Coordinated optimization control method and device for improving active power regulation capability of wind turbine generator

Abstract

The invention discloses a coordinated optimization control method and device for improving the active power regulation capability of a wind turbine generator, and relates to the technical field of wind turbine generator control. The method comprises the following steps: constructing a first-order error active disturbance rejection controller; establishing a controller parameter optimization model with minimum unit fatigue load and optimal control performance as multiple targets; according to the operation parameters of the wind turbine generator, obtaining a fatigue load index value by using a bidirectional long-short-term memory network; according to the fatigue load index value, the output power of the wind turbine generator and the rotor speed, the controller parameter optimization model is solved by adopting a snow-melting optimization algorithm, and the optimal parameters of the two first-order error active-disturbance-rejection controllers are obtained; and respectively controlling the pitch angle and the electromagnetic torque of the wind turbine generator through two first-order error active-disturbance-rejection controllers using the optimal parameters so as to adjust the output power and the rotor speed of the wind turbine generator. The active instruction tracking capability of the wind generating set can be improved, and meanwhile, the fatigue load is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of wind turbine control, and particularly to a coordinated optimization control method and device for improving the active power regulation ability of a wind turbine. Background Art

[0002] With the rapid development of renewable energy, wind energy, as a clean and sustainable energy form, has been increasingly valued by people. As a primary conversion device for wind energy, the stable operation and long-term reliability of a wind turbine are crucial for the development of the entire wind power industry. Traditional wind turbine control is designed to output maximum power while operating safely. However, as the penetration of wind power in the power grid increases, the need for wind power to provide active power control becomes increasingly obvious. In addition, when wind power is connected to the grid, a wind turbine should have the ability to track the active power command of the grid. However, this also leads to a significant increase in the fatigue load of the wind turbine. Therefore, in the active power control of a wind turbine, the wind energy capture efficiency is no longer the core concern.

[0003] In research and engineering applications, the fatigue load is mainly reflected in the unbalanced torque borne by the main shaft and the cyclic bending moment generated by the tower under the action of wind. Therefore, the standard deviation of the main shaft torque and the standard deviation of the tower bending moment are widely regarded as typical fatigue indicators of a wind turbine. However, the calculation of the main shaft torque and the tower bending moment is complex and affected by many parameters, making it difficult to calculate accurate results in practical applications.

[0004] To achieve the active power control of a wind turbine, a controller is required to manipulate the reference values of the pitch angle and the generator torque. The main objective of controller design is to achieve fast and accurate tracking of the active power command by coordinating the pitch angle and torque control. In the design of wind turbine control strategies, the fluctuations of wind speed and power command are considered as external disturbances, and the unmodeled part and modeling uncertainties are usually considered as internal disturbances. Active Disturbance Rejection Control (ADRC) has excellent performance in resisting external and internal disturbances. However, due to the characteristics of ADRC with many parameters and two degrees of freedom, it is difficult to be widely promoted in the industry. And the active power control of wind power exacerbates the fatigue load of the unit, resulting in higher component failure rates and maintenance costs. Therefore, how to improve the active command tracking ability of a wind turbine and reduce the fatigue load of the system has become the main objective that urgently needs to be solved at present. Summary of the Invention

[0005] The purpose of the present application is to provide a coordinated optimization control method and device for improving the active power regulation ability of a wind turbine, which can improve the active command tracking ability of a wind turbine and reduce the fatigue load at the same time.

[0006] To achieve the above object, the present application provides the following solutions:

[0007] In a first aspect, the present application provides a coordinated optimization control method for improving the active power regulation ability of a wind turbine, including: constructing a first-order error active disturbance rejection controller; establishing a controller parameter optimization model with the minimum fatigue load of the unit and the best control performance as multiple objectives; the controller parameter optimization model includes the parameters of two first-order error active disturbance rejection controllers; according to the operating parameters of the wind turbine, using a bidirectional long short-term memory network to obtain the fatigue load index value; according to the fatigue load index value, the output power of the wind turbine and the rotor speed, using a snow melting optimization algorithm to solve the controller parameter optimization model to obtain the optimal parameters of the two first-order error active disturbance rejection controllers; the snow melting optimization algorithm initializes the population using the Sobol sequence, updates the entire population using the golden sine search mechanism, and mutates the optimal individual using the Lévy flight strategy; controlling the pitch angle and electromagnetic torque of the wind turbine respectively through two first-order error active disturbance rejection controllers using the optimal parameters to adjust the output power and rotor speed of the wind turbine.

[0008] Optionally, the first-order error active disturbance rejection controller includes: an extended state observer and a control law; the extended state observer takes the system error and the control signal output by the control law as inputs, and outputs a first state variable and a second state variable according to the system error and the control signal output by the control law; when the system error is the deviation between the output power of the wind turbine and the active power reference value, the control signal output by the control law is the pitch angle command; when the system error is the deviation between the rotor speed of the wind turbine and the rotor speed reference value, the control signal output by the control law is the electromagnetic torque command; the gain of the extended state observer satisfies l 1 = 2ω o , l 1 and l 2 are both the gains of the extended state observer, ω o is the bandwidth of the extended state observer; the control law outputs a control signal according to the first state variable and the second state variable, based on where u is the control signal, k 0 and b 0 are both control gains, k 0 = ω c , ω c is the bandwidth of the first-order error active disturbance rejection controller, z 1 is the first state variable, and z 2 is the second state variable.

[0009] Optionally, the controller parameter optimization model is:

[0010]

[0011] In the formula, is the parameter matrix of the first-order error active disturbance rejection controller, b 01 and b 02 are the control gains of two first-order error active disturbance rejection controllers, ω o1 and ω o2 are the bandwidths of the extended state observers of two first-order error active disturbance rejection controllers, ω c1 and ω c2 are the bandwidths of two first-order error active disturbance rejection controllers; F is the total objective function; F 1 is the control performance sub-objective function; F 2 is the fatigue load sub-objective function; h 1 is the weighting coefficient of the control performance sub-objective function; h 2 is the weighting coefficient of the fatigue load sub-objective function; T 0 is the preset initial time; T max is the maximum simulation time; t is the time variable; is the rotor speed reference value; ω r is the rotor speed of the wind turbine; P cmd is the active power reference value; P e is the output power of the wind turbine; std is the standard deviation; M t is the tower moment; T s is the main shaft torque; β ref and Δβ ref are the pitch angle command and the pitch angle command variation respectively; T g-ref and ΔT g-ref are the electromagnetic torque command and the electromagnetic torque command variation respectively; s is the time unit; Z max and Z min are the upper and lower limits of the parameter variables of the first-order error active disturbance rejection controller respectively.

[0012] Optionally, according to the operating parameters of the wind turbine, the fatigue load index value is obtained by using a bidirectional long short-term memory network, which specifically includes: determining the tower moment and the main shaft torque as the fatigue load indexes; obtaining multiple operating parameter data of the wind turbine; according to the multiple operating parameter data, performing Pearson correlation analysis on the multiple operating parameters, the tower moment and the main shaft torque to obtain the Pearson correlation coefficient between each operating parameter and the tower moment, and the Pearson correlation coefficient between each operating parameter and the main shaft torque; selecting the operating parameters with Pearson correlation coefficients greater than the preset coefficient threshold as the first input feature variables corresponding to the tower moment and the second input feature variables corresponding to the main shaft torque respectively; according to the first input feature variable values of the wind turbine, using the first bidirectional long short-term memory network to obtain the value of the tower moment; according to the second input feature variables of the wind turbine, using the second bidirectional long short-term memory network to obtain the value of the main shaft torque.

[0013] Optionally, the first input feature variables corresponding to the tower moment include: pitch angle, wind speed, mechanical torque, and tip speed ratio; the second input feature variables corresponding to the main shaft torque include: rotor speed, electromagnetic speed, mechanical torque, and electromagnetic torque.

[0014] Optionally, the specific process of the snowmelt optimization algorithm includes: initializing the population parameters; the population parameters include the initial population size, the maximum number of iterations, and the search range; using the Sobol sequence to initialize the initial positions of all individuals in the initial population; according to the initial positions of each individual, calculating the fitness value of each individual in the initial population, and selecting the initial position of the individual with the maximum fitness value as the global optimal solution; constructing an elite pool for the initial population; introducing a dual-population mechanism to divide the entire initial population into two sub-populations; entering the exploration stage, using Brownian motion to simulate the complex process of snow or water turning into steam, and updating the two sub-populations; entering the exploitation stage, using the degree-day method to simulate the process of snow melting into water; using the golden sine search mechanism to update the position of each individual in the current entire population; calculating the fitness value of each individual in the current entire population, and updating the global optimal solution; using the Lévy flight strategy with long and short jump capabilities to mutate the global optimal solution to generate a mutated global optimal solution; using the greedy strategy to determine whether the current global optimal solution is the global optimal solution before mutation or the mutated global optimal solution; updating the elite pool; determining whether the current iteration count has reached the maximum number of iterations to obtain a comparison result; if the comparison result is no, then return to the step "introducing a dual-population mechanism to divide the entire initial population into two sub-populations"; if the comparison result is yes, then output the current global optimal solution.

[0015] Optionally, the elite pool is [Z b ″ est ,Z second (k),Z third (k),Zc (k); where Z b ″ est is the global optimal solution of the population, Z second (k) and Z third (k) are the positions of the second-best and third-best individuals in the population, Z c (k) is the central position of the top 50% of individuals with fitness values, is the overall population after rearranging Z i (k) from best to worst, N 1 is half of the total population size; k = 1, 2LK, and K is the maximum number of iterations.

[0016] Optionally, the greedy strategy is: In the formula, Z b ′ est is the current global optimal solution, Z new is the mutated global optimal solution, Z best is the global optimal solution before mutation, and f is the fitness function.

[0017] Optionally, the pitch angle and electromagnetic torque of the wind turbine are respectively controlled by two first-order error active disturbance rejection controllers using optimal parameters, specifically including: inputting the deviation between the output power of the wind turbine and the active power reference value into the first first-order error active disturbance rejection controller to output a pitch angle command; the optimal parameters of the first first-order error active disturbance rejection controller include b 01 , ω o1 and ω c1 ; inputting the deviation between the rotor speed of the wind turbine and the rotor speed reference value into the second first-order error active disturbance rejection controller to output an electromagnetic torque command; the optimal parameters of the second first-order error active disturbance rejection controller include b 02 , ω o2 and ω c2 .

[0018] In a second aspect, the present application provides a coordinated optimization control device for improving the active power regulation ability of a wind turbine, including: a controller construction module, an optimization model establishment module, an index calculation module, a parameter optimization module, and a coordinated optimization module.

[0019] A controller construction module for constructing a first-order error active disturbance rejection controller; an optimization model establishment module for establishing a controller parameter optimization model with the minimum fatigue load of the unit and the best control performance as multiple objectives; the controller parameter optimization model includes the parameters of two first-order error active disturbance rejection controllers; an index calculation module for obtaining a fatigue load index value by using a bidirectional long short-term memory network according to the operating parameters of the wind turbine; a parameter optimization module for solving the controller parameter optimization model by using a snow melting optimization algorithm according to the fatigue load index value, the output power of the wind turbine and the rotor speed to obtain the optimal parameters of the two first-order error active disturbance rejection controllers; the snow melting optimization algorithm initializes the population by using a Sobol sequence, updates the entire population by using a golden sine search mechanism, and mutates the optimal individual by using a Lévy flight strategy; a coordinated optimization module for controlling the pitch angle and electromagnetic torque of the wind turbine by using two first-order error active disturbance rejection controllers with optimal parameters respectively to adjust the output power and rotor speed of the wind turbine.

[0020] According to the specific embodiments provided in the present application, the present application has the following technical effects:

[0021] The present application provides a coordinated optimization control method and device for improving the active power regulation ability of a wind turbine. When optimizing the parameters of a first-order error active disturbance rejection controller, a controller parameter optimization model with the minimum fatigue load of the unit and the best control performance as multiple objectives is established, which not only considers the tracking performance of the power command but also considers the reduction of the fatigue load of the thermal power unit. The snow melting optimization algorithm is used to quickly and accurately locate the optimal parameters of the first-order error active disturbance rejection controller. Then, the pitch angle and electromagnetic torque of the wind turbine are controlled by using two first-order error active disturbance rejection controllers with optimal parameters respectively, so as to adjust the active power and rotor speed output by the wind turbine, improve the active power command tracking ability of the wind power generation unit, and reduce the fatigue load at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0023] Figure 1 It is a schematic flowchart of a coordinated optimization control method for improving the active power regulation ability of a wind turbine provided by an embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of the structure of a wind turbine provided by another embodiment of the present application;

[0025] Figure 3Schematic diagram of the first-order error active disturbance rejection controller provided by another embodiment of the present application;

[0026] Figure 4 Schematic diagram of the active power coordinated control box of a wind turbine based on the first-order error active disturbance rejection controller provided by another embodiment of the present application;

[0027] Figure 5 Schematic diagram of the long short-term memory network structure provided by another embodiment of the present application;

[0028] Figure 6 Schematic diagram of the bidirectional long short-term memory network structure provided by another embodiment of the present application;

[0029] Figure 7 Schematic diagram of the enhanced snow melting optimization algorithm process provided by another embodiment of the present application;

[0030] Figure 8 Schematic diagram of the principle of a coordinated optimization control method for improving the active power regulation ability of a wind turbine provided by an embodiment of the present application. Detailed implementation manners

[0031] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0032] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0033] In an exemplary embodiment, as Figure 1 shown, a coordinated optimization control method for improving the active power regulation ability of a wind turbine is provided, including the following steps 101 to 105. Among them:

[0034] Step 101: Construct a first-order error active disturbance rejection controller.

[0035] Step 102: Establish a controller parameter optimization model with the minimum fatigue load of the unit and the best control performance as multiple objectives; the controller parameter optimization model includes the parameters of two first-order error active disturbance rejection controllers.

[0036] Step 103: According to the operating parameters of the wind turbine, use the bidirectional long short-term memory network to obtain the fatigue load index value.

[0037] Step 104: According to the fatigue load index value, the output power of the wind turbine, and the rotor speed, use the snow melting optimization algorithm to solve the controller parameter optimization model, and obtain the optimal parameters of the two first-order error active disturbance rejection controllers; the snow melting optimization algorithm initializes the population using the Sobol sequence, updates the entire population using the golden sine search mechanism, and mutates the optimal individual using the Lévy flight strategy.

[0038] Step 105: Control the pitch angle and electromagnetic torque of the wind turbine through two first-order error active disturbance rejection controllers using the optimal parameters to adjust the output power and rotor speed of the wind turbine.

[0039] Implementing the above Steps 101 to 105 not only improves the active power command tracking ability of the wind turbine generator set but also reduces the fatigue load of the system.

[0040] In another exemplary embodiment of the present application, in order to be compatible with standard industrial control function blocks, the two-degree-of-freedom ADRC is reconstructed into a one-degree-of-freedom error active disturbance rejection control (EADRC). In EADRC, the observation gain and control gain are bandwidth-ized, which simplifies the controller structure and parameter tuning problems. In particular, using the error as the input of the controller makes the application and deployment of EADRC in industry more convenient. Therefore, introducing EADRC into the active power control of wind turbines is a good attempt. Therefore, in order to improve the regulation performance of the active power of wind turbines, the present application designs a first-order error active disturbance rejection controller according to the wind power system model.

[0041] The traditional wind power system consists of many interrelated subsystems, mainly including an aerodynamic subsystem, a transmission subsystem, a pitch subsystem, an electrical subsystem, and a control subsystem. The structure of the wind turbine is as Figure 2 shown.

[0042] In the wind power control system, the operation of the system can be regarded as a second-order dynamic process, generally described as:

[0043]

[0044] where a 0 and a 1 are system parameters. u(t) is the input of the system, that is, the control quantity; y(t) is the system output; and are system state variables; b is the control gain; w(t) is the unknown disturbance.

[0045] EADRC is a practical one-degree-of-freedom controller, and the structure of the first-order EADRC system is as Figure 3as shown Figure 3 where r represents the set value, ε is the system error, y is the output of the system, and k 0 and b 0 are control gains, u 0 is the error feedback control quantity, u is the control signal applied to the system after adjustment, and z 1 and z 2 are the outputs of the Extended State Observer (ESO).

[0046] The extended state observer takes the system error and the control signal output by the control law as inputs, and outputs the first state variable and the second state variable according to the system error and the control signal output by the control law; when the system error is the deviation between the output power of the wind turbine and the active power reference value, the control signal output by the control law is the pitch angle command; when the system error is the deviation between the rotor speed of the wind turbine and the rotor speed reference value, the control signal output by the control law is the electromagnetic torque command.

[0047] According to the characteristics of ADRC, all dynamic characteristics different from the series integrator can be regarded as disturbances, and then estimated and compensated by the ESO. Considering that low-order controllers are easier to design and deploy in practical applications, and they can stably control high-order objects. This application designs a first-order EADRC to control the wind power system, simplifying the complexity of the controller structure and parameter adjustment. Based on this analysis, the structure of the wind power system can be rewritten as:

[0048]

[0049] Let

[0050]

[0051] Equation (2) can be rewritten as:

[0052]

[0053] where f is the total disturbance of the system.

[0054] In the framework of the EADRC method, the error is defined as a state variable:

[0055]

[0056] where h is the total disturbance of the system in the EADRC scheme,

[0057] Select the state variable as x = [x 1 x 2 = [ε h]. Then, equation (5) is written as:

[0058]

[0059] The second-order ESO is given by:

[0060]

[0061] where z 1 and z 2 are the estimated values of x 1 and h respectively; l 1 and l 2 are the observer gains.

[0062] The design of the EADRC control method is consistent with the traditional ADRC. To simplify the structure, the control law is designed as follows:

[0063]

[0064] According to the stability analysis, the observer gain and the control gain can be defined by the bandwidth as l 1 = 2ω o , k 0 = ω c . ω o represents the observer bandwidth, and ω c is the controller bandwidth.

[0065] The block diagram of the active power coordinated control of the wind turbine based on EADRC is shown in Figure 4 .

[0066] After the controller design is completed, the adjustment of its parameters is crucial for the performance of the controller. In the wind power system, the objective function of the controller parameter optimization is multiple.

[0067] In another exemplary embodiment of this application, in the active power control of wind power generation, two optimization objectives are considered. The objective F 1 is to minimize the tracking error, which is composed of the sum of the integrals of the absolute errors of the rotor speed and the active power over time. The objective F 2 is to reduce the fatigue load of the unit, which is composed of the sum of the standard deviations of the tower moment and the main shaft torque. The total objective F is composed of the weighted sum of F 1 and F 2 . In addition, the constraints on the pitch angle, the generator torque, and the decision variables are also considered. Then the controller parameter optimization model is:

[0068]

[0069] where is the parameter matrix of the first-order error auto-disturbance rejection controller, b 01 and b 02 are the control gains of two first-order error active disturbance rejection controllers, ω o1 and ω o2 are the bandwidths of the extended state observers of two first-order error active disturbance rejection controllers, ω c1 and ω c2 are the bandwidths of two first-order error active disturbance rejection controllers; F is the total objective function; F 1 is the control performance sub-objective function; F 2 is the fatigue load sub-objective function; h 1 is the weighting coefficient of the control performance sub-objective function; h 2 is the weighting coefficient of the fatigue load sub-objective function; T 0 is the preset initial time; T max is the maximum simulation time; t is the time variable; is the rotor speed reference value; ω r is the rotor speed of the wind turbine; P cmd is the active power reference value; P e is the output power of the wind turbine; std is the standard deviation; M t is the tower bending moment; T s is the main shaft torque; β ref and Δβ ref are the pitch angle command and the pitch angle command change amount respectively; T g-ref and ΔT g-ref are the electromagnetic torque command and the electromagnetic torque command change amount respectively; s is the time unit, second; Z max and Z min are the upper and lower limits of the parameter variables of the first-order error active disturbance rejection controller respectively. is the parameter matrix of the EADRC scheme, represented by the i-th individual in the k-th iteration of the enhanced snowmelt optimization algorithm.

[0070] In another exemplary embodiment of the present application, the calculation of the fatigue load of the unit is relatively complex and has strong nonlinearity. Machine learning represented by bidirectional long short-term memory (BiLSTM) performs well in data prediction and data fitting. Therefore, the intelligent modeling of the main shaft torque and the tower bending moment can be realized by using BiLSTM. The present application designs a fatigue load estimation model based on BiLSTM to realize the real-time and accurate calculation of the fatigue load change of the wind power system during the optimization process.

[0071] The present application takes the periodic changes of the tower bending moment M t and the main shaft torque T s as typical indicators of the fatigue load.

[0072] To facilitate the calculation of M of the wind turbine by using the BiLSTM networkt and T s For the 10,000 groups of data sets constructed and selected in this application, it includes pitch angle β, wind speed V, rotor speed ωr, electromagnetic speed ω g , mechanical torque T r , electromagnetic torque T g , tip speed ratio λ, these 7 characteristic parameters. Although most characteristic parameters have a direct impact on fatigue loads, in order to simplify the complexity of the model.

[0073] This application first performed Pearson correlation analysis on the 7 characteristic parameters with M t and T s , and selected the four characteristic parameters with the highest correlation as the input of the BiLSTM network. The results of the Pearson correlation coefficient are shown in Table 1. From the data in Table 1, β, V, T r and λ were selected as the input characteristic variables of the M t model, ωr, ω g , T r and T g were selected as the input characteristic variables of the T s model.

[0074] Table 1 Pearson correlation analysis

[0075]

[0076] Long short-term memory (LSTM) is a special type of recurrent neural network, specifically designed to solve the problems of vanishing gradients and exploding gradients faced by traditional recurrent neural networks when dealing with long sequence data. Its structure is mainly composed of an input gate, a forget gate, and an output gate, as Figure 5 shown. BiLSTM is an enhanced network that combines forward LSTM and backward LSTM. It has independent hidden layers in both the forward and backward directions, and each hidden layer captures features and information in both the forward and backward directions of the data set. The structure of BiLSTM is as Figure 6 shown.

[0077] During the operation of the wind turbine, we use the BiLSTM network to simulate M t and T s . BiLSTM can accurately capture the complex dependencies between each characteristic parameter and M t and T s , so as to quickly estimate the trends of M t and T s . The model equations of M t and T s are as follows:

[0078]

[0079] Then, the above step 103 can be replaced by the following steps 201 to 206:

[0080] Step 201: Determine the tower moment and the main shaft torque as fatigue load indicators.

[0081] Step 202: Obtain multiple operating parameter data of the wind turbine.

[0082] Step 203: According to the multiple operating parameter data, perform Pearson correlation analysis on the multiple operating parameters, the tower moment, and the main shaft torque to obtain the Pearson correlation coefficient between each operating parameter and the tower moment, and the Pearson correlation coefficient between each operating parameter and the main shaft torque.

[0083] Step 204: Select the operating parameters with Pearson correlation coefficients greater than the preset coefficient threshold as the first input feature variables corresponding to the tower moment and the second input feature variables corresponding to the main shaft torque, respectively.

[0084] Step 205: According to the values of the first input feature variables of the wind turbine, use the first bidirectional long short-term memory network to obtain the value of the tower moment.

[0085] Step 206: According to the second input feature variables of the wind turbine, use the second bidirectional long short-term memory network to obtain the value of the main shaft torque.

[0086] In another exemplary embodiment of the present application, in power tracking control, the optimization of the controller parameters should consider both the tracking performance of the power command and the reduction of the fatigue load of the unit, which is a typical multi-objective problem. To quickly and accurately locate the optimal controller parameters, an enhanced snow melting optimization algorithm is used to solve the multi-objective function.

[0087] The snow melting optimization algorithm is a newly proposed novel bionic optimization algorithm. It draws inspiration from nature and solves complex optimization problems by simulating the natural changes of snow. For the multi-objective optimization problem of the active power control system of wind power generation,

[0088] The present application will propose an enhanced snow melting optimization algorithm to obtain the parameters of the EADRC to optimize the overall performance of the active power control of the wind turbine. The enhanced snow melting optimization algorithm includes three improvements, namely, Sobol sequence initialization of the population, golden sine search mechanism, and Lévy flight strategy. As Figure 7 shown, the detailed process of the enhanced snow melting optimization algorithm is described as follows:

[0089] Step 1: Initialize the population size N, the maximum number of iterations K, the search range [Z max , Z min , and two sub-populations Na = N b = N / 2。

[0090] Step 2: Initialize the population with the Sobol sequence to obtain a set of initial solutions evenly distributed between [0, 1].

[0091] Z i (0) = Z min + G n (Z max - Z min )(i = 1, 2, L N) (11)

[0092] In the formula, Z i (0) is the initial position of the i-th individual; G n is the Sobol random sequence distributed on [0, 1].

[0093] Step 3: Calculate the fitness value of each individual in the initial population and record the global optimal solution Z best . Then construct an elite pool through Equation (10) to prevent excellent individuals from being lost in the next generation.

[0094]

[0095] In the formula, Z elite (k) is the individual randomly selected from the elite pool; Z b ″ est represents the global optimal position; Z second (k) and Z third (k) represent the second-best and third-best individuals in the current population; Z c (k) represents the central position of the top 50% of individuals with fitness values; is the overall population after rearranging Z i (k) from best to worst; N 1 is half of the total population N, N 1 = N / 2.

[0096] Step 4: To balance the exploration and exploitation phases, a dual-population mechanism is introduced, where the entire population N is divided into two sub-populations N a and N b .

[0097] Step 5: Enter the exploration phase, and Brownian motion is used to simulate the complex process of snow or water turning into steam. The formula is as follows:

[0098]

[0099] In the formula, f(·) is the fitness function. Z is the population individual.

[0100] Step 6: Update the Na and Nb subgroups.

[0101] N a = N a + 1, N b = N b - 1, if N a < N (15)

[0102] Step 7: Enter the development stage. Use the degree-day method to simulate the process of snow melting into water. The focus of this stage is to develop high-quality solutions around the optimal individual positions. The single-position update formula is as follows:

[0103]

[0104] In the formula, r 2 represents a random number between [0, 1]; M(k) represents the snow ablation rate; BM i (k) is a set of random numbers symbolizing Brownian motion.

[0105] M(k) is described by the degree-day method, and the expression is as follows:

[0106]

[0107] In the formula, DDF(k) represents the degree-day factor, ranging from 0.35 to 0.6; T(k) represents the average daily temperature.

[0108] Step 8: Use the golden sine search mechanism to update the entire population to prevent the population from prematurely converging near the optimal value.

[0109] Z i (k + 1)= Z i (k)× sinx 1 |- x 2 × sinx 1 × g 1 × Z best - g 2 × Z i (k)(i = 1, 2LN; k = 1, 2LK)(18)

[0110] In the formula, x 1 represents a random number between [0, 2π]; x 2 represents a random number between [0, π]; g 1 and g 2 are the golden ratio coefficients.

[0111] Step 9: After updating the population positions, calculate the fitness of the entire population and update the global optimal position Z bestConsidering that the original algorithm will widely converge to the vicinity of the optimal value in the later iteration, resulting in a lack of randomness. In this application, the Lévy flight strategy with long and short jump capabilities is used to mutate the optimal individual Z best , so as to fully increase the diversity of the population and expand the search range. The Lévy escape strategy formula is as follows:

[0112]

[0113] where Z new is the position of the optimal individual after Lévy flight mutation; α is the control step size parameter, which is taken as 0.01 here; levy(δ) represents the path following the Lévy distribution.

[0114] Step 10: Compare the fitness values before and after perturbation, and use the greedy strategy to determine whether to update the global optimal value.

[0115]

[0116] where, Z b ′ est is the current global optimal solution, Z new is the global optimal solution after mutation, Z best is the global optimal solution before mutation, and f is the fitness function.

[0117] Step 11: Update the elite pool and determine whether the current iteration count has reached the maximum iteration number. If so, Z best is considered as the global optimal solution of the entire optimization process. Otherwise, return to Step 4.

[0118] Under the solution of the enhanced snow melting optimization algorithm, it can quickly and accurately obtain the optimal parameters of the EADRC controller under the constraint conditions of minimizing the fatigue load of the unit and optimizing the control performance. Reasonable parameter selection can give full play to the performance of the EADRC controller, thereby improving the active command tracking ability of the wind power system.

[0119] In another exemplary embodiment of this application, in the above step 105, the pitch angle and electromagnetic torque of the wind turbine are respectively controlled by two first-order error active disturbance rejection controllers using optimal parameters, specifically including:

[0120] Input the deviation between the output power of the wind turbine and the active power reference value into the first first-order error active disturbance rejection controller to output the pitch angle command; the optimal parameters of the first first-order error active disturbance rejection controller include b 01 , ω o1 and ω c1The deviation between the rotor speed of the wind turbine and the rotor speed reference value is input into the second first-order error active disturbance rejection controller, and an electromagnetic torque command is output; the optimal parameters of the second first-order error active disturbance rejection controller include b 02 , ω o2 and ω c2 .

[0121] The overall principle of the method of this application is as Figure 8 shown. The method of this application is roughly divided into the following three parts:

[0122] 1) A coordinated active power control strategy for wind turbines based on EADRC is designed to achieve stable tracking of active power commands.

[0123] 2) The problem of controller parameter optimization is realized by using an enhanced snow melting optimizer to balance the tracking performance of active power and the mitigation of fatigue loads of the unit.

[0124] 3) The establishment of a wind turbine fatigue estimation model is realized through a BiLSTM network.

[0125] Taking a 5MW wind turbine as an example for simulation, the quantization indexes F, F1 and F are respectively counted under three different scenarios of power command tracking, pitch angle actuator failure and model mismatch 2 . The statistical results show that the strategy for improving the active power regulation ability of wind turbines proposed in this application shows excellent power tracking ability in this example, and the fatigue load is also mitigated.

[0126] Based on the same inventive concept, the embodiment of this application also provides a coordinated optimization control device for improving the active power regulation ability of a wind turbine to implement the coordinated optimization control method for improving the active power regulation ability of a wind turbine involved above. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the coordinated optimization control device for improving the active power regulation ability of a wind turbine provided below can refer to the limitations on the coordinated optimization control method for improving the active power regulation ability of a wind turbine in the above text, and will not be repeated here.

[0127] In an exemplary embodiment, a coordinated optimization control device for improving the active power regulation ability of a wind turbine is provided, including: a controller construction module, an optimization model establishment module, an index calculation module, a parameter optimization module, and a coordinated optimization module.

[0128] A controller construction module for constructing a first-order error active disturbance rejection controller; an optimization model establishment module for establishing a controller parameter optimization model with the minimum fatigue load of the unit and the best control performance as multiple objectives; the controller parameter optimization model includes the parameters of two first-order error active disturbance rejection controllers; an index calculation module for obtaining a fatigue load index value by using a bidirectional long short-term memory network according to the operating parameters of the wind turbine; a parameter optimization module for solving the controller parameter optimization model by using a snow melting optimization algorithm according to the fatigue load index value, the output power and the rotor speed of the wind turbine to obtain the optimal parameters of the two first-order error active disturbance rejection controllers; the snow melting optimization algorithm initializes the population by using a Sobol sequence, updates the entire population by using a golden sine search mechanism, and mutates the optimal individual by using a Lévy flight strategy; a coordinated optimization module for respectively controlling the pitch angle and the electromagnetic torque of the wind turbine by using two first-order error active disturbance rejection controllers with optimal parameters to adjust the output power and the rotor speed of the wind turbine.

[0129] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0130] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set, characterized in that: include: Construct a first-order error active disturbance rejection controller; Establishing a controller parameter optimization model with multi-objectives of minimizing fatigue load of the unit and optimizing control performance; the controller parameter optimization model includes parameters of two first-order error auto-disturbance rejection controllers; According to the operating parameters of the wind turbine, the fatigue load index value is obtained using a bidirectional long short-term memory network; According to the fatigue load index value, the output power and rotor speed of the wind turbine, the controller parameter optimization model is solved by using a snowmelt optimization algorithm to obtain the optimal parameters of two first-order error anti-disturbance controllers; the snowmelt optimization algorithm uses a Sobol sequence to initialize the population, uses a golden sine search mechanism to update the entire population, and uses a Levy flight strategy to mutate the optimal individual; The pitch angle and electromagnetic torque of the wind turbine are controlled respectively by two first-order error anti-disturbance controllers using optimal parameters to adjust the output power and rotor speed of the wind turbine.

2. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 1, characterized in that: The first-order error active disturbance rejection controller comprises: an extended state observer and a control law; The extended state observer takes the system error and the control signal output by the control law as input, and outputs the first state variable and the second state variable according to the system error and the control signal output by the control law; when the system error is the deviation between the output power of the wind turbine and the active power reference value, the control signal output by the control law is a pitch angle instruction; when the system error is the deviation between the rotor speed of the wind turbine and the rotor speed reference value, the control signal output by the control law is an electromagnetic torque instruction; The gain of the extended state observer satisfies l1=2ω o , l1 and l2 are the gains of the extended state observer, ω o is the bandwidth of the extended state observer; The control law is based on the first state variable and the second state variable, according to Output control signal; where u is the control signal, k0 and b0 are both control gains, k0 = ω c ,ω c is the bandwidth of the first-order error ADRC, z1 is the first state variable, and z2 is the second state variable.

3. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 1, characterized in that: The controller parameter optimization model is: In the formula, is the parameter matrix of the first-order error ADRC, b 01 and b 02 are the control gains of the two first-order error ADRCs, ω o1 and ω o2 is the extended state observer bandwidth of the two first-order error ADRCs, ω c1 and ω c2 is the bandwidth of the two first-order error ADRCs; F is the total objective function; F1 is the control performance sub-objective function; F2 is the fatigue load sub-objective function; h1 is the weighting coefficient of the control performance sub-objective function; h2 is the weighting coefficient of the fatigue load sub-objective function; T0 is the preset initial time; T max is the maximum simulation time; t is the time variable; is the rotor speed reference value; ω r is the rotor speed of the wind turbine; P cmd is the active power reference value; P e is the output power of the wind turbine; std is the standard deviation; M t is the tower bending moment; T s is the main shaft torque; β ref and Δβ ref are the pitch angle command and the pitch angle command variation respectively; T g-ref and ΔT g-ref They are electromagnetic torque command and electromagnetic torque command variation respectively; s is the time unit; Z max and Z min are the upper and lower limits of the parameter variables of the first-order error active disturbance rejection controller.

4. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 1, characterized in that: According to the operating parameters of the wind turbine, the fatigue load index value is obtained using the bidirectional long short-term memory network, including: Determine tower bending moment and main shaft torque as fatigue load indicators; Obtain multiple operating parameter data of wind turbines; According to the data of multiple operating parameters, a Pearson correlation analysis is performed on the multiple operating parameters, the tower bending moment and the main shaft torque to obtain the Pearson correlation coefficient between each operating parameter and the tower bending moment, and the Pearson correlation coefficient between each operating parameter and the main shaft torque; Selecting operating parameters whose Pearson correlation coefficient is greater than a preset coefficient threshold as the first input characteristic variable corresponding to the tower bending moment and the second input characteristic variable corresponding to the main shaft torque; According to the first input characteristic variable value of the wind turbine generator set, a value of the tower bending moment is obtained by using a first bidirectional long short-term memory network; According to the second input characteristic variable of the wind turbine generator set, a value of the main shaft torque is obtained by utilizing a second bidirectional long short-term memory network.

5. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 4 is characterized in that: The first input characteristic variables corresponding to the tower bending moment include: pitch angle, wind speed, mechanical torque and tip speed ratio; The second input characteristic variable corresponding to the main shaft torque includes: rotor speed, electromagnetic speed, mechanical torque and electromagnetic torque.

6. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 1, characterized in that: The specific process of the snowmelt optimization algorithm includes: Initializing population parameters; the population parameters include initial population size, maximum number of iterations and search range; Initializing the initial positions of all individuals in the initial population using the Sobol sequence; According to the initial position of each individual, the fitness value of each individual in the initial population is calculated, and the initial position of the individual with the largest fitness value is selected as the global optimal solution; Construct an elite pool of the initial population; Introduce a dual population mechanism to divide the entire initial population into two subpopulations; Entering the exploration phase, Brownian motion is used to simulate the complex process of snow or water turning into steam and update the two subpopulations; Entering the development phase, the degree-day method is used to simulate the process of snow melting into water; Use the golden sine search mechanism to update the position of each individual in the current population; Calculate the fitness value of each individual in the current population and update the global optimal solution; The global optimal solution is mutated using the Levy flight strategy with long and short jump capabilities to generate the mutated global optimal solution; Use the greedy strategy to determine whether the current global optimal solution is the global optimal solution before mutation or the global optimal solution after mutation; Updated elite pool; Determine whether the current iteration count has reached the maximum number of iterations and obtain a comparison result; If the comparison result is negative, return to step "introducing a dual population mechanism to divide the entire initial population into two subpopulations"; If the comparison result is yes, the current global optimal solution is output.

7. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 6, characterized in that: The elite pool is [Z b ″ est ,Z second (k),Z third (k),Z c (k)]; Among them, Z b ″ est is the global optimal solution of the population, Z second (k) and Z third (k) is the second and third best individual position in the population, Z c (k) is the center position of the top 50% individuals with fitness values, Rearrange Z from best to worst i (k) after, N1 is half of the total population size; k = 1,2LK, K is the maximum number of iterations.

8. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 6, characterized in that: The greedy strategy is: In the formula, Z b ' est is the current global optimal solution, Z new is the global optimal solution after mutation, Z best is the global optimal solution before mutation, and f is the fitness function.

9. The coordinated optimization control method for improving the active power regulation capability of a wind turbine generator set according to claim 3, characterized in that: The pitch angle and electromagnetic torque of the wind turbine are controlled respectively by two first-order error anti-disturbance controllers using optimal parameters, including: The deviation between the output power of the wind turbine and the active power reference value is input into the first first-order error anti-disturbance controller, and the pitch angle command is output; the optimal parameters of the first first-order error anti-disturbance controller include b 01 ,ω o1 and ω c1 ; The deviation between the rotor speed of the wind turbine and the rotor speed reference value is input into the second first-order error anti-disturbance controller to output the electromagnetic torque command; the optimal parameters of the second first-order error anti-disturbance controller include b 02 ,ω o2 and ω c2 .

10. A coordinated optimization control device for improving the active power regulation capability of a wind turbine generator set, characterized in that: include: Controller building block for building first-order error ADRC controllers; An optimization model building module is used to build a controller parameter optimization model with multiple objectives of minimizing fatigue load of the unit and optimizing control performance; the controller parameter optimization model includes parameters of two first-order error auto-disturbance rejection controllers; An index calculation module is used to obtain fatigue load index values ​​using a bidirectional long short-term memory network according to the operating parameters of the wind turbine generator set; A parameter optimization module is used to solve the controller parameter optimization model using a snowmelt optimization algorithm according to the fatigue load index value, the output power and the rotor speed of the wind turbine, and obtain the optimal parameters of two first-order error anti-disturbance controllers; the snowmelt optimization algorithm uses a Sobol sequence to initialize the population, uses a golden sine search mechanism to update the entire population, and uses a Levy flight strategy to mutate the optimal individual; The coordinated optimization module is used to control the pitch angle and electromagnetic torque of the wind turbine respectively through two first-order error anti-disturbance controllers using optimal parameters to adjust the output power and rotor speed of the wind turbine.

Images