A PID parameter tuning method for wind power generation system

Abstract

The present invention relates to the field of wind power generation technology, and more particularly to a method for PID parameter tuning for a wind power generation system. The method collects wind speed data from a random wind speed model and actual generator speed data, calculates the error between the optimal generator speed and the actual generator speed, uses the calculated error as the input signal for a fractional-order PID controller, and optimizes the parameters of the fractional-order PID controller using a differential evolution algorithm. The method utilizes a differential evolution algorithm, which has fast convergence speed and high search accuracy, improving parameter tuning speed and accuracy while offering greater robustness. Furthermore, the method has low reliance on motor parameters during the tuning process, is highly versatile, and can be widely used in engineering practice.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a PID parameter tuning method for a wind power generation system. Background Art

[0002] In recent years, wind energy, a green, renewable energy source, has been widely used worldwide, becoming a key solution to the global energy crisis. Wind turbines are currently the most efficient way to harvest wind energy. Maximizing the conversion of wind energy into electricity is a key concern in this research field.

[0003] Wind power generation systems cannot convert all captured wind energy into electrical energy, and usually use the wind energy utilization coefficient C p To express the conversion efficiency of wind turbines in converting captured wind energy into electrical energy. Wind turbines have nonlinear characteristics. Wind turbine torque control and output power are directly affected by wind speed. Wind speed changes in wind fields are random and uncertain. The wind energy utilization coefficient C of wind power generation systems is p The maximum power point position of the operating output will change with the change of wind speed v, which is non-linear. The maximum power point position refers to the maximum wind energy utilization coefficient C. p When the generator speed value of the wind power generation system is opt .

[0004] Assuming that the wind speed v is constant, from the wind power equation (ρ is air density, R is the radius of the wind wheel) It can be found that the wind power generation power P w Wind energy utilization coefficient C p When the wind energy utilization coefficient C p When the maximum value is obtained, the wind power generation system can achieve the maximum power under the current wind speed v. In order to improve the conversion efficiency of the wind power generation system and make the wind energy utilization rate reach the ideal state, it is necessary to adopt maximum power point tracking (MPPT) control, which controls the optimal generator speed value ω according to different wind speed v control inputs. opt , the generator speed ω is controlled by the PID control system, so that the speed of the wind power generation system tends to the maximum power point and the maximum power tracking is achieved.

[0005] The control effect of the PID control system on the generator speed ω depends largely on the parameter values ​​of the PID controller. If the parameters are set improperly, the control results will have large deviations.

[0006] The traditional PID control system is generally a linear control system. The setting of its control parameters mostly relies on empirical methods. It needs to refer to the motor speed, d-axis current, q-axis current and other parameters for continuous debugging to obtain the control parameter proportional adjustment coefficient K. p , integral adjustment coefficient K i and differential adjustment coefficient K d , the versatility is poor, and when the PID control system uses the control parameters obtained by the empirical method to control the generator speed, there will still be a large control deviation, and the maximum power tracking cannot be achieved well. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to overcome the deficiencies in the prior art and provide a method for adjusting PID parameters of a wind power generation system.

[0008] The present invention is achieved through the following technical solutions:

[0009] A method for adjusting PID parameters of a wind power generation system comprises the following steps:

[0010] Step S10, building a random wind speed model and a permanent magnet synchronous power generation simulation model including a fractional-order PID controller;

[0011] Step S20: collect wind speed data of the random wind speed model in real time, and calculate the optimal generator speed ω at time t based on the wind speed v(t) at time t. opt (t);

[0012] Step S30: collect the actual speed data of the generator in real time and calculate the optimal speed of the generator ω opt The error e(t) between the actual generator speed ω(t) collected at time t is,

[0013] e(t)=ω opt (t)-ω(t);

[0014] In step S40 , the error e(t) is used as an input signal of the fractional-order PID controller, and the parameters of the fractional-order PID controller are optimized using a differential evolution algorithm.

[0015] Preferably, the control equation of the fractional-order PID controller is:

[0016] u(t)=K p e(t)+K i D -∝ e(t)+K d D μ e(t)

[0017] Where u(t) is the output signal of the fractional-order PID controller, which is used to control the generator speed; ∝,β is the order of the fractional-order control, ∝,μ∈[0,1.5]; K p is the proportional adjustment coefficient; K i is the integral adjustment coefficient; K d is the differential adjustment coefficient; operator D -∝ represents the fractional-order Grünwald-Letnikov type integral operator; D μ represents the fractional-order Grünwald-Letnikov type differential operator,

[0018] The parameters for optimizing the fractional-order PID controller in step S40 include ∝, μ, K p ,K i ,K d .

[0019] Preferably, the step S40 specifically includes:

[0020] Step S41: Generate an initial population, set the parameter dimension M to be optimized, set the population size NP, and represent each individual in the population as an M-dimensional vector

[0021]

[0022] Where i = 0, 1, ..., NP-1; g represents the iteration number of the current population. Let g = 0 and randomly generate the initial population.

[0023] Step S42, set the maximum population iteration number g max , set the comparison constant θ, determine the optimization objective function, the optimization objective function is as follows,

[0024]

[0025] Among them, f is the fitness value,

[0026] If the fitness value of the g-th generation population individual is less than θ, or g=g max , then the output Finish;

[0027] Otherwise, execute step S43;

[0028] Step S43, perform mutation operation, set mutation operator F, and generate mutation vector

[0029]

[0030] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; V ig represents the i-th mutant individual in the g-th generation population; r0,r1,r2∈{0,1,…,NP-1}, and r0,r1,r2r0,r1,r2 are different from each other;

[0031] Step S44, perform a recombination operation to obtain a new temporary population, set the crossover factor CR, and generate a test vector

[0032]

[0033] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th experimental individual in the g-th generation population;

[0034] Step S45, a selection operation is performed. The fitness values ​​of the original population individuals and the temporary population individuals are calculated by optimizing the objective function. Individuals with good fitness values ​​are retained, and the obtained individuals enter the next generation population. The selection process is as follows:

[0035]

[0036] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th selected individual in the g+1 generation population; Represents an individual The fitness value of Represents an individual The fitness value of

[0037] Step S46, let g increase by one, and go to step S42.

[0038] Preferably, for step S41, NP individuals in the initial population are obtained by an initialization formula, which is as follows:

[0039]

[0040] Where j = 0, 1, ..., M-1; is the upper limit of the j-th dimension value of each individual; is the lower limit of the j-th dimension value of each individual; rand i,j (0,1)) is a random number uniformly distributed between [0,1]. Preferably, for step S43, the mutation operator F is adjusted as follows:

[0041]

[0042] Among them, g max is the maximum number of iterations; g is the current number of iterations; F MIN is the minimum value of the mutation operator; FMAX is the maximum value of the mutation operator.

[0043] Preferably, for step S42,

[0044] If the error e(t)<0, then as the optimization objective function.

[0045] Preferably, for step S46, the value of θ is 10 -6 .

[0046] The beneficial effects of the present invention are:

[0047] The present invention adopts the differential evolution algorithm to optimize the parameters of the fractional-order PID controller, which has fast convergence speed and high search accuracy, improves the parameter tuning speed and accuracy, has better robustness, and has little dependence on motor parameters during the tuning process of this method. It has strong versatility and can be widely used in engineering practice. Since the fractional-order PID control has two more degrees of freedom, α and β, than the conventional PID control, it is more flexible and can effectively improve the control performance. The mutation operator F gradually becomes smaller. In the early stage, F takes a larger value to increase the global search ability of the algorithm. In the later stage, F should take a smaller value to increase the ability to search locally and improve the search accuracy and search speed of the algorithm. The penalty mechanism can avoid overshoot and achieve stable operation of the generator. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a structural block diagram of the permanent magnet synchronous power generation simulation model of the present invention;

[0049] Figure 2 This is a flow chart for parameter optimization of fractional-order PID controller based on differential evolution algorithm;

[0050] Figure 3 is the wind energy utilization coefficient C p Function curve graph of . DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and the best embodiments.

[0052] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0053] like Figure 1-Figure 3 As shown, the present invention provides a method for adjusting PID parameters of a wind power generation system, comprising the following steps:

[0054] Step S10, building a wind speed model and a permanent magnet synchronous power generation simulation model including a fractional-order PID controller, wherein the wind speed model is an existing model. Specifically, the permanent magnet synchronous power generation simulation model includes a wind turbine, a generator, a machine-side converter, a grid-side converter and a master control system. The master control system includes an MPPT control system, a machine-side control system and a grid-side control system. The fractional-order PID controller is located in the machine-side control system. The wind speed model is used to generate natural random wind. The wind turbine is used to receive the wind energy of the random wind generated by the wind speed model and convert the wind energy into mechanical energy. The generator is used to convert the mechanical energy converted by the wind turbine into electrical energy. The MPP The T control system is used to obtain the wind speed of the random wind generated by the wind speed model, and to obtain the optimal speed of the generator through the obtained wind speed. The machine-side control system is used to collect the actual speed of the generator. The fractional-order PID controller outputs a generator control signal according to the actual speed of the generator and the optimal speed of the generator obtained by the MPPT control system, thereby adjusting the generator torque and realizing variable-speed operation of the generator, thereby realizing maximum power tracking. At the same time, the machine-side control system is used to control the machine-side converter to convert the AC power with unstable voltage and frequency emitted by the generator into DC power with constant voltage. The grid-side control system is used to control the grid-side converter to incorporate the current processed by the machine-side converter into the power grid;

[0055] Step S20: collect wind speed data from the wind speed model in real time, and calculate the optimal generator speed ω at time t based on the wind speed v(t) at time t. opt (t),

[0056] It is known that when the wind energy utilization coefficient C p When the maximum value is achieved, the wind power generation system can achieve the maximum power under the current wind speed v. According to the aerodynamic theory of wind turbines, there is the following relationship:

[0057]

[0058]

[0059] Where β is the pitch angle; λ is the tip speed ratio,

[0060] By C p Function Curve Figure 3 It can be seen that when the pitch angle β = 0, there is an optimal tip speed ratio λ opt Corresponding maximum wind energy utilization coefficient C p , the optimal generator speed ω opt (t) is obtained by the following formula:

[0061] ωopt =λ opt v / R

[0062] Where v is the current wind speed and R is the radius of the wind turbine blade;

[0063] Step S30: collect the actual speed data of the generator in real time and calculate the optimal speed of the generator ω opt The error e(t) between the actual speed of the generator ω(t) collected at time t,

[0064] e(t)=ω opt (t)-ω(t);

[0065] In step S40, the error e(t) is used as the input signal of the fractional-order PID controller, and the parameters of the fractional-order PID controller are optimized using the differential evolution algorithm. The differential evolution algorithm has fast convergence speed and high search accuracy, which improves the parameter tuning speed and accuracy and has better robustness. In addition, the method has little dependence on motor parameters during the tuning process, is highly versatile, and can be widely used in engineering practice.

[0066] The control equation of the fractional-order PID controller is:

[0067] u(t)=K p e(t)+K i D -∝ e(t)+K d D μ e(t)

[0068] Where u(t) is the output signal of the fractional-order PID controller, which is used to control the generator speed; ∝,μ is the order of the fractional-order control, ∝,μ∈[0,1.5]; K p is the proportional adjustment coefficient; K i is the integral adjustment coefficient; K d is the differential adjustment coefficient; operator D -∝ represents the fractional-order Grünwald-Letnikov type integral operator; D μ represents the fractional-order Grünwald-Letnikov type differential operator,

[0069] The parameters for optimizing the fractional-order PID controller in step S40 include ∝, μ, K p ,K i ,K d Since fractional-order PID control has two more degrees of freedom, ∝ and μ, than conventional PID control, it is more flexible and can effectively improve control performance.

[0070] The step S40 specifically includes:

[0071] Step S41: Generate an initial population, set the dimension of the parameter to be optimized M=5, set the population size NP, and represent each individual in the population as an M-dimensional vector

[0072]

[0073] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population,

[0074] Let g = 0, randomly generate the initial population, and the NP individuals in the initial population are obtained by the initialization formula, which is as follows;

[0075]

[0076] Where j = 0, 1, ..., M-1; is the upper limit of the j-th dimension value of each individual; is the lower limit of the j-th dimension value of each individual; rand i,j (0,1)) is a random number uniformly distributed between [0,1];

[0077] Step S42, set the maximum population iteration number g max , set the comparison constant θ=10 -6 , determine the optimization objective function, the optimization objective function is as follows,

[0078]

[0079] Among them, f is the fitness value, which can well reflect the control effect of the fractional-order PID controller. The smaller the fitness value f, the better the control effect of the fractional-order PID controller.

[0080] Determine whether the termination condition is met. If the fitness value corresponding to the individual in the g-th generation population is less than θ, or g = g max , then output the optimal individual Finish;

[0081] Otherwise, execute step S43;

[0082] Step S43, perform mutation operation, set mutation operator F, and generate mutation vector

[0083]

[0084] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; V i g represents the i-th mutant individual in the g-th generation population; r0,r1,r2∈{0,1,…,NP-1}, and r0,r1,r2r0,r1,r2 are different from each other;

[0085] Step S44, perform a recombination operation to obtain a new temporary population, set the crossover factor CR, CR is used to control the degree of participation of each dimension of the individual parameter in the crossover process, ensure the balance between global and local search capabilities, and generate a test vector

[0086]

[0087] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th experimental individual in the g-th generation population;

[0088] Step S45, a selection operation is performed. The fitness values ​​of the original population individuals and the temporary population individuals are calculated by optimizing the objective function. Individuals with good fitness values ​​are retained, and the obtained individuals enter the next generation population. The selection process is as follows:

[0089]

[0090] Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th selected individual in the g+1 generation population; Represents an individual The fitness value of Represents an individual The fitness value of

[0091] Step S46, let g increase by one, and go to step S42.

[0092] For step S43, the mutation operator F is adjusted as follows:

[0093]

[0094] Among them, g max is the maximum number of iterations; g is the current number of iterations; F MIN is the minimum value of the mutation operator; F MAX is the maximum value of the mutation operator,

[0095] It can realize the gradual reduction of the mutation operator F. In the early stage, F takes a larger value to increase the global search ability of the algorithm. In the later stage, F should take a smaller value to increase the ability to search locally and improve the search accuracy and speed of the algorithm.

[0096] For step S42, if the error e(t) < 0, a penalty mechanism is added. As the optimization objective function, overshoot is avoided and stable operation of the generator is achieved.

[0097] The present invention adopts the differential evolution algorithm to optimize the parameters of the fractional-order PID controller, which has fast convergence speed and high search accuracy, improves the parameter tuning speed and accuracy, has better robustness, and has little dependence on motor parameters during the tuning process of this method. It has strong versatility and can be widely used in engineering practice. Since the fractional-order PID control has two more degrees of freedom, α and μ, than the conventional PID control, it is more flexible and can effectively improve the control performance. The mutation operator F gradually becomes smaller. In the early stage, F takes a larger value to increase the global search ability of the algorithm. In the later stage, F should take a smaller value to increase the ability to search locally and improve the search accuracy and search speed of the algorithm. The penalty mechanism can avoid overshoot and achieve stable operation of the generator.

[0098] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for PID parameter tuning of a wind power generation system, characterized in that: The steps include: Step S10, building a random wind speed model and a permanent magnet synchronous power generation simulation model including a fractional-order PID controller; Step S20, collecting wind speed data of the random wind speed model in real time, and calculating the optimal generator speed ωopt(t) at time t based on the wind speed v(t) at time t; Step S30: collect the actual speed data of the generator in real time, and calculate the error e(t) between the optimal speed ωopt(t) of the generator and the actual speed ω(t) of the generator collected at time t. e(t)=ωopt(t)-ω(t); In step S40, the error e(t) is used as the input signal of the fractional-order PID controller, and the parameters of the fractional-order PID controller are optimized using a differential evolution algorithm. The control equation of the fractional-order PID controller is: u(t) Kpe(t)+KiD -∝ e(t)+KdD μ e(t) Where u(t) is the output signal of the fractional-order PID controller, which is used to control the generator speed; ∝,μ is the order of the fractional-order control, ∝,μ∈[0,1.5]; Kp is the proportional adjustment coefficient; Ki is the integral adjustment coefficient; Kd is the differential adjustment coefficient; operator D -∝ represents the fractional-order Grünwald-Letnikov type integral operator; D μ represents the fractional-order Grünwald-Letnikov type differential operator, The parameters for optimizing the fractional-order PID controller in step S40 include ∝, μ, Kp, Ki, Kd The step S40 specifically includes: step S41, generating an initial population, setting the dimension M of the parameter to be optimized, setting the population size NP, and representing each individual in the population as an M-dimensional vector Where i = 0, 1, ..., NP-1; g represents the iteration number of the current population. Let g = 0 and randomly generate the initial population; Step S42, set the maximum population iteration number gmax, set the comparison constant θ, and determine the optimization objective function. The optimization objective function is as follows: Among them, f is the fitness value, If the fitness value of the g-th generation population individual is less than θ, or g=gmax, then the output is Finish; Otherwise, execute step S43; Step S43, perform mutation operation, set mutation operator F, generate mutation vector Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th mutant individual in the g-th generation population; r0, r1, r2∈{0,1,…,NP-1}, and r0, r1, r2 are different from each other; Step S44, perform a recombination operation to obtain a new temporary population, set the crossover factor CR, and generate a test vector Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th experimental individual in the g-th generation population; Step S45, a selection operation is performed. The fitness values ​​of the original population individuals and the temporary population individuals are calculated by optimizing the objective function. Individuals with good fitness values ​​are retained, and the obtained individuals enter the next generation population. The selection process is as follows: Where i = 0, 1, ..., NP-1; g represents the number of iterations of the current population; represents the i-th selected individual in the g+1 generation population; express The fitness value of the individual; express The fitness value of the individual; Step S46, let g increase by one, and go to step S42.

2. A method for PID parameter tuning of a wind power generation system according to claim 1, characterized in that: For step S41, the NP individuals in the initial population are obtained by the initialization formula, which is as follows: Where j = 0, 1, ..., M-1; is the upper limit of the j-th dimension value of each individual; is the lower limit of the j-th dimension value of each individual; randi,j(0,1) is a random number uniformly distributed between [0,1].

3. A method for PID parameter tuning of a wind power generation system according to claim 1, characterized in that: For step S43, the mutation operator F is adjusted as follows: Among them, gmax is the maximum number of iterations; g is the current number of iterations; FMIN is the minimum value of the mutation operator; FMAX is the maximum value of the mutation operator.

4. A method for PID parameter tuning of a wind power generation system according to claim 1, characterized in that: For step S42, If the error e(t)<0, then as the optimization objective function.

5. A method for PID parameter tuning of a wind power generation system according to claim 1, characterized in that: For step S42, the value of θ is 10 -6 .

Images