A Dual-Mode Dynamic Active Wake Control Method, System and Medium for Wind Farms

Abstract

The present invention discloses a dual-mode dynamic active wake control method, system and medium for a wind farm. The present invention includes: if the current control mode is the power enhancement mode, obtaining an optimal control instruction by solving with the optimization objectives of maximum power output, power balance and fatigue load suppression; otherwise, obtaining an optimal control instruction by solving with the optimization objective of maximum power output; sending the optimal control instruction to the virtual wind farm and obtaining a delayed control instruction; if it is the power enhancement mode, directly sending it to the real wind farm; if it is the power tracking mode, solving the final optimal control instruction with the optimization objectives of power reference tracking, control reference tracking and fatigue load suppression and sending it to the real wind farm. The purpose of the present invention is to weaken the influence of wake effect in the wind farm through dynamic active wake control, improve power generation and maintain fatigue load balance, so as to bring higher economic benefits to the operator.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation, and particularly relates to a dual-mode dynamic active wake control method, system and medium for a wind farm. Background Art

[0002] With the continuous growth of the world's demand for renewable energy, the scale of wind farms is increasing. To accommodate a certain number of wind turbines within a limited area, the wake effect has gradually become more and more significant and cannot be ignored. The wake effect refers to the interference effect of upstream wind turbines on the downstream wind speed field and turbulence characteristics, usually manifested as power loss and increased loads of downstream turbines. The accumulation of this effect not only significantly reduces the overall power generation efficiency of the wind farm but also exacerbates the fatigue loss of equipment, posing new technical challenges to the layout optimization and control strategies of wind farms. Therefore, the wake effect has become a key research topic in the field of optimal design and intelligent control of large-scale wind farms. The operation modes of wind farms are usually divided into two types: one is the "maximum power output" mode, aiming to output as much electrical energy as possible; the other is the "power reference tracking" mode, with the main goal of meeting the power demand given by the power grid. In the "maximum power output" mode, traditional control methods set each wind turbine at its respective maximum power point. However, due to the existence of the wake effect, the preferential absorption of wind energy by upstream turbines significantly reduces the inflow wind speed of downstream turbines, resulting in a decrease in the power output of downstream turbines and power loss. In addition, the wake effect causes an uneven distribution of fatigue loads within the wind farm, increasing the fatigue load pressure on downstream turbines and being unfavorable to the long-term stable operation of equipment. In the "power reference tracking" mode, traditional control methods calculate the power reference value of each turbine according to the proportional distribution strategy based on the current available power of each turbine to achieve the overall power tracking of the wind farm. However, due to the uncontrollability of wind speed changes, each turbine is affected by wind speed fluctuations during actual operation, resulting in severe fluctuations in fatigue loads, which not only affect the equipment life but also may cause the wind farm to be difficult to meet the power demand of the power grid when the wind energy is insufficient. In this case, how to achieve maximum power, balanced fatigue loads, and tracking of the power demand of the power grid under different operation modes has become an important topic in the current research on wind farm control strategies. Therefore, in the current field of wind farm operation optimization control, there is an urgent need for a wind farm dynamic active wake control method applicable to dynamic inflow wind speeds and considering the dynamic characteristics of wind turbines, which can improve the overall power generation of the wind farm and suppress the fatigue loads of turbines under different working modes of the wind farm, thereby providing a more applicable method for wind farm operation optimization control. Summary of the Invention

[0003] Technical problems to be solved by the present invention: In view of the above problems of the prior art, a dual-mode dynamic active wake control method, system and medium for a wind farm are provided. The present invention aims to improve the power generation of the wind farm under different working modes, balance the fatigue load distribution, and weaken the influence of the wake effect in the wind farm through dynamic active wake control, so as to improve power generation and maintain the balance of fatigue loads, bringing higher economic benefits to the operator.

[0004] To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0005] A dual-mode dynamic active wake control method for a wind farm, comprising the following steps:

[0006] S1. For the jointly deployed virtual wind farm and real wind farm, the virtual wind farm controller determines the current control mode of the virtual wind farm. If it is the power enhancement mode, the optimal control instruction is obtained by solving the wind farm state space model with the optimization objectives of maximum power output, power balance and fatigue load suppression; if it is the power tracking mode, the optimal control instruction is obtained by solving the wind farm state space model with the optimization objective of maximum power output; the optimal control instruction is sent to the virtual wind farm, and after being delayed according to the wake delay time, the obtained delayed control instruction is sent to the real wind farm controller;

[0007] S2. The real wind farm controller determines the current control mode of the real wind farm. If it is the power enhancement mode, the delayed control instruction is directly output and sent to the real wind farm; if it is the power tracking mode, the final optimal control instruction is obtained by solving the wind farm state space model with the optimization objectives of power reference tracking, control reference tracking and fatigue load suppression, and is sent to the real wind farm.

[0008] Optionally, when obtaining the optimal control instruction by solving the wind farm state space model with the optimization objectives of maximum power output, power balance and fatigue load suppression in step S1, the function expression of the adopted optimization objective is:

[0009] ,

[0010] In the above formula, is the optimization objective of the virtual wind farm in the power enhancement mode, is the optimization objective of maximum power output, is the optimization objective of power balance, is the optimization objective of wind farm fatigue load suppression, is the optimization objective of suppressing the fluctuation of the control quantity, and there are:

[0011] ,

[0012] ,

[0013] ,

[0014] ,

[0015] wherein, is the rated power of the wind farm, is the total number of wind turbines in the wind farm, is the initial power value of the j th wind turbine, is the power increment of the j th wind turbine, is a constant set manually, and are respectively the front - back and left - right displacement increments of the j th wind turbine tower, , , , and are respectively the weights of each optimization objective, and are respectively the voltage increments at the k - th and (k - 1)-th moments, where k is the current optimization time.

[0016] Optionally, when solving the wind farm state - space model to obtain the optimal control command with the maximum power output as the optimization objective in step S1, the functional expression of the optimization objective adopted is:

[0017] ,

[0018] wherein, is the optimization problem of the virtual wind farm in the power reference tracking mode, is the optimization objective of the maximum power output, is the optimization objective of suppressing the fluctuation of the control quantity, and there is:

[0019] ,

[0020] ,

[0021] wherein, is the rated power of the wind farm, is the total number of wind turbines in the wind farm, is the initial power value of the j th wind turbine, is the power increment of the j th wind turbine, and are respectively the weights of each optimization objective, and The voltage increments at times k and k - 1 respectively, where k is the current optimization time.

[0022] Optionally, the functional expression for calculating the delay time for the delay processing according to the wake delay time in step S1 is:

[0023] ,

[0024] In the above formula, is the delay time of the j th wind turbine in the real wind farm, is the geographical distance along the wind direction between the j th wind turbine and the first wind turbine in the real wind farm, is the average inflow wind speed.

[0025] Optionally, when solving the wind farm state - space model with power reference tracking, control reference tracking, and fatigue load suppression as the optimization objectives to obtain the final optimal control command and send it to the real wind farm in step S2, the functional expression of the optimization objective adopted is:

[0026] ,

[0027] Among them, is the optimization objective of the real wind farm in the power reference tracking mode, is the optimization objective of the wind farm power reference tracking, is the optimization objective of the wind farm control reference tracking, is the optimization objective of the wind farm fatigue load suppression, is the optimization objective of suppressing the fluctuation of the control quantity, and there are:

[0028] ,

[0029] ,

[0030] ,

[0031] ,

[0032] Among them, is the power reference command of the wind farm, is the total number of wind turbines in the wind farm, is the j th wind turbine's initial power value, is the j th wind turbine's power increment, is the j th wind turbine's generator electromagnetic torque increment, is the jThe initial value of the generator electromagnetic torque of the is the reference value of the generator electromagnetic torque of the j th real wind turbine, is the j th increment of the blade pitch angle of the wind turbine, is the j th initial value of the blade pitch angle of the wind turbine, is the j th reference value of the blade pitch angle of the real wind turbine, and are respectively the front - back and left - right displacement increments of the j th wind turbine tower, and are respectively the voltage increments at times k and k - 1, where k is the current optimization time, 、 、 、 、 and are respectively the weights of each optimization objective.

[0033] Optionally, the functional expression of the wind farm state - space model is:

[0034] ,

[0035] ,

[0036] In the above formula, is the system matrix of the wind farm 's first - order derivative, 、 、 、 、 、 and are all coefficient matrices of the wind farm, is the output matrix of the wind farm, is the identity matrix, is the control matrix, and there are , , are respectively the state matrix, output matrix and control matrix of the wind farm, ~ are respectively the state matrices of the 1st to nth wind turbines, ~ are respectively the output matrices of the 1st to nth wind turbines, ~ are respectively the control matrices of the 1st to nth wind turbines. The state matrix 、output matrix and the control matrix The expression of

[0037] ,

[0038] ,

[0039] ,

[0040] wherein, are respectively the increment of the generator speed of the i-th wind turbine the increment of the generator electromagnetic torque the increment of the blade pitch angle the increment of the front-back moving speed of the tower the increment of the front-back displacement of the tower the increment of the left-right moving speed of the tower the increment of the left-right displacement of the tower the increment of are respectively the increment of the generator electromagnetic torque reference value of the i-th wind turbine and the increment of the blade pitch angle reference value of is the initial value of the electromagnetic power of the i-th wind turbine.

[0041] Optionally, before step S1, it also includes the step of constructing a wind farm state space model:

[0042] S101, construct a non-linear model of a wind turbine unit for a two-degree-of-freedom control method shown by the following formula:

[0043] ,

[0044] In the above formula, is the first derivative of the generator speed , is the gearbox ratio is the equivalent inertia of the impeller and the generator is the impeller mechanical torque is the generator electromagnetic torque is the blade pitch angle the first derivative of is the pitch angle reference value is the time constant of the pitch mechanism is the first derivative of the generator electromagnetic torque , is the generator electromagnetic torque reference value is the time constant of the generator , is the air density is the impeller radius is the wind energy utilization coefficient of the wind turbine, is the tip speed ratio, is the incoming flow wind speed, is the impeller rotational speed;

[0045] S102. Construct a linearized model of the wind turbine according to the non - linear model of the wind turbine:

[0046] ,

[0047] In the above formula, Δ represents the increment, is the initial value of the impeller mechanical torque, T e0 is the initial value of the generator electromagnetic torque, and the increment of the impeller mechanical torque is:

[0048] ,

[0049] In the above formula, is the increment of the incoming flow wind speed;

[0050] S103. Set the fatigue load index of the wind turbine as the front - back and left - right displacements of the tower:

[0051] , ,

[0052] Among them, and are the front - back and left - right displacements of the tower respectively, and are the front - back and left - right moving speeds of the tower respectively, , are the second - order and first - order derivatives of the front - back displacement of the tower respectively, , are the second - order and first - order derivatives of the left - right displacement of the tower respectively, , , are the mass, damping and stiffness of the tower respectively, is the tower thrust, is the force on the nacelle, and there is: , , where is the air density, is the impeller radius, is the thrust coefficient, is the tip speed ratio, is the blade pitch angle, is the incoming flow wind speed, is the tower height, is the gearbox ratio, is the low - speed shaft torque of the wind turbine, and there is:

[0053] ,

[0054] Among them, and are the inertia of the impeller and the generator respectively;

[0055] S104. Construct a linearized form of the fatigue load index shown in the following formula:

[0056] ,

[0057] In the above formula, Δ represents an increment, and there are:

[0058] ,

[0059] ;

[0060] S105. Include the i th wind turbine state space model of the fatigue load index shown in the following formula:

[0061] ,

[0062] ,

[0063] Among them, is the state matrix of the is the first derivative of , is the control matrix of the is the first derivative of , is the increment of the inflow wind speed of the wind turbine i , is the output matrix of the is the first derivative of , and and and and are the coefficient matrices of the

[0064] ,

[0065] ,

[0066] ,

[0067] ,

[0068] , ,

[0069] , ,

[0070] Among them, are the increments of the generator speed, the increment of the generator electromagnetic torque, the increment of the blade pitch angle, the increment of the front - rear moving speed of the tower, the increment of the front - rear displacement of the tower, the increment of the left - right moving speed of the tower, the increment of the left - right displacement of the tower, the increment, are the increment of the reference value of the generator electromagnetic torque and the increment of the reference value of the blade pitch angle of the i - th wind turbine, is the increment of the electromagnetic power of the i - th wind turbine, is the impeller mechanical torque of the i - th wind turbine, is the initial value of the impeller mechanical torque of the i - th wind turbine, is the initial value of the generator electromagnetic torque of the i - th wind turbine, is the generator speed of the i - th wind turbine, is the initial value of the generator speed of the i - th wind turbine, is the blade pitch angle of the i - th wind turbine, is the tower thrust of the i - th wind turbine, is the incoming flow wind speed of the i - th wind turbine, μ is the generator efficiency;

[0071] S106. Construct the linearized form of the wind farm wake model shown by the following formula:

[0072] ,

[0073] ,

[0074] Among them, is the change in the wake wind speed caused by the i -th wind turbine on the n -th wind turbine, is the i -th wind turbine's wake wind speed on the n -th wind turbine, is the change in the thrust coefficient of the i -th wind turbine, is the i -th wind turbine's incoming flow wind speed change, and there is:

[0075] ,

[0076] ,

[0077] wherein, is the wake expansion coefficient of the Jensen wake model, is the distance along the wind direction between the i-th wind turbine and the n th virtual wind turbine, is the impeller radius, is the wake coverage area, is the impeller swept area;

[0078] S107. Determine the incremental form of the predicted values of the incoming wind speeds of the 1st to nth wind turbines according to the linearized form of the wind farm wake model:

[0079] ,

[0080] wherein, is the incremental matrix of the predicted wind speed values, is the state matrix of the wind farm, and are coefficient matrices, and there are:

[0081] ,

[0082] ,

[0083] Coefficient matrix M and N Any element and The expression of is:

[0084] , ,

[0085] Based on the coefficient matrix N is a lower triangular matrix and all diagonal elements are 0, simplify the incremental form of the predicted values of the incoming wind speeds of the 1st to nth wind turbines to:

[0086] ,

[0087] wherein, is the identity matrix;

[0088] S108. Construct the wind farm state space model:

[0089] ,

[0090] ,

[0091] Among them, , , are the state matrix, output matrix, and control matrix of the wind farm, respectively, , , , , are the coefficient matrices of the wind farm state space model, respectively, ~ are the state matrices of the 1st to nth wind turbines, respectively, ~ are the output matrices of the 1st to nth wind turbines, respectively, ~ are the control matrices of the 1st to nth wind turbines, respectively, ~ are the coefficient matrices of the 1st to nth wind turbines, respectively , ~ are the coefficient matrices of the 1st to nth wind turbines, respectively , ~ are the coefficient matrices of the 1st to nth wind turbines, respectively , ~ are the coefficient matrices of the 1st to nth wind turbines, respectively , ~ are the coefficient matrices of the 1st to nth wind turbines, respectively ; Substitute the incremental form of the predicted values of the inflow wind speeds of the 1st to nth wind turbines to obtain the simplified form of the wind farm state space model:

[0092] ,

[0093] ,

[0094] to be used to solve the wind farm state space model in combination with the specified objective to obtain the optimal control instruction.

[0095] In addition, the present invention also provides a wind farm dual-mode dynamic active wake control system, including a microprocessor and a memory connected to each other, and the microprocessor is programmed or configured to execute the wind farm dual-mode dynamic active wake control method.

[0096] In addition, the present invention also provides a computer-readable storage medium, in which a computer program or instruction is stored, and the computer program or instruction is programmed or configured to execute the wind farm dual-mode dynamic active wake control method through a processor.

[0097] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the dual-mode dynamic active wake control method for a wind farm through a processor.

[0098] Compared with the prior art, the present invention mainly has the following advantages:

[0099] 1. The present invention provides a dynamic active wake control method applicable to dynamic inflow wind speeds and considering the dynamic characteristics of wind turbines, which solves the problem of wake time delay in the optimal operation control of a wind farm considering wake effects.

[0100] 2. In the "power enhancement" mode of the wind farm, compared with the traditional method, the present invention can improve the overall power generation of the wind farm, while balancing the fatigue load distribution of the wind farm and suppressing the fluctuation of the fatigue load of the wind turbines.

[0101] 3. In the "power tracking" mode of the wind farm, compared with the traditional method, the present invention can suppress the load fluctuation of the wind turbines under the condition of abundant wind energy, and improve the power generation of the wind farm under the condition of scarce wind energy to achieve better power tracking performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1 It is a schematic diagram of the basic process of the method according to the embodiment of the present invention.

[0103] Figure 2 It is a schematic diagram of the principle of the method according to the embodiment of the present invention.

[0104] Figure 3 It is a structural diagram of the simulation system of the simulation case in the embodiment of the present invention.

[0105] Figure 4 It is a simulation result diagram of the total power generation of the wind farm in the "power enhancement" mode of the present invention.

[0106] Figure 5 It is the average value and standard deviation of the fatigue load (displacement) of wind turbines 1-6 in the "power enhancement" mode of the present invention, where (a) is the average value of the front and rear displacements of wind turbines 1-6, (b) is the standard deviation of the front and rear displacements of wind turbines 1-6, (c) is the average value of the left and right displacements of wind turbines 1-6, and (d) is the standard deviation of the left and right displacements of wind turbines 1-6.

[0107] Figure 6 It is a simulation result diagram of the total power generation of the wind farm when the wind speed is abundant in the "power tracking" mode of the present invention.

[0108] Figure 7 It is a result diagram of the fatigue load of wind turbine 3 when the wind speed is abundant in the "power tracking" mode of the present invention.

[0109] Figure 8 This is a simulation result diagram of the total power generation of a wind farm when the wind speed is scarce in the "power tracking" mode of the present invention.

[0110] Figure 9 This is a result diagram of the fatigue load of wind turbine 5 when the wind speed is scarce in the "power tracking" mode of the present invention. Detailed implementation manners

[0111] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0112] As Figure 1 shown, the dual-mode dynamic active wake control method for a wind farm in this embodiment includes the following steps:

[0113] S1. For the jointly deployed virtual wind farm and real wind farm, the virtual wind farm controller determines the current control mode of the virtual wind farm. If it is the power enhancement mode, the optimal control instruction is obtained by solving the state space model of the wind farm with the optimization objectives of maximum power output, power balance, and fatigue load suppression; if it is the power tracking mode, the optimal control instruction is obtained by solving the state space model of the wind farm with the optimization objective of maximum power output; the optimal control instruction is sent to the virtual wind farm, and after being delayed according to the wake delay time, the obtained delayed control instruction is sent to the real wind farm controller;

[0114] S2. The real wind farm controller determines the current control mode of the real wind farm. If it is the power enhancement mode, the delayed control instruction is directly output and sent to the real wind farm; if it is the power tracking mode, the final optimal control instruction is obtained by solving the state space model of the wind farm with the optimization objectives of power reference tracking, control reference tracking, and fatigue load suppression and sent to the real wind farm.

[0115] As Figure 2As shown in the figure, in this embodiment, the real wind farm includes real wind turbines and a real wind farm controller. The real wind farm has the following characteristics: the real wind farm includes real wind turbines, collector lines, medium-voltage transformers, high-voltage transformers, an external power grid, a wake effect with time-delay characteristics, a state information measurement system, etc.; the real wind farm controller outputs the control instructions of the real wind farm according to the current working mode and the actual wind conditions of the real wind farm. The virtual wind farm includes virtual wind turbines and a virtual wind farm controller. The virtual wind farm has the following characteristics: the virtual wind farm includes virtual wind turbines, and the virtual wind turbines are programs, codes, or digital models that can reflect the dynamic characteristics of real wind turbines. The number and spatial arrangement of the virtual wind turbines are consistent with those of the real wind turbines; the incoming wind speed of the virtual wind turbines is calculated by an engineering wake model, and the engineering wake model does not consider the time-delay characteristics; the virtual wind farm controller outputs the control instructions of the virtual wind farm according to the current working mode of the real wind farm and the actual wind conditions of the virtual wind farm.

[0116] In step S1 of this embodiment, when solving the state space model of the wind farm to obtain the optimal control instructions with maximum power output, power balance, and fatigue load suppression as the optimization objectives, the functional expression of the adopted optimization objective is:

[0117] ,

[0118] In the above formula, is the optimization objective of the virtual wind farm in the power enhancement mode, is the optimization objective of maximum power output, is the optimization objective of power balance, is the optimization objective of suppressing the fatigue load of the wind farm, is the optimization objective of suppressing the fluctuation of the control quantity, and there are:

[0119] ,

[0120] ,

[0121] ,

[0122] ,

[0123] Among them, is the rated power of the wind farm, is the total number of wind turbines in the wind farm, is the initial power value of the j th wind turbine, is the power increment of the j th wind turbine, is a manually set constant, and are the front - back and left - right displacement increments of the j th wind turbine tower respectively, , , , and are the weights of each optimization objective respectively, and are the voltage increments at the k - th and (k - 1)-th moments respectively, where k is the current optimization time.

[0124] When obtaining the optimal control command by solving the state - space model of the wind farm with the maximum power output as the optimization objective in step S1 of this embodiment, the functional expression of the adopted optimization objective is:

[0125] ,

[0126] where is the optimization problem of the virtual wind farm in the power reference tracking mode, is the optimization objective of maximum power output, is the optimization objective of suppressing the fluctuation of the control quantity, and there is:

[0127] ,

[0128] ,

[0129] where is the rated power of the wind farm, is the total number of wind turbines in the wind farm, is the initial power value of the j th wind turbine, is the power increment of the j th wind turbine, and are the weights of each optimization objective respectively, and are the voltage increments at the k - th and (k - 1)-th moments respectively, where k is the current optimization time.

[0130] The functional expression of the delay time calculation for the delay processing according to the wake delay time in step S1 of this embodiment is:

[0131] ,

[0132] In the above formula, is the delay time of the j th wind turbine in the real wind farm, is the geographical distance along the wind direction between the j th wind turbine and the first wind turbine in the real wind farm, is the average inflow wind speed.

[0133] In step S2 of this embodiment, when obtaining the final optimal control instruction by solving the state - space model of the wind farm with power reference tracking, control reference tracking, and fatigue load suppression as the optimization objectives and sending it to the actual wind farm, the functional expression of the optimization objective adopted is:

[0134] ,

[0135] where, is the optimization objective in the power reference tracking mode of the actual wind farm, is the optimization objective of the power reference tracking of the wind farm, is the optimization objective of the control reference tracking of the wind farm, is the optimization objective of the fatigue load suppression of the wind farm, is the optimization objective of suppressing the fluctuation of the control quantity, and there are:

[0136] ,

[0137] ,

[0138] ,

[0139] ,

[0140] where, is the power reference command of the wind farm, is the total number of wind turbines in the wind farm, is the initial power value of the j th wind turbine, is the power increment of the j th wind turbine, is the generator electromagnetic torque increment of the j th wind turbine, is the initial generator electromagnetic torque value of the j th wind turbine, is the generator electromagnetic torque reference value of the j th real (superscript R indicates) wind turbine, is the blade pitch angle increment of the j th wind turbine, is the initial blade pitch angle value of the j th wind turbine, is the blade pitch angle reference value of the j th real wind turbine, and are the front - back and left - right displacement increments of the tower of the j th wind turbine respectively, and The voltage increments at times k and k - 1 respectively, where k is the current optimization time, 、 、 、 、 and are the weights of each optimization objective respectively.

[0141] In this embodiment, the functional expression of the wind farm state - space model is:

[0142] ,

[0143] ,

[0144] In the above formula, is the system matrix of the wind farm 's first - order derivative, 、 、 、 、 、 and are all coefficient matrices of the wind farm, is the output matrix of the wind farm, is the identity matrix, is the control matrix, and there are , , are the state matrix, output matrix and control matrix of the wind farm respectively, ~ are the state matrices of the 1st to nth wind turbines respectively, ~ are the output matrices of the 1st to nth wind turbines respectively, ~ are the control matrices of the 1st to nth wind turbines respectively. The state matrix 、output matrix and control matrix of any ith wind turbine have the following expressions:

[0145] ,

[0146] ,

[0147] ,

[0148] Wherein, are the increments of the generator speed 、the increment of the generator electromagnetic torque 、the increment of the blade pitch angle 、the front - and - back moving speed of the tower Increment of, front - rear displacement of the tower Increment of, left - right moving speed of the tower Increment of, left - right displacement of the tower Increment of, Are respectively the reference value of the generator electromagnetic torque of the i - th wind turbine Increment of and the reference value of the blade pitch angle Increment of, Is the increment of the electromagnetic power of the i - th wind turbine. Therefore, by solving the optimization problem of the virtual wind farm, the virtual wind farm controller can obtain the reference value of the generator electromagnetic torque of the i - th wind turbine Increment of And the reference value of the blade pitch angle Increment of , and combined with the reference value of the generator electromagnetic torque of the previous step And the reference value of the blade pitch angle The current reference value of the generator electromagnetic torque can be obtained And the reference value of the blade pitch angle , and then directly send them to each virtual wind turbine in the virtual wind farm, and delay this instruction according to the wake delay time and then send it to the real - wind - farm controller. Its function expression is:

[0149] ,

[0150] In the above formula, And Are respectively the reference value of the generator electromagnetic torque before and after the delay, And Are respectively the reference value of the blade pitch angle before and after the delay, Is the current moment, Is the delay time. Similarly, by solving the optimization problem of the virtual wind farm, the real - wind - farm controller can obtain the reference value of the generator electromagnetic torque of the i - th real wind turbine Increment of and the reference value of the blade pitch angle Increment of , and send them to each real wind turbine in the real wind farm.

[0151] Before step S1 in this embodiment, there is also a step of constructing a wind farm state - space model:

[0152] S101, construct the non - linear model of the wind turbine of the two - degree - of - freedom control method shown in the following formula:

[0153] ,

[0154] In the above formula, Is the generator speed The first derivative of is the gearbox ratio, is the equivalent inertia of the impeller and the generator, is the mechanical torque of the impeller, is the electromagnetic torque of the generator, is the blade pitch angle The first derivative of is the pitch angle reference value, is the time constant of the pitch mechanism, is the electromagnetic torque of the generator The first derivative of is the electromagnetic torque reference value of the generator, is the time constant of the generator, and there is:

[0155] , is the air density, is the impeller radius, is the wind energy utilization coefficient of the wind turbine, is the tip speed ratio, is the incoming wind speed, is the impeller speed; the superscript ref is the reference command;

[0156] S102. Construct a linearized model of the wind turbine according to the non - linear model of the wind turbine:

[0157] ,

[0158] In the above formula, Δ represents the increment, is the initial value of the mechanical torque of the impeller, T e0 is the initial value of the electromagnetic torque of the generator, and the increment of the mechanical torque of the impeller is:

[0159] ,

[0160] In the above formula, is the increment of the incoming wind speed;

[0161] S103. Set the fatigue load index of the wind turbine to the front - back and left - right displacements of the tower:

[0162] ,

[0163] ,

[0164] Among them, and are the front - back and left - right displacements of the tower respectively, and are the front - back and left - right moving speeds of the tower respectively, , are the second and first derivatives of the front and rear displacements of the tower respectively, and are the second and first derivatives of the left and right displacements of the tower respectively, and and are the mass, damping and stiffness of the tower respectively, is the tower thrust, is the force on the nacelle, and there is: and where is the air density, is the impeller radius, is the thrust coefficient, is the tip speed ratio, is the blade pitch angle, is the inflow wind speed, is the tower height, is the gearbox ratio, is the torque of the low-speed shaft of the wind turbine, and there is:

[0165] and

[0166] where and are the inertias of the impeller and the generator respectively;

[0167] S104, construct the linearized form of the fatigue load index shown in the following formula:

[0168] and

[0169] In the above formula, Δ represents the increment, and there is:

[0170] and

[0171] ;

[0172] S105, include the state space model of the i th wind turbine shown in the following formula:

[0173] and

[0174] and

[0175] where is the state matrix of the i th wind turbine, is the control matrix of the i th wind turbine, is the wind turbine i is the increment of the inflow wind speed, is the output matrix of the i-th wind turbine, is the first derivative of, 、 、 、 、 are the coefficient matrices of the state space model of the i-th wind turbine, and there are:

[0176] ,

[0177] ,

[0178] ,

[0179] ,

[0180] , ,

[0181] , ,

[0182] Among them, are respectively the increment of the generator speed of the i-th wind turbine, the increment of the generator electromagnetic torque of the i-th wind turbine, the increment of the blade pitch angle of the i-th wind turbine, the increment of the fore-aft moving speed of the tower of the i-th wind turbine, the increment of the fore-aft displacement of the tower of the i-th wind turbine, the increment of the left-right moving speed of the tower of the i-th wind turbine, the increment of the left-right displacement of the tower of the i-th wind turbine, are respectively the increment of the reference value of the generator electromagnetic torque of the i-th wind turbine and the increment of the reference value of the blade pitch angle of the i-th wind turbine, is the increment of the electromagnetic power of the i-th wind turbine, is the impeller mechanical torque of the i-th wind turbine, is the initial value of the impeller mechanical torque of the i-th wind turbine, is the initial value of the generator electromagnetic torque of the i-th wind turbine, is the generator speed of the i-th wind turbine, is the initial value of the generator speed of the i-th wind turbine, is the blade pitch angle of the i-th wind turbine, is the tower thrust of the i-th wind turbine, is the inflow wind speed of the i-th wind turbine, μis the generator efficiency;

[0183] S106, construct the linearized form of the wind farm wake model shown by the following formula:

[0184] ,

[0185] ,

[0186] where, is the change in the wake wind speed caused by the i th wind turbine on the n th wind turbine, is the wake wind speed caused by the i th wind turbine on the n th wind turbine, is the change in the thrust coefficient of the i th wind turbine, is the change in the incoming wind speed i of the th wind turbine, and there is:

[0187] ,

[0188] ,

[0189] where, is the wake expansion coefficient of the Jensen wake model, is the distance along the wind direction between the n th virtual wind turbine and the th wind turbine, is the wake coverage area, is the swept area of the impeller;

[0190] S107, according to the linearized form of the wind farm wake model, determine the incremental form of the predicted incoming wind speed of the 1st to nth wind turbines:

[0191] ,

[0192] where, is the incremental matrix of the wind speed prediction value, is the state matrix of the wind farm, and are the coefficient matrices, and there is:

[0193] ,

[0194] ,

[0195] Coefficient matrix M and N any element in and The expression is:

[0196] , ,

[0197] Based on the coefficient matrix N being a lower triangular matrix and all diagonal elements being 0, the incremental form of the predicted values of the inflow wind speeds of the 1st to nth wind turbines is simplified to:

[0198] ,

[0199] wherein, is the identity matrix;

[0200] S108. Construct the state - space model of the wind farm:

[0201] ,

[0202] ,

[0203] wherein, , , are respectively the state matrix, output matrix and control matrix of the wind farm, , , , , are respectively the coefficient matrices of the wind - farm state - space model, ~ are respectively the state matrices of the 1st to nth wind turbines, ~ are respectively the output matrices of the 1st to nth wind turbines, ~ are respectively the control matrices of the 1st to nth wind turbines, ~ are respectively the coefficient matrices of the 1st to nth wind turbines , ~ are respectively the coefficient matrices of the 1st to nth wind turbines , ~ are respectively the coefficient matrices of the 1st to nth wind turbines , ~ are respectively the coefficient matrices of the 1st to nth wind turbines , ~ are respectively the coefficient matrices of the 1st to nth wind turbines ; Substitute the incremental forms of the predicted values of the incoming flow wind speeds of the first to nth wind turbines to obtain a simplified form of the wind farm state space model:

[0204] ,

[0205] ,

[0206] for obtaining an optimal control instruction by combining with a specified target to solve the wind farm state space model. Moreover, under the power tracking mode and the power enhancement mode, the wind farm state space model optimization problems in the real wind farm controller and the virtual wind farm controller both have the following constraint conditions: (1) To protect the pitch system and the generator, the amplitudes and action rates of the electromagnetic torque and the pitch angle need to be restricted:

[0207] ,

[0208] wherein, is the initial value of the electromagnetic torque of the ith wind turbine, is the initial value of the pitch angle of the ith wind turbine, is the increment of the electromagnetic torque of the ith wind turbine, is the increment of the pitch angle of the ith wind turbine, and are the upper limits of the electromagnetic torque and the pitch angle, and are respectively the upper limits of the change rates of the electromagnetic torque and the pitch angle. (2) The power outputs of each wind turbine need to be restricted below the rated power and be higher than a certain value to prevent the wind turbines from shutting down; at the same time, the motor speeds need to be restricted within a reasonable range to ensure operation safety, that is:

[0209] ,

[0210] wherein, is the power lower limit matrix, is the rated power matrix, , , is the power lower limit, is the rated power, is an n×1 matrix with all elements being 1, is the power initial value matrix, ~ are respectively the initial values of the electromagnetic powers of wind turbines 1 to n; is the speed lower limit matrix, is the speed upper limit matrix, , , is the speed lower limit, is the speed upper limit, is the initial rotational speed matrix, ~ which are the initial rotational speeds of wind turbines 1 to n respectively; is the power sensitivity matrix, and , ~ which are the power sensitivity matrices of wind turbines 1 to n respectively, is the generator efficiency, is the initial electromagnetic torque of wind turbine n; is the rotational speed sensitivity matrix, and , ~ which are the rotational speed sensitivity matrices of wind turbines 1 to respectively, is the number of wind turbines.

[0211] To verify the effectiveness of the above dual-mode dynamic active wake control method for wind farms aiming at power enhancement and load alleviation, an improved virtual-real combined control framework for wind farms was built using SimWindFarm software and MATLAB / Simulink software, as shown in Figure 3 . In this simulation system, SimWindFarm is regarded as a real offshore wind farm, including 6×6 NREL 5MW wind turbines arranged along the wind direction, and the front, rear and lateral distances between each wind turbine are all 630 meters. The simulation environment is set as a time-varying wind condition with an average wind speed of 10m / s and a turbulence intensity of 10%.

[0212] Figure 4 This is the simulation result diagram of the total power generation of the wind farm in the "power enhancement" mode of this embodiment. Among them, "the proposed strategy" is the dynamic wake control method of the wind farm in this embodiment (the proposed strategy hereinafter all refers to the dynamic wake control method of the wind farm in this embodiment), "maximum power output" is the control method without considering the optimization objectives of power balance and fatigue load alleviation, and "greedy control" is the control method in which all wind turbines operate in the maximum power point tracking mode. It can be seen that the total power generation of the wind farm under the maximum power output strategy control is the highest among the three methods, while the total power generation of the wind farm under the proposed strategy control is slightly reduced, and the greedy control strategy is the lowest, with average powers of 80.62, 79.84 and 74.51 MW respectively. Figure 5This is the average value and standard deviation of the fatigue loads (displacements) of wind turbines 1 to 6 in the "power enhancement" mode of this embodiment, where the fatigue loads (displacements) include left-right displacements and front-back displacements. The blue, red, and yellow solid lines are the average values and standard deviations of the fatigue loads of each unit under the proposed strategy, the maximum power output control strategy, and the greedy control strategy, respectively. The dashed lines are the average values of the standard deviations of the fatigue loads of the wind farm. It can be seen that the average value and standard deviation of the fatigue loads of wind turbines 1 to 6 under the greedy control decrease in turn, and the average value of the standard deviation of the fatigue loads of the wind farm is the highest, indicating that the fatigue load distribution is the most uneven and the fluctuation is the largest under this control strategy; while the proposed strategy can obtain the most balanced fatigue load distribution and the lowest fatigue load fluctuation. Overall Figure 4 and Figure 5 it can be summarized that: compared with the greedy control strategy, the proposed strategy can improve the total power generation of the wind farm while making the fatigue load distribution of the wind farm more balanced; compared with the maximum power output strategy, the proposed strategy sacrifices a little power output and can obtain a more balanced fatigue load distribution.

[0213] Figure 6 This is the simulation result diagram of the total power generation of the wind farm when the wind speed is sufficient in the "power tracking" mode of this embodiment. Among them, the "classical two-degree-of-freedom" strategy is a control method that does not consider the wake effect, and the "proportional distribution" strategy is a control method that calculates the power reference value according to the current available power of all wind turbines in proportion. It can be seen that due to the two-degree-of-freedom control of the wind turbines, the total power generation of the wind farm fluctuates greatly under the proposed strategy and the classical two-degree-of-freedom strategy. The root mean square errors of the total power generation of the wind farm for the power reference under the three control methods are 0.0303, 0.0307, and 0.0097 MW, respectively, and the fluctuations are still within the acceptable range. Figure 7 This is the result diagram of the fatigue load of wind turbine 3 when the wind speed is sufficient in the "power tracking" mode of this embodiment. It can be seen that compared with the proportional distribution strategy, both the proposed strategy and the classical two-degree-of-freedom strategy can significantly reduce the tower displacement of the wind turbine. In addition, the control effects of the proposed strategy and the classical two-degree-of-freedom strategy are similar when the wind speed is sufficient.

[0214] Figure 8This is the simulation result diagram of the total power generation of the wind farm when the wind speed is scarce in the "power tracking" mode of this embodiment. It can be seen that when the wind speed is scarce, the proportional allocation strategy fails to achieve the tracking of the reference power starting from 900 seconds. Since the classical two-degree-of-freedom strategy itself will actively adjust the pitch angle and electromagnetic torque of the upstream wind turbines, to a certain extent, it changes the wake distribution of the wind farm. Therefore, the total power generation of the wind farm under the condition of scarce wind speed has a certain increase. In contrast, the proposed strategy can increase the total power generation of the wind farm the most, can track the power reference for a longer time, and achieve better power tracking performance. From 900 to 1400 seconds when the power reference is the highest, the total power generation of the wind farm under the control of the proposed strategy can reach 83.11 MW, while that under the control of the classical two-degree-of-freedom strategy is only 80.41 MW. Figure 9 This is the result diagram of the fatigue load of wind turbine 5 when the wind speed is scarce in the "power tracking" mode of this embodiment. It can be seen that from 0 to 900 seconds when the power reference is relatively low, the fluctuation of the tower displacement under the proportional allocation strategy is the largest among the three strategies, while the fluctuations of the tower displacement under the proposed strategy and the classical two-degree-of-freedom strategy are similar. From 900 to 1400 seconds when the power reference is relatively high, under the condition of scarce wind speed, the wind farm is required to generate higher power. Therefore, the fluctuation of the tower displacement under the classical two-degree-of-freedom strategy is very serious. However, the proposed strategy improves the wind speed distribution inside the wind farm through active wake control, making the inflow wind speed of the downstream turbines higher. Therefore, the fluctuation of the tower displacement can be kept within an acceptable range. Comprehensive Figure 8 and Figure 9 It can be concluded that the proposed strategy can improve the power output of the wind farm under the condition of scarce wind speed, so as to achieve better power tracking performance, and at the same time can suppress the load fluctuation of the wind turbines.

[0215] In addition, this embodiment also provides a dual-mode dynamic active wake control system for a wind farm, including a microprocessor and a memory connected to each other. The microprocessor is programmed or configured to execute the dual-mode dynamic active wake control method for the wind farm. In addition, this embodiment also provides a computer-readable storage medium storing a computer program or instruction, which is programmed or configured to execute the dual-mode dynamic active wake control method for the wind farm through a processor. In addition, this embodiment also provides a computer program product including a computer program or instruction, which is programmed or configured to execute the dual-mode dynamic active wake control method for the wind farm through a processor. Those skilled in the art should understand that the technical solutions provided by the embodiments of the present application can be in the form of a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes. The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, as well as the combination of processes and / or blocks in the flowchart and / or block diagram. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0216] The above are only the preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A dual-mode dynamic active wake control method for a wind farm, characterized in that: The steps include: S1, for the jointly deployed virtual wind farm and real wind farm, the virtual wind farm controller determines the current control mode of the virtual wind farm. If it is the power efficiency mode, the wind farm state space model is solved with the maximum power output, power balance and fatigue load smoothing as the optimization objectives to obtain the optimal control instruction; if it is the power tracking mode, the wind farm state space model is solved with the maximum power output as the optimization objective to obtain the optimal control instruction; the optimal control instruction is sent to the virtual wind farm, and delayed according to the wake delay time, and the obtained delayed control instruction is sent to the real wind farm controller; S2, the real wind farm controller determines the current control mode of the real wind farm. If it is the power efficiency mode, it directly outputs the delayed control instruction and sends it to the real wind farm; If it is the power tracking mode, the wind farm state space model is solved with power reference tracking, control reference tracking and fatigue load smoothing as the optimization objectives to obtain the final optimal control instructions and send them to the real wind farm.

2. The wind farm dual-mode dynamic active wake control method according to claim 1 is characterized in that: In step S1, when solving the wind farm state space model to obtain the optimal control instruction with maximum power output, power balance and fatigue load smoothing as the optimization objectives, the function expression of the optimization objective adopted is: ， In the above formula, is the optimization target of the virtual wind farm in the power efficiency mode. For the optimization goal of maximum power output, For the optimization goal of power balance, The optimization goal of fatigue load suppression in wind farms is is the optimization goal of suppressing the fluctuation of the control quantity, and: ， ， ， ， in, is the rated power of the wind farm, is the total number of wind turbines in the wind farm, For the j The initial power value of the typhoon generator, For the j The power increment of the typhoon turbine, is a manually set constant, and Respectively j The displacement increments of the typhoon tower, front and back, left and right, , , , and are the weights of each optimization objective, and are the voltage increments at time k and k-1 respectively, and k is the current optimization time.

3. The wind farm dual-mode dynamic active wake control method according to claim 1, characterized in that: When solving the wind farm state space model to obtain the optimal control command with the maximum power output as the optimization target in step S1, the function expression of the optimization target used is: ， in, is the optimization target of the virtual wind farm in power tracking mode, For the optimization goal of maximum power output, is the optimization goal of suppressing the fluctuation of the control quantity, and: ， ， in, is the rated power of the wind farm, is the total number of wind turbines in the wind farm, For the j The initial power value of the typhoon generator, For the j The power increment of the typhoon turbine, and are the weights of each optimization objective, and are the voltage increments at time k and k-1 respectively, and k is the current optimization time.

4. The wind farm dual-mode dynamic active wake control method according to claim 1, characterized in that: The function expression for calculating the delay time for delay processing according to the wake delay time in step S1 is: ， In the above formula, For the real wind farm j The delay time of the typhoon, For the real wind farm j The geographical distance between the typhoon and the first typhoon along the wind direction, is the average inflow wind speed.

5. The wind farm dual-mode dynamic active wake control method according to claim 1, characterized in that: In step S2, when solving the wind farm state space model with power reference tracking, control reference tracking and fatigue load smoothing as optimization objectives to obtain the final optimal control instructions and send them to the real wind farm, the function expression of the optimization objective used is: ， in, is the optimization target of a real wind farm in power tracking mode, is the optimization target of wind farm power reference tracking, is the optimization target of wind farm control reference tracking, The optimization goal of fatigue load suppression in wind farms is is the optimization goal of suppressing the fluctuation of the control quantity, and: ， ， ， ， in, is the power reference command of the wind farm, is the total number of wind turbines in the wind farm, For the j The initial power value of the typhoon generator, For the j The power increment of the typhoon turbine, For the j The electromagnetic torque increment of the wind turbine generator, For the j The initial value of the electromagnetic torque of the wind turbine generator, For the j The reference value of the electromagnetic torque of the generator of a real wind turbine, For the j The blade pitch angle increment of the typhoon turbine, For the j The initial value of the blade pitch angle of the typhoon turbine, For the j The blade pitch angle reference value of a real wind turbine, and Respectively j The displacement increments of the typhoon tower, front and back, left and right, and are the voltage increments at time k and k-1 respectively, k is the current optimization time, , , , , and are the weights of each optimization objective.

6. The wind farm dual-mode dynamic active wake control method according to claim 1, characterized in that: The functional expression of the wind farm state space model is: ， ， In the above formula, is the state matrix of the wind farm The first derivative of , , , , , and are the coefficient matrices of wind farms, is the output matrix of the wind farm, is the identity matrix, is the control matrix, and , , are the state matrix, output matrix and control matrix of the wind farm respectively. ~ are the state matrices of the 1st to nth wind turbines, ~ are the output matrices of the 1st to nth fans respectively, ~ are the control matrices of the 1st to nth fans respectively, and the state matrix of any i-th fan is , output matrix and control matrix The expression is: ， ， ， in, are the generator speed of the i-th wind turbine respectively The increment of the generator electromagnetic torque The increment of blade pitch angle The increment of the tower's forward and backward movement speed The increment of the tower's front and rear displacement The increment of the tower's left and right movement speed The increment and the left and right displacement of the tower The increment of are the reference values ​​of the electromagnetic torque of the generator of the i-th wind turbine The increment and blade pitch angle reference value The increment of is the increment of electromagnetic power of the i-th fan.

7. The wind farm dual-mode dynamic active wake control method according to claim 6, characterized in that: Before step S1, the step of constructing a wind farm state space model is also included: S101, construct a nonlinear model of the wind turbine generator system using a two-degree-of-freedom control method as shown in the following formula: ， In the above formula, is the generator speed The first derivative of is the gearbox ratio, is the equivalent inertia of the impeller and generator, is the impeller mechanical torque, is the electromagnetic torque of the generator, is the blade pitch angle The first derivative of is the pitch angle reference value, is the time constant of the pitch mechanism, is the electromagnetic torque of the generator The first derivative of is the reference value of the electromagnetic torque of the generator, is the time constant of the generator, , is the air density, is the impeller radius, is the wind energy utilization coefficient of the fan, is the tip speed ratio, is the inflow wind speed, is the impeller speed; S102, constructing a linearized model of the wind turbine generator set according to the nonlinear model of the wind turbine generator set: ， In the above formula, Indicates the increment, for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of is the initial value of the impeller mechanical torque, is the initial value of the generator electromagnetic torque, and the increment of the impeller mechanical torque is: ， In the above formula, is the increment of inflow wind speed; S103, construct the fatigue load index of the wind turbine as the front-to-back and left-to-right displacement of the tower: ， ， in, and are the front-to-back and left-to-right displacements of the tower, and are the forward and backward and left and right moving speeds of the tower respectively, , are the second and first order derivatives of the tower's front and rear displacements, , are the second and first order derivatives of the left and right displacements of the tower, , , are the mass, damping and stiffness of the tower respectively, is the tower thrust, is the force on the nacelle, and there are: , ,in is the air density, is the impeller radius, is the thrust coefficient, is the tip speed ratio, is the blade pitch angle, is the inflow wind speed, is the tower height, is the gearbox ratio, is the low speed shaft torque of the fan, and: ， in, , are the inertias of impeller and generator respectively; S104, construct a linearized form of the fatigue load index shown in the following formula: ， In the above formula, Δ represents the increment, for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of for The increment of , and: ， ； S105, construct the fatigue load index including the following formula i Typhoon turbine state space model: ， ， in, is the state matrix of the i-th fan, for The first derivative of is the control matrix of the i-th fan, for The first derivative of For fans i The increase in inflow wind speed, is the output matrix of the i-th fan, , , , , is the coefficient matrix of the state space model of the i-th wind turbine, and: ， ， ， ， , ， , , in, are the generator speed of the i-th wind turbine respectively The increment of the generator electromagnetic torque The increment of blade pitch angle The increment of the tower's forward and backward movement speed The increment of the tower's front and rear displacement The increment of the tower's left and right movement speed The increment and the left and right displacement of the tower The increment of are the reference values ​​of the electromagnetic torque of the generator of the i-th wind turbine The increment and blade pitch angle reference value The increment of is the increment of electromagnetic power of the i-th fan, is the mechanical torque of the impeller of the i-th fan, is the initial value of the turbine mechanical torque of the i-th fan, is the initial value of the electromagnetic torque of the generator of the i-th wind turbine, is the generator speed of the i-th wind turbine, is the initial value of the generator speed of the i-th wind turbine, is the blade pitch angle of the i-th wind turbine, is the tower thrust of the i-th wind turbine, is the inflow wind speed of the i-th fan, is the generator efficiency; S106, construct the linearized form of the wind farm wake model shown in the following formula: ， ， in, For the i Typhoon machine n Wake wind speed caused by typhoon The amount of change, For the i Typhoon machine n The wind speed of the typhoon's wake is For the i Thrust coefficient of typhoon The amount of change, For the i The thrust coefficient of the typhoon, For the i Inflow wind speed of typhoon The amount of change, For the i The inflow wind speed of the typhoon is: ， ， in, is the wake expansion coefficient of Jensen wake model, is the i-th fan and the n The distance between virtual wind turbines along the wind direction, is the impeller radius, is the wake coverage area, is the impeller swept area; S107, determining the incremental form of the predicted values ​​of the inflow wind speeds of the first to nth wind turbines according to the linearized form of the wind farm wake model: ， in, is the incremental matrix of wind speed prediction values, is the state matrix of the wind farm, and is a coefficient matrix, and has: ， ， Coefficient Matrix M and N Any element in and The expression is: ， ， Based on the coefficient matrix N is a lower triangular matrix with all diagonal elements being 0. The incremental form of the predicted inflow wind speed of the 1st to nth wind turbines is simplified to: ， in, is the identity matrix; S108, constructing a wind farm state space model: , ， in, , , are the state matrix, output matrix and control matrix of the wind farm respectively. , , , , are the coefficient matrices of the wind farm state space model, ~ are the state matrices of the 1st to nth wind turbines, ~ are the output matrices of the 1st to nth fans respectively, ~ are the control matrices of the 1st to nth fans respectively, ~ are the coefficient matrices of the 1st to nth fans respectively , ~ are the coefficient matrices of the 1st to nth fans respectively , ~ are the coefficient matrices of the 1st to nth fans respectively , ~ are the coefficient matrices of the 1st to nth fans respectively , ~ are the coefficient matrices of the 1st to nth fans respectively ; Substitute the incremental form of the predicted inflow wind speed of the 1st to nth wind turbines to obtain the simplified form of the wind farm state space model: ， ， It is used to solve the wind farm state space model in combination with the specified objectives to obtain the optimal control instructions.

8. A wind farm dual-mode dynamic active wake control system, comprising a microprocessor and a memory connected to each other, characterized in that: The microprocessor is programmed or configured to execute the wind farm dual-mode dynamic active wake control method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program or instruction stored therein, characterized in that: The computer program or instruction is programmed or configured to execute the wind farm dual-mode dynamic active wake control method according to any one of claims 1 to 7 through a processor.

10. A computer program product comprising a computer program or instructions, characterized in that The computer program or instruction is programmed or configured to execute the wind farm dual-mode dynamic active wake control method according to any one of claims 1 to 7 through a processor.

Images