation between each energy storage system and the average SOC of the overall energy storage system, as shown in the following formula:

[0196]

[0197] In the formula, SOC avg (t) represents the mean SOC of the entire multi-energy storage system at time t; SOC total,bias Used to evaluate the degree of deviation between each energy storage system and the average SOC of the overall energy storage system. total,bias The smaller the value, the better the balance of each energy storage SOC, and the stronger the sustainability of the overall energy storage's participation in frequency regulation;

[0198] The fourth step is to calculate the root mean of the frequency deviation, which indicates the frequency fluctuation, as shown in the following formula:

[0199]

[0200] In the formula, △f i The frequency deviation of the system at time i represents the root mean of the system's frequency deviation, f. rms The smaller the value, the smaller the frequency fluctuation and the better the frequency tuning effect of the strategy.

[0201] Step 4: Perform corresponding simulation analysis on specific examples using the analysis method of this invention.

[0202] In MATLAB, a regional power grid containing thermal power units and multiple types of energy storage systems suitable for frequency regulation is simulated. Then, a two-level optimization model for frequency regulation power of the thermal-multi-storage system is constructed to perform frequency regulation. The frequency regulation effect of thermal-storage system is analyzed by comparing it with the frequency regulation scenario under the traditional strategy.

[0203] Example:

[0204] In MATLAB Figure 4 The regional power grid shown is simulated and analyzed, including thermal power units and three different types of energy storage systems suitable for frequency regulation. Frequency regulation is carried out according to the strategy designed by the optimization analysis method of this invention, and the effect of thermal power-energy storage frequency regulation is analyzed by comparing it with the frequency regulation scenario under the traditional strategy.

[0205] First, simulate the actual power generation control command data of a certain region, with a sampling interval of 1 minute. Figure 3 The steps shown first involve signal decomposition and determining the filter order, then outputting the thermal power-storage frequency modulation output power command. The power output of the thermal power unit is as follows: Figure 5 As shown, the overall energy storage output is as follows: Figure 6 As shown in the figure, compared with traditional control strategies, the superiority of this method can be clearly seen through economic analysis.

[0206] Secondly, the power commands for each energy storage system are optimized and allocated using a multi-objective genetic algorithm. The power allocated to each system is as follows: Figure 7 As shown, compared with traditional control methods, the total frequency regulation cost analysis shows that the proposed optimization method effectively reduces the frequency regulation cost. Then, the technical indicators are compared and analyzed with traditional optimization methods, such as... Figure 8 The SOC state change curves of each energy storage power station shown demonstrate how this method effectively maintains SOC consistency. Figure 9 The system frequency deviations under different strategies shown indicate that the frequency modulation effect of this method is superior.

[0207] The embodiments of the present invention are not exhaustive and do not constitute a limitation on the scope of protection of the claims. Those skilled in the art, upon learning from the embodiments of the present invention, can conceive of other substantially equivalent alternatives without inventive effort, all of which are within the scope of protection of the present invention.

Claims

1. A two-level optimization method for frequency regulation power in a thermal power-multiple energy storage system, characterized by: The steps include the following steps, and the following steps are performed in sequence: Step 1: Establish a two-tiered control center in the fire-multiple storage system and determine the relationship between the two control centers. The dual-layer control center includes an upper-layer thermal-storage frequency regulation system control center and a lower-layer multi-energy storage frequency regulation system control center. The thermal-storage frequency regulation system control center receives the overall automatic generation control (AGC) command, generates thermal-storage frequency regulation power commands through Empirical Mode Decomposition (EEMD), and distributes them between the thermal power units and the energy storage systems. The thermal-storage frequency regulation power commands include those for the thermal power units and those for the energy storage systems. The multi-energy storage frequency regulation system control center receives the frequency regulation power commands for the energy storage systems from the upper-layer thermal-storage frequency regulation system control center and distributes them among the various energy storage systems. Step 2: Construct a two-level optimization model for the frequency regulation power of the thermal power-multiple energy storage system. The construction of the dual-layer optimization model for the frequency regulation power of the thermal-multi-storage system includes the construction of a thermal-storage frequency regulation power optimization layer and a multi-storage system frequency regulation power optimization layer. 1) Construct a thermal-storage frequency modulation power optimization layer; ① Construct a spatiotemporal filter based on ensemble empirical mode decomposition (EEMD); ② Determine the optimal order of the spatiotemporal filter ; ③ Based on the determination of the optimal order The spatiotemporal filter performs frequency regulation power command allocation between thermal power units and energy storage systems to obtain the frequency regulation power command undertaken by the thermal power unit as a whole and the frequency regulation power command undertaken by the energy storage system as a whole, and adjusts the frequency regulation power command according to different situations; 2) Construct a frequency regulation power optimization layer for multi-energy storage systems; ①To address the two optimization objectives of the total frequency regulation cost of the multi-energy storage system and the overall SOC state of the multi-energy storage system, a mathematical formula for the multi-objective optimization problem is established; ② Construct a frequency regulation power optimization model for multiple energy storage systems and obtain the objective function for the frequency regulation cost of each energy storage system; ③ Construct a comprehensive control model of the state of charge (SOC) of the energy storage system and obtain the objective function of the comprehensive SOC deviation; ④ Apply conditional constraints to each objective function; 3) Use genetic algorithms to solve the mathematical formulas of multi-objective optimization problems; ① Use fuzzy set theory to find the optimal solution set of the mathematical formula for multi-objective optimization problems, express the satisfaction corresponding to each objective function in each solution in the optimal solution set as a function, and obtain the objective function satisfaction matrix of each set of solutions in the optimal solution set; ② Construct an adaptive weighting coefficient matrix for the system frequency regulation demand and the overall deviation of SOC, and then obtain the overall satisfaction matrix; ③The solution corresponding to the maximum value of the comprehensive satisfaction matrix is ​​the optimal output of the fire-multiple storage system; Step 3: Determine whether the remaining frequency regulation capacity of the energy storage is sufficient; If the remaining operable frequency regulation capacity of all energy storage systems is greater than or equal to the frequency regulation power command value allocated to the upper-level multi-energy storage system as a whole, it is determined that the remaining frequency regulation capacity of energy storage is sufficient. Then, the dual-layer optimization model of the fire-multi-energy storage system frequency regulation power outputs the optimal output as the frequency regulation power command of each energy storage system. If the remaining operable frequency regulation capacity of all energy storage systems is less than the frequency regulation power command value allocated to the upper-level multi-energy storage system as a whole, it is determined that the remaining frequency regulation capacity of energy storage is insufficient. Then, the dual-layer optimization model of the fire-multi-energy storage system frequency regulation power output takes the maximum operable frequency regulation power output as the frequency regulation power command of each energy storage system for output. The specific method for constructing the spatiotemporal filter based on ensemble empirical mode decomposition in step 2 is as follows: The spatiotemporal filter based on ensemble empirical mode decomposition is expressed as: , In the formula, The original signal, These are the intrinsic mode components obtained by ensemble empirical mode decomposition of the original signal. To decompose the remainder, The number of intrinsic modal components; Construct spatiotemporal filters based on different feature scales of the original signal. The low-pass filter is represented as: , A high-pass filter is represented as: , In the formula, The low-frequency signal of the original signal. The high-frequency signal of the original signal. The order of the spatiotemporal filter; In step 2, the optimal order of the spatiotemporal filter is determined. The specific method is as follows: Establish the order with the goal of achieving the overall stable maximum output of thermal power units. The method for determining it is as follows: ， , In the formula, The order of the spatiotemporal filter is Low-frequency signals at that time The order of the spatiotemporal filter is High-frequency signals at that time; Record the maximum value of the frequency regulation power command borne by the thermal power unit as a whole under each filtering result. The specific expression is as follows: , In the formula, This is the matrix representing the maximum frequency regulation power command undertaken by the entire thermal power unit. The order of the spatiotemporal filter is The maximum frequency regulation power command undertaken by the entire thermal power unit at that time; Calculate and obtain the root mean of the maximum frequency regulation power command and the maximum ramp rate of the thermal power unit under each filtering result, and select the order corresponding to the minimum root mean as the order of the spatiotemporal filter for this application. The expression for the root mean is: , In the formula, is the root mean, which also represents the filter order selection function; This represents the maximum gradient rate for the entire thermal power unit. The frequency regulation power commands for the thermal power unit as a whole and the frequency regulation power commands for the energy storage system as a whole obtained in step 2, and the specific steps for adjusting the frequency regulation power commands according to different situations are as follows: The frequency regulation power command undertaken by the thermal power unit as a whole is: , The frequency regulation power command undertaken by the energy storage system as a whole is: , When the overall frequency regulation power command of the thermal power unit exceeds the limit, but the overall frequency regulation power command of the energy storage system does not exceed the limit, the frequency regulation power command is adjusted as follows: , In the formula, This is the revised overall output command for thermal power units. for At any given time, the thermal power unit can output its maximum power command. This is the revised overall output command for the energy storage system. for The energy storage system can output maximum power commands at any time. When the overall frequency regulation power command of the thermal power unit does not exceed the limit, but the overall frequency regulation command of the energy storage system exceeds the limit, the frequency regulation power command is adjusted as follows: , When both the overall frequency regulation power command of the thermal power unit and the overall energy storage system exceed the limit, the frequency regulation power command is adjusted as follows: ； The overall satisfaction matrix in step 2 is as follows: , In the formula, for A comprehensive satisfaction matrix at any given time; for The objective function satisfaction matrix of each solution in the Pareto optimal solution set at time step; for The weighting coefficients of the objective function for frequency modulation cost at any given time. for The weighting coefficients of the objective function of the overall deviation of SOC at time step.

2. The two-layer optimization method for frequency regulation power of a fire-multiple energy storage system according to claim 1, characterized in that: The relationship between the control centers of each frequency modulation system in step 1 is as follows: ， , In the formula, for Total power demand for frequency modulation at any given time. for The frequency regulation power allocated to the entire thermal power unit at all times. for The frequency regulation power allocated to the entire energy storage system at all times. for Real-time energy storage system The allocated frequency modulation power.

3. The two-layer optimization method for frequency regulation power of a fire-multiple-storage system according to claim 1, characterized in that: The mathematical formula for the multi-objective optimization problem in step 2 is: , In the formula, Let the objective function be 1, representing Total frequency regulation cost of a multi-energy storage system at any time; Let the objective function be 2, representing The overall SOC status of the multi-energy storage system at all times; The objective function is the cost function, which is related to... time It is related to the frequency modulation power; Let SOC be the objective function for equalization control, and its relationship with... time It is related to the frequency modulation power; This is an inequality constraint function; This is an equality function constraint.

4. The two-layer optimization method for frequency regulation power of a fire-multiple-storage system according to claim 1, characterized in that: The objective functions for frequency regulation costs of each energy storage system in step 2 are as follows: ， ， , In the formula, For the first The capacity investment cost of an energy storage system For the first Energy loss cost of an energy storage system For the first The lifespan depreciation cost of an energy storage system For the first Unit capacity cost of an energy storage system For the first The grid connection price of an energy storage system For the first Unit power cost of an energy storage system For the first The charging power of an energy storage system For the first The discharge power of an energy storage system For the first The rated power of an energy storage system For the first The charging efficiency of an energy storage system For the first Discharge efficiency of an energy storage system For the first The rated capacity of an energy storage system The annual interest rate is Take 8%, For the first The float charge lifespan of an energy storage system For the first The number of cycles a single energy storage system performs under full charging and discharging conditions. For the number of scheduling, To optimize the cycle; The objective function for the frequency regulation cost of each frequency regulation power station is shown in the following formula: ， In the formula, express Real-time energy storage system The sum of frequency modulation costs; Regional power grid contains When there is an energy storage system, then The objective function for the frequency regulation cost of all energy storage systems at any given time is as follows: 。 5. The two-layer optimization method for frequency regulation power of a fire-multiple-storage system according to claim 1, characterized in that: The specific method for obtaining the SOC comprehensive deviation objective function in step 2 is as follows: The SOC integrated control model of the energy storage system is constructed as shown in the following equation: ， , In the formula, For energy storage systems SOC overall deviation, For energy storage systems The previous SOC, For energy storage systems The change in SOC at that moment, This represents the average SOC of the entire energy storage system at the previous moment; The SOC (Solution of Cost) overall deviation objective function is as follows: , In the formula, express The sum of the SOC balance deviations of all energy storage systems at any given time.

6. The two-layer optimization method for frequency regulation power of a fire-multiple-storage system according to claim 1, characterized in that: In step 2, the objective functions are subject to the following constraints: , In the formula, The frequency regulation power command undertaken by the energy storage system as a whole; for Real-time energy storage system The allocated frequency modulation power; The output of each energy storage system during each frequency regulation optimization cycle should be less than the maximum output limit, as shown in the following formula: , In the formula, For energy storage systems Minimum output power limit; For energy storage systems Maximum output power limit; The capacity of each energy storage system must be maintained within the threshold range during each frequency regulation optimization cycle to reduce lifespan reduction, as shown in the following formula: , In the formula, For energy storage systems exist SOC value at time t; For energy storage systems Minimum SOC limit; For energy storage systems Maximum SOC limit.

Images