A method for optimizing control parameters of virtual synchronous generator

Abstract

The present invention is a method for optimizing the control parameters of a virtual synchronous generator, comprising the following steps: establishing a mathematical model of a virtual synchronous generator; constructing an objective function for the optimization problem of the control parameters of a virtual synchronous generator; optimizing the control parameters of the virtual synchronous generator using an improved artificial jellyfish search algorithm; inputting the optimized control parameters into the virtual synchronous generator; establishing an inverter control simulation model based on the virtual synchronous generator to verify the control effect of the virtual synchronous generator after the control parameters are optimized; and displaying the control results of the virtual synchronous generator after the control parameters are optimized. The present invention improves the control performance of the traditional virtual synchronous generator on the inverter, is more conducive to improving the inertia and damping characteristics of the power system, enhancing the anti-disturbance capability of the power system, and is helpful to improve the grid-connected penetration rate of renewable energy.

Description

Technical Field

[0001] The invention belongs to the technical field of electric power systems, and in particular to an optimization method for controlling parameters of a virtual synchronous generator. Background Art

[0002] With the rapid development of global science and technology and economy, energy demand has increased dramatically, and fossil energy supply is insufficient, making the development and utilization of renewable energy more and more important in today's energy industry. As representatives of renewable energy, wind and solar energy have attracted much attention due to their low cost, clean and environmentally friendly, and recyclable advantages. The installed capacity of wind and photovoltaic power generation has been increasing worldwide.

[0003] At present, wind power and photovoltaic power generation are greatly affected by seasons, weather and environmental factors. The change of non-continuous natural wind will affect the output power of wind power generation, and the uncertainty of solar radiation intensity will cause the power fluctuation of photovoltaic power generation, that is, the output power of wind power and photovoltaic power generation has obvious volatility. The inverter is the key device responsible for managing the power transmission between wind power and photovoltaic power generation and the power grid, but compared with the traditional synchronous generator, the inverter is very sensitive to the external operating environment and lacks anti-disturbance ability. With the development of renewable energy, traditional synchronous generators are gradually being replaced by inverters, resulting in a decrease in the damping and inertia characteristics of the power system. Therefore, when wind power and photovoltaic power generation are connected to the grid through the inverter, if the control strategy of the inverter is improper, it will cause the frequency and voltage parameters of the power system to fluctuate, resulting in a decrease in power quality, which is not conducive to the stable operation of the power system. Virtual synchronous generator technology is introduced into the control of the inverter. The virtual synchronous generator can simulate the transient characteristics of the synchronous generator, and can give the inverter inertia and damping characteristics when controlling, which improves the anti-disturbance ability of the system to a certain extent. Among them, the selection of the control parameters of the virtual synchronous generator directly affects the control performance of the virtual synchronous generator. However, with the increasing penetration rate of renewable energy grid connection, the fixed control parameters in traditional virtual synchronous generators limit the control ability of virtual synchronous generators over inverters. Traditional virtual synchronous generators can no longer meet the damping and inertia requirements for stable operation of power systems.

[0004] In summary, the strong volatility of wind and photovoltaic power output power and the weak anti-disturbance capability of inverters have brought difficulties to the grid connection of renewable energy. How to improve the control capability of traditional virtual synchronous generators over inverters and enhance the anti-disturbance capability of power systems has become an urgent problem to be solved. Summary of the invention

[0005] The technical problem to be solved by the present invention is: to provide a method for optimizing the control parameters of a virtual synchronous generator, to establish a virtual synchronous generator model when the penetration rate of renewable energy grid connection increases, resulting in reduced inertia and damping of the power system and reduced system operation stability, to enhance the inertia and damping characteristics of the system and improve the voltage and frequency regulation capabilities of the system; to combine the virtual synchronous generator with an improved artificial jellyfish search algorithm, and to propose a method for optimizing the control parameters of a virtual synchronous generator based on the improved artificial jellyfish search algorithm; to apply the improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator, so that the virtual synchronous generator has better inertia and damping characteristics, to enhance the control performance of the virtual synchronous generator, and to further enhance the anti-disturbance capability of the power system.

[0006] The present invention solves the technical problem by adopting the following technical solutions:

[0007] A method for optimizing control parameters of a virtual synchronous generator comprises the following steps:

[0008] Step 1, establishing a mathematical model of a virtual synchronous generator;

[0009] Step 2: construct the objective function of the virtual synchronous generator control parameter optimization problem;

[0010] Step 3, using an improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator;

[0011] Step 4, inputting the optimally selected control parameters into the virtual synchronous generator;

[0012] Step 5, establishing an inverter control simulation model based on a virtual synchronous generator to verify the control effect of the virtual synchronous generator after the control parameters are optimized;

[0013] Step 6, displaying the control result of the virtual synchronous generator after the control parameters are optimized;

[0014] Step 1: Establish a mathematical model of the virtual synchronous generator

[0015] (1.1) Establishing the mathematical model of the virtual synchronous generator reactive power-voltage control structure

[0016] The reactive power-voltage control structure of the virtual synchronous generator adjusts the transmitted reactive power when the voltage amplitude of the AC grid changes, which is equivalent to simulating the excitation system of the synchronous generator. The reactive power-voltage control equations are shown in Equations (1) and (2):

[0017]

[0018] ΔQ=K q (U N -U)+Qref -Q (2)

[0019] In the formula, E m represents the virtual induced electromotive force amplitude output by the virtual synchronous generator, E0 is the no-load electromotive force, ΔQ represents the reactive power deviation, 1 / (K·s) represents an integral link, K is the reactive integral gain coefficient, which is used to simulate the excitation system of the synchronous generator, K q is the reactive power droop coefficient, Q ref Represents the reactive power command value, U represents the input voltage of the reactive-voltage control structure, and U N is the reference voltage of the reactive-voltage control structure, Q represents the input reactive power of the reactive-voltage control structure;

[0020] (1.2) Establishing the mathematical model of the active power-frequency control structure of the virtual synchronous generator

[0021] The active power-frequency control structure of the virtual synchronous generator adjusts the output active power when the phase of the AC power grid changes, which is equivalent to simulating the speed governor and rotor movement of the synchronous generator. The active power-frequency control mechanism of the virtual synchronous generator includes two parts: the virtual speed governor and the virtual rotor. The control equation of the virtual speed governor is shown in equations (3) and (4):

[0022] P m =P ref +K p (w N -w L ) (3)

[0023] w L =2πf (4)

[0024] Where P m Indicates the active power output by the virtual speed regulator, P ref Indicates the active power command value, K p is the active power droop coefficient, w N represents the rated angular velocity, f represents the AC grid frequency, w L represents the angular velocity calculated according to the AC grid frequency;

[0025] The motion equations of the virtual rotor are shown in equations (5), (6) and (7):

[0026]

[0027]

[0028] ΔP=P m -P (7)

[0029] Where J represents the moment of inertia of the virtual rotor, D represents the damping coefficient of the virtual rotor, w represents the output angular velocity of the rotor motion equation, Δw represents the error angular velocity, ΔT represents the error torque, and T m represents virtual mechanical torque, T represents virtual electromagnetic torque, P m is the output power of the virtual speed regulator and also the input active power of the virtual rotor motion equation, P is the active power of the AC power grid, and ΔP represents the difference between the output power of the virtual speed regulator and the active power of the AC power grid;

[0030] Step 2: Construct the objective function of the virtual synchronous generator control parameter optimization problem

[0031] Select the control parameter K of the virtual synchronous generator q , D and J as optimization targets, where K q is the reactive droop coefficient in reactive-voltage control, D is the damping coefficient of the virtual rotor in active-frequency control, and J is the moment of inertia of the virtual rotor in active-frequency control. The objective function of the virtual synchronous generator control parameter optimization problem based on time multiplied by the integral of absolute error (ITAE) is constructed. The time multiplied by the integral of absolute error is the integral of the absolute value of the error multiplied by the time term over time. The objective function of the control parameter optimization is set to the sum of the ITAE of frequency and voltage, and the weight coefficient a is introduced. The expression of the objective function is shown in formula (8):

[0032] Obj=a×ITAE f +(1-a)×ITAE u (8)

[0033] Where Obj represents the objective function value, a and 1-a are the weight coefficients of frequency and voltage ITAE respectively, and ITAE f and ITAE u They represent the ITAE of frequency and voltage respectively. The calculation methods of the ITAE of frequency and voltage are shown in equations (9) and (10):

[0034]

[0035]

[0036] In the formula, t0 represents the control time, Δf(t0) represents the frequency error value, Δu(t0) represents the voltage error value, f(t0) represents the real-time frequency effective value, u(t0) represents the real-time voltage effective value, f N The rated frequency is 50Hz, u N The rated voltage is 220V;

[0037] Step 3: Use the improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator

[0038] (3.1) Setting the parameters of the improved artificial jellyfish search algorithm and the range of optimization targets

[0039] The maximum number of iterations of the improved artificial jellyfish search algorithm is Iter max , set Iter max = 100 times, set the population size M = 30, set the control parameter K selected by the improved artificial jellyfish search algorithm in the virtual synchronous generator q , D and J value range, where K q The value range of is set to [1, 500], the value range of D is set to [5, 50], the value range of J is set to [0.01, 5], and the population dimension of the improved artificial jellyfish search algorithm is set to d = 3;

[0040] (3.2) Initialize the jellyfish population using the Tent chaos mapping method

[0041] Record the initial position of the jellyfish in the improved artificial jellyfish search algorithm and save it as matrix X t , as shown in formula (11):

[0042]

[0043] In the formula, is the position of the jth jellyfish individual in the i-th jellyfish population in the t-th iteration calculation; d is the jellyfish population dimension. In the virtual synchronous generator control parameter optimization problem, d=3 is set. Each jellyfish population contains 3 jellyfish individuals. The 3 jellyfish individuals are used to search for the performance parameters K of the virtual synchronous generator. q , D and J; m represents the number of jellyfish populations, set m = M = 30; t is the current iteration number of the improved artificial jellyfish search algorithm, when the initial position matrix is ​​used as the position matrix of the current iteration, t = 0, when the updated position matrix is ​​used as the position matrix of the next iteration, t = 1, 2, ..., Iter max ;

[0044] The chaotic mapping method is used to initialize the jellyfish population, generate a uniformly distributed jellyfish population in the solution space, and obtain the position of the initialized jellyfish. Here, the original population initialization strategy is improved by introducing Tent mapping instead of Logistic mapping. The population initialization formulas before and after the improvement are shown in equations (12) and (13):

[0045] Before improvement:

[0046] X i+1 =ηX i (1-X i ), 0<X i<1 (12)

[0047] Where, X i represents the jellyfish in the generated chaotic sequence, where i∈[1, 2, ..., m], m represents the number of jellyfish populations to be initialized, m=M=30 is set, η is the typing parameter of the Logistic chaotic mapping, and η=4 is set in the artificial jellyfish search algorithm;

[0048] After improvement:

[0049]

[0050] Where, α is the adjustment parameter of Tent chaotic mapping, and its value range is [0, 1]. The value of α determines the distribution characteristics of the generated chaotic sequence. In the improved artificial jellyfish search algorithm, α is set to 0.5;

[0051] (3.3) Calculate the fitness values ​​of all jellyfish positions and record the position of the jellyfish with the best current fitness

[0052] The fitness function of the improved artificial jellyfish search algorithm for the virtual synchronous generator control parameter optimization problem is established. The fitness function of the algorithm is set equal to the objective function Obj. The solution with a smaller fitness value has a more suitable corresponding control parameter for the virtual synchronous generator. The fitness function Fit is shown in formula (14):

[0053] Fit=Obj=a×ITAE f +(1-a)×ITAE u (14)

[0054] Calculate the fitness values ​​of all jellyfish positions and record the position of the jellyfish with the best current fitness. The position of the jellyfish with the best current fitness is represented by X best ; Save the fitness value of the jellyfish position as the fitness matrix F (t) , as shown in formula (15):

[0055]

[0056] In the formula, F (t) is the fitness matrix of the jellyfish population in the tth iteration, m represents the number of jellyfish populations to be initialized, and d is the dimension of the jellyfish population;

[0057] (3.4) Calculate the time control function value, determine the movement mode of the jellyfish and update the position of the jellyfish population, calculate the fitness value, record and update the global optimal individual position

[0058] The time control function c(t) of the improved artificial jellyfish search algorithm is shown in formula (16):

[0059]

[0060] Where t is the current iteration number of the improved artificial jellyfish search algorithm, Iter max is the maximum number of iterations of the improved artificial jellyfish search algorithm, rand(0,1) represents a random number between 0 and 1;

[0061] According to the calculated time control function value, the movement mode of the jellyfish is determined, and different position update strategies are executed to update the position of the jellyfish;

[0062] When the value of c(t) is greater than or equal to 0.5, the jellyfish follows the ocean current and updates its position through position update strategy 1. When the value of c(t) is less than 0.5, the jellyfish moves within the population and performs a secondary discrimination, comparing the value of 1-c(t) with a randomly generated random number rand(0, 1) between 0 and 1. If rand(0, 1) is greater than 1-c(t), the jellyfish performs improved type A movement and updates its position through position update strategy 2. If rand(0, 1) is less than or equal to 1-c(t), the jellyfish performs type B movement and updates its position through position update strategy 3.

[0063] Specifically, the position update strategy 1 is a position update formula of the jellyfish when the jellyfish follows the ocean current, as shown in formula (17):

[0064] X i (t+1)=X i (t)+rand(0,1)×(X best -β×rand(0,1)×μ) (17)

[0065] Where, X i (t) represents the current position of the jellyfish, X i (t+1) represents the next moment's position after the jellyfish's position is updated, X best is the position of the jellyfish with the best fitness at the moment, i.e., the control parameter K of the virtual synchronous generator in the current iteration q , D, and J; β represents the distribution coefficient, set β = 3, rand (0, 1) represents a random number between 0 and 1, and μ represents the average position of all jellyfish in the population;

[0066] The position update strategy 2 is the position update formula of the jellyfish when the jellyfish performs the improved type A motion within the population. Here, the original position update formula is improved by introducing a sinusoidal dynamic adaptive factor. The position update formulas before and after the improvement are shown in formulas (18) and (19):

[0067] Before improvement:

[0068] Xi (t+1)=X i (t)+γ×rand(0,1)×(U m -L m ) (18)

[0069] Where γ represents the motion coefficient of jellyfish, set γ = 0.1; U m represents the upper limit of the range of the virtual synchronous generator control parameters, L m It represents the lower bound of the range of control parameters of the virtual synchronous generator;

[0070] After improvement:

[0071]

[0072] Where S represents the sinusoidal dynamic adaptive factor, Iter max represents the maximum number of iterations of the improved artificial jellyfish search algorithm, and t represents the current number of iterations of the improved artificial jellyfish search algorithm;

[0073] The position update strategy 3 is a position update formula for jellyfish when they perform type B motion in the population. When jellyfish perform type B motion, the jellyfish whose position is currently to be updated is jellyfish i. Another jellyfish is randomly selected from the population, called jellyfish j. The position of jellyfish i is updated according to the position of jellyfish j. The position update formula for jellyfish when they perform type B motion is shown in equations (20) and (21):

[0074]

[0075]

[0076] Where, X i (t) represents the current position of jellyfish i, X j (t) represents the current position of jellyfish j, X i (t+1) represents the next moment's position after the jellyfish's position is updated, S T represents the step length of jellyfish i, D ir represents the moving direction of jellyfish i, f(X i (t)) represents the objective function for jellyfish i, f(X j (t)) represents the objective function with respect to jellyfish j;

[0077] (3.5) Update the current position of the jellyfish population through mutation operation, calculate the fitness value of the population, record and update the position of the optimal individual

[0078] After the jellyfish population completes the position update according to the position update formula, the mutation operation is introduced. The specific operations are as follows:

[0079] Randomly select a jellyfish individual X from the jellyfish population k , and then randomly select three jellyfish individuals, calculate the fitness values ​​of the individuals and sort them according to the fitness. The three jellyfish individuals are represented as X a , X b , X c , the fitness values ​​are f a 、f b 、f c , X a Is the individual with the best fitness; update X k The current position of X k The position update formula is shown in equation (22) and equation (23):

[0080] X k =X a +δ(X b -X c ) (twenty two)

[0081]

[0082] In the formula, δ is the mutation operator, δ u represents the upper bound of the mutation rate, δ l Represents the lower bound of the mutation rate, set δ u =0.9,δ l =0.1; Repeat the mutation operation to form a mutant jellyfish population;

[0083] Generate a random number R between 0 and 1 and compare R with the random crossover parameter R C The size of the random number R is greater than R C , then use the jellyfish individual before mutation to replace the jellyfish individual corresponding to the mutant population; if the value of the random number R is less than R C , then the position of the individual jellyfish in the mutant population remains unchanged;

[0084] Calculate the fitness value of the individual jellyfish in the jellyfish population after the mutation and crossover operations, and compare it with the fitness value of the jellyfish population before the mutation. If the fitness of the individual jellyfish after the mutation and crossover operations is better, replace the corresponding position of the individual jellyfish before the mutation, and complete the position update of the jellyfish population after the mutation operation.

[0085] (3.6) Determine whether the algorithm's end condition is met

[0086] If the current number of iterations t does not reach the maximum number of iterations Iter max , then start the next iteration and repeat steps (3.4) and (3.5); if the current number of iterations t reaches the maximum number of iterations Iter max, the algorithm ends and executes step (3.7);

[0087] (3.7) Output optimization results

[0088] Output the optimal parameter value corresponding to the optimal jellyfish position and obtain the control parameter K of the virtual synchronous generator q , the optimal solution of D and J;

[0089] Step 4: Input the optimal control parameters into the virtual synchronous generator

[0090] Step 5: Establish an inverter control simulation model based on a virtual synchronous generator to verify the control effect of the virtual synchronous generator after the control parameters are optimized.

[0091] An inverter control simulation model based on a virtual synchronous generator is built on the MATLAB / Simulink platform to simulate the operation of the power system; the simulation model includes a DC voltage source, an inverter, an RL filter module, an RC filter module, an island / grid-connected switch, an AC power grid, a load power module, an active-reactive power calculation module, a virtual synchronous generator module, and a PWM module;

[0092] When the load power changes, the traditional virtual synchronous generator and the virtual synchronous generator with optimized control parameters are used to control the inverter in the simulation model respectively. The system frequency change and system voltage change of the simulation model are compared to verify the control effect of the virtual synchronous generator with optimized control parameters.

[0093] Step 6: Display the control results of the virtual synchronous generator after the control parameters are optimized

[0094] The comparison results of the system frequency change and the system voltage change under the control of the traditional virtual synchronous generator obtained in step five and the virtual synchronous generator after the control parameters are optimized are displayed on the display screen of the computer.

[0095] The above-mentioned method for optimizing the control parameters of a virtual synchronous generator, the idea of ​​the improved artificial jellyfish search, and the establishment of an inverter control simulation model based on a virtual synchronous generator are existing technologies and are well known to those skilled in the art.

[0096] In the above-mentioned preferred method for controlling parameters of a virtual synchronous generator, the computer, display and MATLAB computer software are all commercially available.

[0097] The beneficial effects of the present invention are:

[0098] (1) The present invention provides a method for optimizing the control parameters of a virtual synchronous generator. The method combines the virtual synchronous generator with an improved artificial jellyfish search algorithm, and proposes a method for optimizing the control parameters of a virtual synchronous generator based on the improved artificial jellyfish search algorithm. The improved artificial jellyfish search algorithm is used to optimize the control parameters of the virtual synchronous generator, and the optimal moment of inertia, damping coefficient and reactive droop coefficient of the virtual synchronous generator are searched, so that the control effect of the virtual synchronous generator is better. The present invention improves the control performance of the traditional virtual synchronous generator on the inverter, is more conducive to improving the inertia and damping characteristics of the power system, enhances the anti-disturbance capability of the power system, and helps to increase the grid-connected penetration rate of renewable energy.

[0099] (2) The present invention introduces the Tent chaotic mapping initialization method into the traditional artificial jellyfish search algorithm, so that the initial position distribution of jellyfish individuals in the population is more uniform; a sinusoidal dynamic adaptive factor is introduced into the position update formula to enhance the local search ability of the algorithm; and a population mutation operation is introduced to improve the diversity of the population, which is conducive to preventing the algorithm from falling into a local optimal solution;

[0100] (3) The method for optimizing the control parameters of a virtual synchronous generator of the present invention is not limited to optimizing the control parameters of a virtual synchronous generator, but can also be extended to optimizing the control parameters of other devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0102] Figure 1 It is a flow chart of a method for optimizing control parameters of a virtual synchronous generator according to the present invention.

[0103] Figure 2 It is a flow chart of the method of the present invention using an improved artificial jellyfish search algorithm to optimize the control parameters of a virtual synchronous generator.

[0104] Figure 3 It is a reactive power-voltage control structure diagram of a virtual synchronous generator provided in an embodiment of the present invention.

[0105] Figure 4 It is a diagram of the active power-frequency control structure of a virtual synchronous generator provided in an embodiment of the present invention.

[0106] Figure 5 It is a schematic diagram of an inverter control structure based on a virtual synchronous generator provided in an embodiment of the present invention.

[0107] Figure 6 It is a comparison diagram of system frequency changes under the control of a traditional virtual synchronous generator provided by an embodiment of the present invention and a virtual synchronous generator after control parameter optimization.

[0108] Figure 7 It is a comparison diagram of system voltage changes under the control of a traditional virtual synchronous generator provided by an embodiment of the present invention and a virtual synchronous generator after control parameters are optimized. DETAILED DESCRIPTION

[0109] Figure 1 The flow of the virtual synchronous generator control parameter optimization method of the present invention is as follows: start → establish a mathematical model of the virtual synchronous generator reactive power-voltage control structure → establish a mathematical model of the virtual synchronous generator active power-frequency control structure → construct an objective function of the virtual synchronous generator control parameter optimization problem → use an improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator → input the optimized control parameters into the virtual synchronous generator → verify the control effect of the virtual synchronous generator after the control parameters are optimized → display the control result of the virtual synchronous generator after the control parameters are optimized → end;

[0110] Figure 2 The process of optimizing the control parameters of the virtual synchronous generator by using the improved artificial jellyfish search algorithm of the present invention is as follows: start → set the value range of the algorithm parameters and the optimization target → use the Tent chaotic mapping method to initialize the jellyfish population → calculate the fitness values ​​of all jellyfish positions, and record the position of the jellyfish with the best fitness at present → start iteration → calculate the time control function value c(t) → c(t) < 0.5? → No, the jellyfish follows the ocean current; Yes, rand(0,1) > 1-c(t)? → Yes, the jellyfish performs the improved A-type movement; No, the jellyfish performs B-type movement → calculate the fitness value, record and update the position of the individual with the best fitness at present → perform mutation operation on the jellyfish population → calculate the fitness value, record and update the position of the individual with the best fitness at present → t < Iter max ? →Yes, increase the number of iterations t by 1 and start the next iteration; No, end and output the optimization results.

[0111] Example 1

[0112] A method for optimizing control parameters of virtual synchronous generator

[0113] Step 1: Establish a mathematical model of the virtual synchronous generator

[0114] (1.1) Establishing the mathematical model of the virtual synchronous generator reactive power-voltage control structure

[0115] The reactive power-voltage control equations of the virtual synchronous generator are shown in equations (1) and (2);

[0116] (1.2) Establishing the mathematical model of the active power-frequency control structure of the virtual synchronous generator

[0117] The active power-frequency control equations of the virtual synchronous generator are shown in equations (3) and (4);

[0118] Step 2: Construct the objective function of the virtual synchronous generator control parameter optimization problem

[0119] In this embodiment, the control parameter K of the virtual synchronous generator is selected q , D and J as optimization targets, where K q is the reactive droop coefficient in reactive-voltage control, D is the damping coefficient of the virtual rotor in active-frequency control, and J is the moment of inertia of the virtual rotor in active-frequency control;

[0120] The objective function of the virtual synchronous generator control parameter optimization problem based on time multiplied by the integral of absolute error (ITAE) is constructed. Time multiplied by the integral of absolute error is the integral of the absolute value of the error multiplied by the time term over time. The objective function of the control parameter optimization is set to the sum of the ITAE of frequency and voltage, and the weight coefficient a is introduced. The expression of the objective function is shown in formula (8); where Obj represents the objective function value, a and 1-a are the weight coefficients of frequency and voltage ITAE respectively, and ITAE f and ITAE u They represent the ITAE of frequency and voltage respectively. The calculation methods of the ITAE of frequency and voltage are shown in equations (9) and (10); where t0 represents the control time, Δf(t0) represents the frequency error value, Δu(t0) represents the voltage error value, f(t0) represents the real-time frequency effective value, u(t0) represents the real-time voltage effective value, f N The rated frequency is 50Hz, u N The rated voltage is 220V;

[0121] Step 3: Use the improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator

[0122] (3.1) Setting the parameters of the improved artificial jellyfish search algorithm and the range of optimization targets

[0123] In this embodiment, the maximum number of iterations of the improved artificial jellyfish search algorithm is Iter max , set Iter max = 100 times; set the population size M = 30; set the performance parameter K optimized by the improved artificial jellyfish search algorithm in the virtual synchronous generator q , D and J, where the virtual synchronous generator reactive power droop coefficient K q The value range of is set to [1, 500], the value range of the damping coefficient D is set to [5, 50], and the value range of J is set to [0.01, 5]; the population dimension d of the improved artificial jellyfish search algorithm is set to 3;

[0124] (3.2) Initialize the jellyfish population using the Tent chaos mapping method

[0125] In this embodiment, the initial position of the jellyfish in the improved artificial jellyfish search algorithm is recorded and saved as a matrix X t , as shown in formula (11); where d is the jellyfish population dimension. In this embodiment, d=3 is set. Each jellyfish population contains 3 jellyfish individuals. The 3 jellyfish individuals are used to search for the performance parameters K of the virtual synchronous generator. q , D and J; m represents the number of jellyfish populations, set m = M = 30; t is the current iteration number of the improved artificial jellyfish search algorithm, when the initial position matrix is ​​used as the position matrix of the current iteration, t = 0, when the updated position matrix is ​​used as the position matrix of the next iteration, t = 1, 2, ..., Iter max ;

[0126] The Tent chaotic mapping method is used to initialize the jellyfish population and obtain the position of the initialized jellyfish. The population initialization formula is shown in formula (13); where α is the adjustment parameter of the Tent chaotic mapping, and the value range is [0, 1]. The value of α determines the distribution characteristics of the generated chaotic sequence. In the improved artificial jellyfish search algorithm, α is set to 0.5;

[0127] (3.3) Calculate the fitness values ​​of all jellyfish positions and record the position of the jellyfish with the best current fitness

[0128] In this embodiment, the fitness function of the algorithm is set to be equal to the objective function Obj, and the fitness function Fit is shown in formula (14); the position of the jellyfish with the best fitness is expressed as X best ; Save the fitness value of the jellyfish position as the fitness matrix F (t) , as shown in formula (15); where F (t) is the fitness matrix of the jellyfish population in the tth iteration, m represents the number of jellyfish populations to be initialized, and d is the dimension of the jellyfish population;

[0129] (3.4) Calculate the time control function value, determine the movement mode of the jellyfish and update the position of the jellyfish population, calculate the fitness value, record and update the global optimal individual position

[0130] In this embodiment, the time control function c(t) of the improved artificial jellyfish search algorithm is shown in formula (16); where t is the current iteration number of the improved artificial jellyfish search algorithm, Iter max is the maximum number of iterations of the improved artificial jellyfish search algorithm, rand(0,1) represents a random number between 0 and 1;

[0131] In this embodiment, the specific steps of determining the movement mode of the jellyfish according to the time control function value are as follows:

[0132] When the value of c(t) is greater than or equal to 0.5, the jellyfish follows the ocean current, and the jellyfish position update formula is shown in formula (17); where X i (t) represents the current position of the jellyfish, X i (t+1) represents the next moment's position after the jellyfish's position is updated, X best is the position of the jellyfish with the best fitness at the moment, i.e., the control parameter K of the virtual synchronous generator in the current iteration q , D, and J; β represents the distribution coefficient, set β = 3, rand (0, 1) represents a random number between 0 and 1, and μ represents the average position of all jellyfish in the population;

[0133] When the value of c(t) is less than 0.5, the jellyfish moves within the population and performs a secondary judgment, comparing the value of 1-c(t) with the randomly generated random number rand(0, 1) between 0 and 1; if rand(0, 1) is greater than the value of 1-c(t), the jellyfish performs improved type A motion, and the jellyfish position update formula is shown in formula (19); where S represents the sinusoidal dynamic adaptive factor; if rand(0, 1) is less than or equal to 1-c(t), the jellyfish performs type B motion, and the jellyfish whose current position is to be updated is jellyfish i. Another jellyfish is randomly selected from the population, called jellyfish j, and the position of jellyfish i is updated according to the position of jellyfish j. The jellyfish position update formula is shown in formulas (20) and (21); where X i (t) represents the current position of jellyfish i, X j (t) represents the current position of jellyfish j, X i (t+1) represents the next moment's position after the jellyfish's position is updated, S T represents the step length of jellyfish i, Dir represents the moving direction of jellyfish i, f(X i (t)) represents the objective function for jellyfish i, f(X j (t)) represents the objective function with respect to jellyfish j;

[0134] (3.5) Update the current position of the jellyfish population through mutation operation, calculate the fitness value of the population, record and update the position of the optimal individual

[0135] In this embodiment, the specific steps of updating the position of the jellyfish population through the mutation operation are as follows:

[0136] Randomly select a jellyfish individual X from the jellyfish population k , and then randomly select three jellyfish individuals, calculate the fitness values ​​of the individuals and sort them according to the fitness. The three jellyfish individuals are represented as X a , X b , X c, the fitness values ​​are f a 、f b 、f c , X a Is the individual with the best fitness; update X k The current position of X k The position update formula of is shown in equation (22) and equation (23); where δ is the mutation operator, δ u represents the upper bound of the mutation rate and δ l Represents the lower bound of the mutation rate, set δ u =0.9,δ l =0.1; Repeat the mutation operation to form a mutant jellyfish population;

[0137] Generate a random number R between 0 and 1 and compare R with the random crossover parameter R C The size of the random number R is greater than R C , then use the jellyfish individual before mutation to replace the jellyfish individual corresponding to the mutant population; if the value of the random number R is less than R C , then the position of the individual jellyfish in the mutant population remains unchanged;

[0138] Calculate the fitness value of the individual jellyfish in the jellyfish population after the mutation and crossover operations, and compare it with the fitness value of the jellyfish population before the mutation. If the fitness of the individual jellyfish after the mutation and crossover operations is better, replace the corresponding position of the individual jellyfish before the mutation, and complete the position update of the jellyfish population after the mutation operation.

[0139] (3.7) Output optimization results

[0140] Output the optimal parameter value corresponding to the optimal jellyfish position and obtain the reactive power droop coefficient K of the virtual synchronous generator q , the optimal solution of damping coefficient D and moment of inertia J;

[0141] Step 4: Input the optimal control parameters into the virtual synchronous generator

[0142] Specifically, the control parameters include reactive power droop coefficient K q , damping coefficient D and moment of inertia J;

[0143] Step 5: Establish an inverter control simulation model based on a virtual synchronous generator to verify the control effect of the virtual synchronous generator after the control parameters are optimized.

[0144] In this embodiment, according to Figure 5The inverter control structure diagram shown in the figure builds an inverter control simulation model based on a virtual synchronous generator on the MATLAB / Simulink platform to simulate the operation of the power system; the simulation model includes a DC voltage source, an inverter, an RL filter module, an RC filter module, an island / grid-connected switch, a load power module, an AC power grid, an active-reactive power calculation module, a virtual synchronous generator module, and a PWM module;

[0145] Optionally, the reactive power droop coefficient K of the virtual synchronous generator is optimized by an improved artificial jellyfish search algorithm q , damping coefficient D and moment of inertia J, and obtain the virtual synchronous generator after the control parameters are optimized;

[0146] In this embodiment, the main parameter settings in the inverter control simulation model based on the virtual synchronous generator are shown in Table 1:

[0147] Table 1 Parameters of the inverter control simulation model based on virtual synchronous generator

[0148]

[0149] The inverter in the simulation model is controlled by using a traditional virtual synchronous generator and a virtual synchronous generator after parameter optimization. The system frequency change and system voltage change of the simulation model are compared to verify the control effect of the virtual synchronous generator after parameter optimization.

[0150] In this embodiment, the reactive power droop coefficient K of the traditional virtual synchronous generator is q The setting parameter values ​​of the damping coefficient D and the moment of inertia J are shown in Table 1. The K of the virtual synchronous generator after the control parameter optimization is q , D and J are set as the optimal parameter values ​​obtained by the improved artificial jellyfish search algorithm; the system of the simulation model is set to an island operation state, and the system operates stably under the active load of 8kW; at time t=1s, the load with active power of 2kW is increased, and the load power increases to 10kW, causing the system load to fluctuate; at t=2s, the load with active power of 2kW is removed, and the load power is reduced to 8kW, and the frequency and voltage changes of the system are observed;

[0151] The system frequency changes under the control of the traditional virtual synchronous generator and the virtual synchronous generator after the control parameters are optimized are shown in the figure. Figure 6 As shown, Figure 6 The horizontal axis is time and the vertical axis is frequency; Figure 6 The solid line in the figure represents the system frequency change when the load power remains stable, the dotted line represents the system frequency change under the control of the traditional virtual synchronous generator when the load fluctuates, and the dashed line represents the system frequency change under the control of the virtual synchronous generator after the control parameters are optimized when the load fluctuates. Figure 6 It can be seen that when the load power decreases, the system frequency under the control of the traditional virtual synchronous generator exceeds 50Hz three times, and the oscillation amplitude is greater than the system frequency change of the virtual synchronous generator control after the control parameters are optimized;

[0152] The system voltage changes under the control of the traditional virtual synchronous generator and the virtual synchronous generator after the control parameters are optimized are shown in the figure. Figure 7 As shown, Figure 7 The horizontal axis is time and the vertical axis is voltage; Figure 7 The solid line in the figure represents the system voltage change when the load power remains stable, the dotted line represents the system voltage change under the control of the traditional virtual synchronous generator when the load fluctuates, and the dashed line represents the system voltage change under the control of the virtual synchronous generator after the control parameters are optimized when the load fluctuates. Figure 7 It can be seen that when the system load decreases, the system voltage under the control of the traditional virtual synchronous generator drops to 219.5V, while the system voltage under the control of the virtual synchronous generator after the control parameters are optimized only drops to 219V. Compared with the system under the control of the traditional virtual generator, the voltage fluctuation is smaller;

[0153] In this embodiment, the verification results show that: when the system load power fluctuates, compared with the traditional virtual synchronous generator, the virtual synchronous generator after the control parameter optimization has a better control effect on the inverter, the oscillation amplitude of the system frequency and the system voltage is smaller, and the stability is better. The optimization method of the virtual synchronous generator control parameters proposed in the present invention is conducive to improving the control ability of the virtual synchronous generator, and is more conducive to the safe and stable operation of the system;

[0154] Step 6: Display the control results of the virtual synchronous generator after the control parameters are optimized

[0155] The comparison results of the system frequency change and the system voltage change under the control of the traditional virtual synchronous generator obtained in step five and the virtual synchronous generator after the control parameters are optimized are displayed on the display screen of the computer.

[0156] It should be emphasized that the embodiments described in the present invention are illustrative rather than restrictive, and therefore the present invention is not limited to the embodiments described in the specific implementation manners. Any other implementation manners derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. A method for optimizing the control parameters of a virtual synchronous generator, characterized in that The virtual synchronous generator control parameter optimization method is based on the improved artificial jellyfish search algorithm, and its steps are as follows: Step 1: Establish a mathematical model of the virtual synchronous generator (1.1) Establishing the mathematical model of the virtual synchronous generator reactive power-voltage control structure The reactive power-voltage control structure of the virtual synchronous generator adjusts the transmitted reactive power when the voltage amplitude of the AC grid changes, which is equivalent to simulating the excitation system of the synchronous generator. The reactive power-voltage control equations are shown in Equations (1) and (2): ΔQ=K q (U N -U)+Q ref -Q (2) In the formula, E m represents the virtual induced electromotive force amplitude output by the virtual synchronous generator, E0 is the no-load electromotive force, ΔQ represents the reactive power deviation, 1 / (K·s) represents an integral link, K is the reactive integral gain coefficient, which is used to simulate the excitation system of the synchronous generator, K q is the reactive power droop coefficient, Q ref Represents the reactive power command value, U represents the input voltage of the reactive-voltage control structure, and U N is the reference voltage of the reactive-voltage control structure, Q represents the input reactive power of the reactive-voltage control structure; (1.2) Establishing the mathematical model of the active power-frequency control structure of the virtual synchronous generator The active power-frequency control structure of the virtual synchronous generator adjusts the output active power when the phase of the AC power grid changes, which is equivalent to simulating the speed governor and rotor movement of the synchronous generator. The active power-frequency control mechanism of the virtual synchronous generator includes two parts: the virtual speed governor and the virtual rotor. The control equation of the virtual speed governor is shown in equations (3) and (4): P m =P ref +K p (w N -w L ) (3) In L =2πf (4) Where P m Indicates the active power output by the virtual speed regulator, P ref Indicates the active power command value, K p is the active power droop coefficient, w N represents the rated angular velocity, f represents the AC grid frequency, w L represents the angular velocity calculated according to the AC grid frequency; the motion equations of the virtual rotor are shown in equations (5), (6) and (7): ΔP=P m -P (7) Where J represents the moment of inertia of the virtual rotor, D represents the damping coefficient of the virtual rotor, w represents the output angular velocity of the rotor motion equation, Δw represents the error angular velocity, ΔT represents the error torque, and T m represents virtual mechanical torque, T represents virtual electromagnetic torque, P m is the output power of the virtual speed regulator and also the input active power of the virtual rotor motion equation, P is the active power of the AC power grid, and ΔP represents the difference between the output power of the virtual speed regulator and the active power of the AC power grid; Step 2: Construct the objective function of the virtual synchronous generator control parameter optimization problem Select the control parameter K of the virtual synchronous generator q , D and J as optimization targets, where K q is the reactive droop coefficient in reactive-voltage control, D is the damping coefficient of the virtual rotor in active-frequency control, and J is the moment of inertia of the virtual rotor in active-frequency control. The objective function of the virtual synchronous generator control parameter optimization problem based on time multiplied by the integral of absolute error (ITAE) is constructed. The time multiplied by the integral of absolute error is the integral of the absolute value of the error multiplied by the time term over time. The objective function of the control parameter optimization is set to the sum of the ITAE of frequency and voltage, and the weight coefficient a is introduced. The expression of the objective function is shown in formula (8): Obj=a×ITAE f +(1-a)×ITAE u (8) Where Obj represents the objective function value, a and 1-a are the weight coefficients of frequency and voltage ITAE respectively, and ITAE f and ITAE u They represent the ITAE of frequency and voltage respectively. The calculation methods of the ITAE of frequency and voltage are shown in equations (9) and (10): In the formula, t0 represents the control time, Δf(t0) represents the frequency error value, Δu(t0) represents the voltage error value, f(t0) represents the real-time frequency effective value, u(t0) represents the real-time voltage effective value, f N The rated frequency is 50Hz, u N The rated voltage is 220V; Step 3, using an improved artificial jellyfish search algorithm to optimize the control parameters of the virtual synchronous generator; (3.1) Setting the parameters of the improved artificial jellyfish search algorithm and the range of optimization targets The maximum number of iterations of the improved artificial jellyfish search algorithm is Iter max , set Iter max = 100 times; set the population size M = 30; set the control parameter K selected by the improved artificial jellyfish search algorithm in the virtual synchronous generator q , D and J value range, where K q The value range of is set to [1, 500], the value range of D is set to [5, 50], and the value range of J is set to [0.01, 5]; the population dimension of the improved artificial jellyfish search algorithm is set to d = 3; (3.2) Initialize the jellyfish population using the Tent chaos mapping method Record the initial position of the jellyfish in the improved artificial jellyfish search algorithm and save it as matrix X t , as shown in formula (11): In the formula, is the position of the jth jellyfish individual in the i-th jellyfish population in the t-th iteration calculation; d is the jellyfish population dimension. In the virtual synchronous generator control parameter optimization problem, d=3 is set. Each jellyfish population contains 3 jellyfish individuals. The 3 jellyfish individuals are used to search for the performance parameters K of the virtual synchronous generator. q , D and J; m represents the number of jellyfish populations, set m = M = 30; t is the current iteration number of the improved artificial jellyfish search algorithm, when the initial position matrix is ​​used as the position matrix of the current iteration, t = 0, when the updated position matrix is ​​used as the position matrix of the next iteration, t = 1, 2, ..., Iter max ; The chaotic mapping method is used to initialize the jellyfish population, generate a uniformly distributed jellyfish population in the solution space, and obtain the position of the initialized jellyfish. Here, the original population initialization strategy is improved by introducing Tent mapping instead of Logistic mapping. The population initialization formulas before and after the improvement are shown in equations (12) and (13): Before improvement: X i+1 =ηX i (1-X i ),0<X i <1 (12) Where, X i represents the jellyfish in the generated chaotic sequence, where i∈[1, 2, ..., m], m represents the number of jellyfish populations to be initialized, m=M=30 is set, η is the typing parameter of the Logistic chaotic mapping, and η=4 is set in the artificial jellyfish search algorithm; After improvement: Where, α is the adjustment parameter of Tent chaotic mapping, and its value range is [0, 1]. The value of α determines the distribution characteristics of the generated chaotic sequence. In the improved artificial jellyfish search algorithm, α is set to 0.5; (3.3) Calculate the fitness values ​​of all jellyfish positions and record the position of the jellyfish with the best current fitness The fitness function of the improved artificial jellyfish search algorithm for the virtual synchronous generator control parameter optimization problem is established. The fitness function of the algorithm is set equal to the objective function Obj. The solution with a smaller fitness value has a more suitable corresponding control parameter for the virtual synchronous generator. The fitness function Fit is shown in formula (14): Fit=Obj=a×ITAE f +(1-a)×ITAE u (14) Calculate the fitness values ​​of all jellyfish positions and record the position of the jellyfish with the best current fitness. The position of the jellyfish with the best current fitness is represented by X best ; Save the fitness value of the jellyfish position as the fitness matrix F (t) , as shown in formula (15): In the formula, F (t) is the fitness matrix of the jellyfish population in the tth iteration, m represents the number of jellyfish populations to be initialized, and d is the dimension of the jellyfish population; (3.4) Calculate the time control function value, determine the movement mode of the jellyfish and update the position of the jellyfish population, calculate the fitness value, record and update the global optimal individual position The time control function c(t) of the improved artificial jellyfish search algorithm is shown in formula (16): Where t is the current iteration number of the improved artificial jellyfish search algorithm, Iter max is the maximum number of iterations of the improved artificial jellyfish search algorithm, rand(0,1) represents a random number between 0 and 1; According to the calculated time control function value, the movement mode of the jellyfish is determined, and different position update strategies are executed to update the position of the jellyfish; When the value of c(t) is greater than or equal to 0.5, the jellyfish follows the ocean current and updates its position through position update strategy 1. When the value of c(t) is less than 0.5, the jellyfish moves within the population and performs a secondary discrimination, comparing the value of 1-c(t) with a randomly generated random number rand(0, 1) between 0 and 1. If rand(0, 1) is greater than 1-c(t), the jellyfish performs improved type A movement and updates its position through position update strategy 2. If rand(0, 1) is less than or equal to 1-c(t), the jellyfish performs type B movement and updates its position through position update strategy 3. Specifically, the position update strategy 1 is a position update formula of the jellyfish when the jellyfish follows the ocean current, as shown in formula (17): X i (t+1)=X i (t)+rand(0,1)×(X best -β×rand(0,1)×μ) (17) Where, X i (t) represents the current position of the jellyfish, X i (t+1) represents the next moment's position after the jellyfish's position is updated, X best is the position of the jellyfish with the best fitness at the moment, i.e., the control parameter K of the virtual synchronous generator in the current iteration q , D, and J; β represents the distribution coefficient, set β = 3, rand (0, 1) represents a random number between 0 and 1, and μ represents the average position of all jellyfish in the population; The position update strategy 2 is the position update formula of the jellyfish when the jellyfish performs the improved type A motion within the population. Here, the original position update formula is improved by introducing a sinusoidal dynamic adaptive factor. The position update formulas before and after the improvement are shown in formulas (18) and (19): Before improvement: X i (t+1)=X i (t)+γ×rand(0,1)×(U m -L m ) (18) Where γ represents the motion coefficient of jellyfish, set γ = 0.1; U m represents the upper limit of the range of virtual synchronous generator control parameters, L m It represents the lower bound of the range of control parameters of the virtual synchronous generator; After improvement: Where S represents the sinusoidal dynamic adaptive factor, Iter max represents the maximum number of iterations of the improved artificial jellyfish search algorithm, and t represents the current number of iterations of the improved artificial jellyfish search algorithm; The position update strategy 3 is a position update formula for jellyfish when they perform type B motion in the population. When jellyfish perform type B motion, the jellyfish whose position is currently to be updated is jellyfish i. Another jellyfish is randomly selected from the population, called jellyfish j. The position of jellyfish i is updated according to the position of jellyfish j. The position update formula for jellyfish when they perform type B motion is shown in equations (20) and (21): Where, X i (t) represents the current position of jellyfish i, X j (t) represents the current position of jellyfish j, X i (t+1) represents the next moment's position after the jellyfish's position is updated, S T represents the step length of jellyfish i, D ir represents the moving direction of jellyfish i, f(X i (t)) represents the objective function for jellyfish i, f(X j (t)) represents the objective function with respect to jellyfish j; (3.5) Update the current position of the jellyfish population through mutation operation, calculate the fitness value of the population, record and update the position of the optimal individual After the jellyfish population completes the position update according to the position update formula, the mutation operation is introduced. The specific operations are as follows: Randomly select a jellyfish individual X from the jellyfish population k , and then randomly select three jellyfish individuals, calculate the fitness values ​​of the individuals and sort them according to the fitness. The three jellyfish individuals are represented as X a , X b , X c , the fitness values ​​are f a 、f b 、f c , X a is the individual with the best fitness; update X k The current position of X k The position update formula is shown in equation (22) and equation (23): X k =X a +δ(X b -X c ) (22) In the formula, δ is the mutation operator, δ u represents the upper bound of the mutation rate, δ l Represents the lower bound of the mutation rate, set δ u =0.9,δ l =0.1; Repeat the mutation operation to form a mutant jellyfish population; Generate a random number R between 0 and 1 and compare R with the random crossover parameter R C The size of the random number R is greater than R C , then use the jellyfish individual before mutation to replace the jellyfish individual corresponding to the mutant population; if the value of the random number R is less than R C , then the position of the individual jellyfish in the mutant population remains unchanged; Calculate the fitness value of the individual jellyfish in the jellyfish population after the mutation and crossover operations, and compare it with the fitness value of the jellyfish population before the mutation. If the fitness of the individual jellyfish after the mutation and crossover operations is better, replace the corresponding position of the individual jellyfish before the mutation, and complete the position update of the jellyfish population after the mutation operation. (3.6) Determine whether the algorithm's end condition is met If the current number of iterations t does not reach the maximum number of iterations Iter max , then start the next iteration and repeat steps (3.4) and (3.5); if the current number of iterations t reaches the maximum number of iterations Iter max , the algorithm ends and executes step (3.7); (3.7) Output optimization results Output the optimal parameter value corresponding to the optimal jellyfish position and obtain the control parameter K of the virtual synchronous generator q , the optimal solution of D and J; Step 4, inputting the optimally selected control parameters into the virtual synchronous generator; Step 5: Establish an inverter control simulation model based on a virtual synchronous generator to verify the control effect of the virtual synchronous generator after the control parameters are optimized; An inverter control simulation model based on a virtual synchronous generator is built on the MATLAB / Simulink platform to simulate the operation of the power system; the simulation model includes a DC voltage source, an inverter, an RL filter module, an RC filter module, an island / grid-connected switch, an AC power grid, a load power module, an active-reactive power calculation module, a virtual synchronous generator module, and a PWM module; When the load power changes, the traditional virtual synchronous generator and the virtual synchronous generator with optimized control parameters are used to control the inverter in the simulation model respectively. The system frequency change and system voltage change of the simulation model are compared to verify the control effect of the virtual synchronous generator with optimized control parameters. Step 6, displaying the control result of the virtual synchronous generator after the control parameters are optimized; The comparison results of the system frequency change and the system voltage change under the control of the traditional virtual synchronous generator obtained in step five and the virtual synchronous generator after the control parameters are optimized are displayed on the display screen of the computer.

2. A method for optimizing control parameters of a virtual synchronous generator according to claim 1, characterized in that: The preferred parameters of the improved artificial jellyfish search algorithm are obtained based on the mathematical model of the reactive power-voltage control structure and the active power-frequency control structure of the virtual synchronous generator.

Images