A wind power active support frequency drop suppression method for optimizing frequency modulation parameters

Abstract

The application provides a wind power active support frequency modulation parameter optimization method for suppressing frequency drop, comprising the following steps: establishing a system model of a doubly-fed wind turbine connected to a power grid; determining frequency drop characteristics of the doubly-fed wind turbine when participating in power grid frequency modulation, wherein the doubly-fed wind turbine participates in power grid frequency modulation through a comprehensive inertia control strategy, and the frequency drop characteristics include primary frequency drop characteristics, secondary frequency drop characteristics and system frequency characteristics when the doubly-fed wind turbine exits power grid frequency modulation; selecting frequency modulation parameters to be optimized based on the frequency drop characteristics of the doubly-fed wind turbine when participating in power grid frequency modulation; and searching for optimal values of the frequency modulation parameters by using an improved particle swarm algorithm, wherein constraint conditions of the improved particle swarm algorithm include a coupling term of rotor safety and power output safety of the doubly-fed wind turbine. The frequency modulation parameters obtained by using the method provided by the application can effectively suppress the secondary frequency drop problem of the doubly-fed wind turbine when participating in power grid frequency modulation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent control of wind power generation, and relates to a wind power grid-connected frequency modulation optimization technology, and particularly provides a wind power active support frequency modulation parameter optimization method for suppressing frequency drop. BACKGROUND

[0002] The doubly-fed wind turbine is the most widely used variable-speed constant-frequency wind turbine at present, which is connected to the power grid through a converter, and the speed is decoupled from the frequency of the power grid, so it cannot release the shaft energy and automatically provide inertia response like a synchronous machine. Therefore, when the frequency of the power grid system fluctuates, in order to use the doubly-fed wind turbine to participate in the frequency modulation of the power system and provide active support to the power grid, a comprehensive inertia control strategy is generally used for the frequency modulation of the power grid by the doubly-fed wind turbine.

[0003] When the above strategy is used to control the doubly-fed wind turbine to participate in the frequency modulation of the power grid, in the late stage of frequency regulation, the doubly-fed wind turbine must absorb the system active power to restore the speed, which causes the problem of secondary frequency drop. In order to alleviate the problem of secondary frequency drop of the power grid during the speed recovery process of the doubly-fed wind turbine as much as possible, the frequency first drop, the frequency second drop and the system frequency after the maximum power point tracking (MPPT) mode of the doubly-fed wind turbine participating in the frequency modulation of the power grid are comprehensively considered, so as to reasonably set the multiple control parameters of the doubly-fed wind turbine participating in the frequency modulation of the power grid.

[0004] However, the existing control strategy research for the wind turbine participating in frequency modulation generally studies the power change or energy storage control, which can to some extent achieve the purpose of suppressing the secondary drop of the system frequency, but does not conduct in-depth research on the dynamic characteristics of the frequency first drop and the frequency second drop of the wind turbine participating in the frequency modulation of the power grid, and cannot optimize the value range of the droop coefficient and the inertia response coefficient in combination with the operating state of the wind turbine, so the frequency support function of the wind turbine itself cannot be fully exerted. SUMMARY

[0005] To solve the problems in the prior art, the application provides a wind power active support frequency modulation parameter optimization method for suppressing frequency drop, which comprises the following steps:

[0006] S1, a system model of grid-connected generation of a doubly-fed wind turbine is established;

[0007] S2, frequency drop characteristics of the doubly-fed wind turbine participating in the frequency modulation of the power grid are analyzed, wherein the doubly-fed wind turbine participates in the frequency modulation of the power grid through a comprehensive inertia control strategy, and the frequency drop characteristics include frequency first drop characteristics, frequency second drop characteristics and system steady-state frequency characteristics after the doubly-fed wind turbine exits the frequency modulation of the power grid;

[0008] S3, selecting frequency modulation parameters that need to be optimized based on the frequency drop characteristics of the doubly-fed wind turbine when participating in grid frequency modulation;

[0009] S4, using an improved particle swarm algorithm to search for an optimal value of the frequency modulation parameter, wherein the constraint conditions of the improved particle swarm algorithm include a coupling term between rotor safety and power output safety of the doubly fed wind turbine.

[0010] Furthermore, the system model of the doubly fed wind turbine grid-connected power generation includes a doubly fed wind turbine, a rotor-side converter, a grid-side converter and a phase-locked loop; the rotor-side converter and the grid-side converter are connected back-to-back through a DC bus capacitor; the doubly fed wind turbine converts the captured wind energy into electrical energy, and its rotor side is connected to the grid through the back-to-back connected rotor-side converter and grid-side converter, and its stator side is directly connected to the grid; the phase-locked loop is used to synchronize the frequency of the doubly fed wind turbine with the grid system frequency.

[0011] Furthermore, the doubly-fed wind turbine participates in grid frequency regulation through a comprehensive inertia control strategy, specifically, the doubly-fed wind turbine provides an electromagnetic power increment to the grid based on the following formula:

[0012]

[0013] Where ΔP W is the electromagnetic power increment provided by the doubly fed wind turbine to the grid, K W1 is the droop coefficient, K W2 is the inertia response coefficient, Δf is the system frequency f of the power grid and the system frequency reference value f ref The frequency deviation between .

[0014] Furthermore, the frequency primary drop characteristics include a time domain expression of the system frequency of the power grid during the frequency primary drop phase, a time instant of the lowest point of the system frequency primary drop, and a maximum deviation of the system frequency primary drop;

[0015] The time domain expression of the system frequency of the power grid during the first frequency drop phase is as follows:

[0016]

[0017] Where ΔP L is the load increment in the power grid system, T J is the inertia time constant of the synchronous unit, T G K is the time constant of the speed regulator of the synchronous unit, G Adjust power for equivalent units of synchronous units;

[0018] The moment when the system frequency drops to the lowest point is determined by the following formula:

[0019]

[0020] wherein t nadir1 is the moment of the first system frequency drop minimum point, t0 is the moment when the doubly-fed wind turbine starts to participate in the grid frequency modulation;

[0021] The maximum value of the first system frequency drop deviation is determined by the following formula:

[0022] Δf max1 = a5-a1exp[a2(θ-a4) / a3]cosθ,

[0023] wherein Δf max1 is the maximum value of the first system frequency drop deviation.

[0024] Further, the second frequency drop characteristic includes a time-domain expression of the system frequency of the grid in the second frequency drop stage, a moment of the second frequency drop minimum point, and a maximum value of the second frequency drop deviation;

[0025] The time-domain expression of the system frequency of the grid in the second frequency drop stage is shown in the following formula:

[0026]

[0027] wherein f1 is the system frequency at the moment t1 when the doubly-fed wind turbine exits the grid frequency modulation, P u is the system power shortage caused by the speed recovery of the doubly-fed wind turbine;

[0028] The moment of the second frequency drop minimum point is determined by the following formula:

[0029]

[0030] wherein t nadir2 is the moment of the second frequency drop minimum point;

[0031] The maximum value of the second frequency drop deviation is determined by the following formula:

[0032] Δf max2 = 1-f1+b5-b1exp[b2(θ1-b4) / b3]cosθ1,

[0033] wherein Δf max2 is the maximum value of the second frequency drop deviation.

[0034] Further, the steady-state frequency characteristic of the system after the doubly-fed wind turbine exits the grid frequency modulation is determined by the following formula:

[0035]

[0036] wherein Δf sP W0 is a steady-state frequency deviation value of the power grid system after the doubly-fed wind turbine is restored to the maximum power point tracking mode.

[0037] Preferably, the frequency modulation parameters to be optimized include K W1 , K W2 , t1 and ΔP d , wherein ΔP d is a difference between the mechanical power and the electromagnetic power of the doubly-fed wind turbine at the time t1 when the doubly-fed wind turbine exits the power grid frequency modulation.

[0038] Preferably, the objective function of the improved particle swarm algorithm is:

[0039] min{max{Δf max1 ,Δf max2}};

[0040] The constraint conditions of the improved particle swarm algorithm include:

[0041]

[0042] wherein H W is an inertia time constant of the rotor of the doubly-fed wind turbine, ω r (t0) is an initial rotational speed of the doubly-fed wind turbine, ω min is a lower limit of the rotational speed of the doubly-fed wind turbine, K W1,max is an upper limit of the value of K W1 , K W2,max is an upper limit of the value of K W2 , and K W1,max and K W2,max are rotor safety and power output safety coupling terms of the doubly-fed wind turbine.

[0043] Preferably, K W1,max and K W2,max are determined by the following steps:

[0044] A kinetic energy calculation factor k m of the doubly-fed wind turbine at the current rotational speed is determined based on the following formula:

[0045]

[0046] wherein and are per-unit values of an actual value ω r , an upper limit value ω r_max and a lower limit value ω r_min of the rotational speed of the rotor of the doubly-fed wind turbine, and a reference value thereof is a synchronous speed ω rn of the doubly-fed wind turbine.

[0047] The power calculation factor k of the doubly fed wind turbine at the current speed is determined based on the following formula: c :

[0048]

[0049] Among them, k opt is the overspeed load reduction curve coefficient, Its per-unit value, ΔP e is the frequency modulation active power increment, is its per-unit value, K is the coefficient of frequency modulation active power increment, P w 、P rated are the actual value and rated value of the active output of the doubly fed wind turbine, The per-unit value is the lower limit of the active power output of the doubly-fed wind turbine;

[0050] K is determined based on the following formula W1,max With K W2,max :

[0051]

[0052] Where ε is the FM participation coefficient, and K0, K1, K2 and n are all constants.

[0053] Preferably, the improved particle swarm algorithm uses the following formula to adaptively update the inertia weight factor in the particle swarm algorithm:

[0054]

[0055] Among them, w is the inertia weight factor, w d 、w u is the initial value and the final value of w, k is the current number of iterations, k max is the total number of iterations;

[0056] The improved particle swarm algorithm uses the following formula to adaptively update the learning factor in the particle swarm algorithm:

[0057]

[0058] Among them, c1 and c2 are learning factors, c 1d 、c 1u is the initial value and final value of c1, c 2d 、c 2u are the initial and final values ​​of c2.

[0059] The optimization method provided in this application conducts a time-domain analytical analysis of the primary drop, secondary drop and steady-state recovery characteristics of the system frequency during the process of the wind power system participating in the grid frequency regulation, and determines the optimal frequency regulation control parameter combination (including exit frequency regulation time, droop coefficient, inertia response coefficient, etc.) under different wind power penetration rates based on the analysis results, so as to ensure that the wind power system can provide stable frequency support under different operating conditions; at the same time, in the parameter optimization process, the safety requirements of the operating state of the doubly fed wind turbine are taken into consideration, and the value range of the droop coefficient and the inertia response coefficient is more reasonably constrained by the coupling terms including rotor rotation safety and power output safety, thereby improving the convergence speed and solution accuracy of the algorithm, so that the optimal frequency regulation control parameters suitable for different wind power penetration rates can be found quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flow chart of a method for optimizing wind power active support frequency regulation parameters for suppressing frequency drops provided by the present application;

[0061] Figure 2 Schematic diagram of the architecture of a system model for grid-connected power generation of a doubly-fed wind turbine according to an embodiment of the present application;

[0062] Figure 3 Schematic diagram of the relationship between the mechanical power and rotor speed of a doubly-fed wind turbine at different wind speeds according to an embodiment of the present application;

[0063] Figure 4 Schematic diagram of controlling a doubly-fed wind turbine using a comprehensive inertia control strategy;

[0064] Figure 5 This is a schematic diagram of the power variation curve of a doubly-fed wind turbine using a comprehensive inertia control strategy for grid frequency regulation;

[0065] Figure 6 This is a schematic diagram of system frequency changes during the process of doubly-fed wind turbines participating in grid frequency regulation;

[0066] Figure 7 A schematic diagram of the three-dimensional relationship between the maximum value of the primary drop deviation of the system frequency and the inertial response coefficient of the doubly-fed wind turbine droop coefficient provided in an embodiment of the present application;

[0067] Figure 8 A schematic diagram of the three-dimensional relationship between the maximum value of the secondary drop deviation of the system frequency, the system frequency at the time when the doubly fed wind turbine exits frequency regulation, and the power shortage in the system according to an embodiment of the present application;

[0068] Figure 9 A schematic diagram of the three-dimensional relationship between the system steady-state frequency deviation and the droop coefficient and inertial response coefficient of the doubly-fed wind turbine according to an embodiment of the present application;

[0069] Figure 10 Schematic diagram of the three-dimensional relationship between the maximum value of the system frequency change rate and the droop coefficient and inertia response coefficient of the doubly fed wind turbine according to an embodiment of the present application;

[0070] Figure 11 ω is the value of the doubly fed wind turbine at different speeds provided by the embodiment of the present application. r -ε relationship diagram;

[0071] Figure 12 This is a schematic diagram of the architecture of a 3-machine 9-node power system model provided according to specific embodiment 1 of the present application;

[0072] Figure 13 The different K values ​​provided in the specific embodiment 1 of the present application are W1 Schematic diagram of the corresponding system frequency response curve;

[0073] Figure 14 The different K values ​​provided in the specific embodiment 1 of the present application are W1 Schematic diagram of the corresponding doubly fed wind turbine speed curve;

[0074] Figure 15 The different K values ​​provided in the specific embodiment 1 of the present application are W2 Schematic diagram of the corresponding system frequency response curve;

[0075] Figure 16 The different K values ​​provided in the specific embodiment 1 of the present application are W2 Schematic diagram of the corresponding doubly fed wind turbine speed curve;

[0076] Figure 17 Schematic diagram of system frequency response curves corresponding to different t1 provided in specific embodiment 1 of the present application;

[0077] Figure 18 A schematic diagram of an algorithm convergence curve provided according to specific embodiment 1 of the present application;

[0078] Figure 19 A schematic diagram of system frequency response curves corresponding to different frequency modulation parameter combinations provided in specific embodiment 1 of the present application;

[0079] Figure 20 A schematic diagram of a doubly-fed wind turbine speed curve corresponding to different frequency modulation parameter combinations provided in specific embodiment 1 of the present application;

[0080] Figure 21 This is a schematic diagram of the electromagnetic power curves of a doubly-fed wind turbine corresponding to different frequency modulation parameter combinations provided in specific embodiment 1 of the present application. DETAILED DESCRIPTION

[0081] Hereinafter, the present application will be further described based on preferred embodiments with reference to the accompanying drawings.

[0082] The present application provides a method for optimizing wind power active support frequency regulation parameters for suppressing frequency drop, which is used to determine the optimal frequency regulation parameters for a doubly fed wind turbine to participate in the frequency regulation of a power grid system, such as Figure 1 As shown, the method includes the following steps:

[0083] S1, establish a system model of doubly fed wind turbine grid-connected power generation;

[0084] S2, analyzing the frequency drop characteristics of the doubly-fed wind turbine when participating in grid frequency regulation, wherein the doubly-fed wind turbine participates in grid frequency regulation through a comprehensive inertia control strategy, and the frequency drop characteristics include a primary frequency drop characteristic, a secondary frequency drop characteristic, and a system steady-state frequency characteristic after the doubly-fed wind turbine exits grid frequency regulation;

[0085] S3, selecting frequency modulation parameters that need to be optimized based on the frequency drop characteristics of the doubly-fed wind turbine when participating in grid frequency modulation;

[0086] S4, using an improved particle swarm algorithm to search for an optimal value of the frequency modulation parameter, wherein the constraint conditions of the improved particle swarm algorithm include a coupling term between rotor safety and power output safety of the doubly fed wind turbine.

[0087] The specific implementation of this method is described in detail below with reference to the accompanying drawings.

[0088] <1. Establishing a system model for doubly-fed wind turbine power generation>

[0089] Step S1 is used to establish a system model of a doubly-fed induction generator (DFIG) grid-connected power generation, thereby providing a basis for analyzing the time domain variation characteristics of the system frequency when the DFIG participates in grid frequency regulation and extracting frequency regulation parameters.

[0090] Figure 2 FIG. 1 shows a schematic diagram of a system model architecture for a double-fed wind turbine grid-connected power generation system in a specific embodiment. Figure 2 As shown, the doubly-fed wind turbine grid-connected power generation system includes a doubly-fed wind turbine, a rotor-side converter, a grid-side converter and a phase-locked loop.

[0091] Specifically, the main structure of a doubly fed wind turbine includes an impeller, a gearbox, a stator, a rotor, and a transmission system connecting the above parts. The impeller rotates by capturing wind energy, converts the wind energy into mechanical energy through the gearbox, drives the rotor to rotate relative to the stator, and converts the mechanical energy into electrical energy according to the principle of electromagnetic induction.

[0092] Furthermore, the power output by the doubly fed wind turbine needs to be connected to the grid through a converter, such as Figure 2 As shown, the rotor side is connected to the grid through the rotor-side converter and the grid-side converter connected back to back, while the stator side is directly connected to the grid. The rotor-side converter and the grid-side converter are connected back to back through the DC bus capacitor.

[0093] A phase-locked loop (PLL) is connected between the doubly fed wind turbine and the grid. It consists of a phase detector, a low-pass filter, and a voltage-controlled oscillator, and is used to synchronize the frequency of the doubly fed wind turbine with the grid frequency.

[0094] In the specific process of controlling the converter switching devices, the rotor-side converter can adopt the power outer loop control plus current inner loop control method, and the grid-side converter can adopt the DC voltage control plus current inner loop control method. Combined with the phase command output by the phase-locked loop, SPWM signals for controlling each switching device of the rotor-side converter and the grid-side converter are generated respectively.

[0095] <2. Analysis of frequency drop characteristics when doubly-fed wind turbines participate in grid frequency regulation>

[0096] In an embodiment of the present application, after establishing a system model of a doubly fed wind turbine connected to the grid through step S1, the frequency drop characteristics of the doubly fed wind turbine when participating in grid frequency regulation are further analyzed through step S2 to select appropriate frequency regulation parameters to optimize the frequency regulation effect and achieve effective suppression of frequency drops.

[0097] Specifically, in an embodiment of the present application, a doubly-fed wind turbine participates in grid frequency regulation through a comprehensive inertia control strategy. Its frequency drop characteristics when participating in grid frequency regulation include primary frequency drop characteristics, secondary frequency drop characteristics, and system frequency characteristics when exiting grid frequency regulation. Step S2 is described in detail below.

[0098] 2.1 Comprehensive inertia control strategy

[0099] When the power grid is in a stable operating state, the doubly fed wind turbine generally operates in the maximum power point tracking (MPPT) mode. Figure 3 The relationship between mechanical power and wind turbine rotor speed (pitch angle β = 0°) at different wind speeds v (v is the wind speed entering the impeller scanning surface) is shown, and the fitting curve P of the MPPT point at each wind speed is shown. MPPT As shown by the dotted line in the figure, its expression is:

[0100]

[0101] Among them, ω r is the angular velocity of the doubly fed wind turbine, ρ is the air density, R W is the impeller radius, C pmax is the wind energy utilization coefficient C p The maximum value of λopt is the optimal tip speed ratio.

[0102] The rotor side of the doubly fed wind turbine is connected to the power grid via a rotor-side converter. In order to enable it to quickly respond to changes in system frequency, a certain frequency control strategy needs to be adopted. In an embodiment of the present application, the doubly fed wind turbine adopts a comprehensive inertia control strategy to participate in system frequency regulation. Under this control strategy, the doubly fed wind turbine has both system frequency regulation capability and inertia response characteristics.

[0103] Figure 4 shows the principle diagram of the integrated inertia control strategy. Figure 4 As shown, when the wind turbine is operating normally in MPPT mode, its electromagnetic power P W and mechanical power P m Phase balance, when the power grid system has load disturbance, the system frequency f and the system frequency reference value f ref Generate frequency deviation Δf, at this time the double-fed wind turbine adopts integrated inertia control to participate in system frequency regulation, by providing additional electromagnetic power increment ΔP to the system W To achieve active support for the system frequency, specifically, the electromagnetic power increment ΔP W There are the following expressions:

[0104]

[0105] In the above formula, K W1 is the droop coefficient, K W2 is the inertial response coefficient.

[0106] Figure 5 The power variation curve of the double-fed wind turbine using the integrated inertia control strategy for grid frequency regulation is shown in Figure 2. Figure 5 As shown in Figure 2, the process of doubly fed wind turbines participating in grid frequency regulation can be divided into three periods:

[0107] Phase 1: From t0 to t1, the doubly fed wind turbine actively performs system frequency modulation. At this time, the rotor releases kinetic energy and converts it into electrical energy, providing active support for the power grid system.

[0108] The second stage: from t1 to t2, the doubly fed wind turbine exits the system frequency regulation and absorbs electrical energy from the grid and converts it into rotor kinetic energy;

[0109] In the third stage, after time t2, the doubly fed wind turbine returns to MPPT mode operation.

[0110] Specifically, at time t0, the system experiences a power disturbance, causing the system frequency to fluctuate. The doubly fed wind turbine participates in the system frequency modulation, and the shaft kinetic energy released by it is converted into the electromagnetic power added by the wind turbine itself. Due to the continuous fluctuation of the system frequency, the electromagnetic power added by the doubly fed wind turbine will also change continuously, which can be expressed as ΔP W (t) = P W(t)-P W0 , where P W (t) is the electromagnetic power of the wind turbine at time t, P W0 is the initial electromagnetic power of the doubly fed wind turbine. At the same time, the release of the shaft kinetic energy of the doubly fed wind turbine will cause its rotor speed to decrease, that is, its mechanical power P m (t)Continuously decreasing.

[0111] Since the mechanical frequency of the doubly fed wind turbine is continuously decreasing during the frequency regulation of the power grid system, when it reaches the lower limit of the mechanical frequency, it will no longer be able to provide active support for the system frequency. That is, at time t1, the doubly fed wind turbine exits the frequency regulation of the system and enters the rotor speed recovery process. During this process, the doubly fed wind turbine absorbs energy from the power grid. In the embodiment of the present application, the doubly fed wind turbine adopts a constant output electromagnetic power strategy during its speed recovery phase, and its power relationship satisfies the expression: P W (t1) = P m (t1)-ΔP d , where ΔP d It is the difference between the mechanical power and electromagnetic power of the doubly fed wind turbine at the time of exiting frequency regulation (time t1).

[0112] As the speed of the doubly fed wind turbine continues to increase, after the speed of the wind turbine rotor gradually increases to the rated state, the doubly fed wind turbine returns to the maximum power point tracking operation mode at time t2, thus completing a complete system frequency regulation process.

[0113] Figure 6 The figure shows the changes of system frequency in a complete process of doubly fed wind turbines participating in system frequency regulation. Figure 6 As shown in the figure, from t0 to t1, the system frequency is supported by the doubly fed wind turbine, and its frequency first drops and then rebounds, so there is a frequency minimum, which can be called the first frequency drop stage; from t1 to t2, the system frequency drops and rebounds again because the doubly fed wind turbine exits frequency regulation and absorbs power from the grid, which can be called the second frequency drop stage; after t2, the doubly fed wind turbine operates in MPPT mode, and the steady-state frequency deviation value of the system is Δf s , after which the system frequency relies on synchronous units to maintain stability.

[0114] Obviously, although the participation of the doubly fed wind turbine in the grid frequency regulation can provide active support for the system in some periods, it is at the expense of reducing the wind turbine kinetic energy. The process of restoring the speed to return to MPPT will inevitably cause a secondary drop in the system frequency and a system frequency deviation after restoring the steady state. Therefore, it is necessary to reasonably select the control parameters to comprehensively control the maximum deviation Δf of the system frequency drop. max1 , the maximum deviation of the second drop Δf max2and the steady-state frequency deviation value Δf of the system after frequency stabilization s The ideal suppression is performed.

[0115] To this end, in step S2, the system frequency drop characteristics in the complete process of the doubly-fed wind turbine participating in grid frequency modulation are analyzed to determine reasonable control parameters and objective functions, thereby ensuring optimal control of the doubly-fed wind turbine participating in grid frequency modulation.

[0116] Specifically, corresponding to the aforementioned frequency first drop phase, frequency second drop phase, and frequency steady-state phase after MPPT recovery, the frequency drop characteristics analyzed include frequency first drop characteristics, frequency second drop characteristics, and system steady-state frequency characteristics after exiting grid frequency modulation.

[0117] 2.2 Time-domain analytical analysis of frequency first drop characteristics

[0118] For a grid system containing a doubly-fed wind turbine, the original load in the system is set as P L0 At time t0, the load in the system is cut in, so that the load increment in the system is ΔP L , and the system frequency drops. In the power system, synchronous units and doubly-fed wind turbines provide system frequency support to alleviate the fluctuation of the system frequency. In this phase, the state of the synchronous unit in the system can be represented by the following formula:

[0119]

[0120] where P W0 is the initial value of the electromagnetic power output of the doubly-fed wind turbine at time t0, P G0 is the initial value of the electromagnetic power output of the synchronous unit at time t0, ΔP G is the incremental power of the synchronous unit, ΔP W is the incremental power of the doubly-fed wind turbine, ω g is the speed of the synchronous unit, and T J is the inertia time constant of the synchronous unit. When using per-unit calculation, the synchronous unit speed ω g in the above formula can be replaced by the system frequency f, and the initial value of the system frequency f0=f(t0).

[0121] When the doubly-fed wind turbine operates in the MPPT mode, the power of the doubly-fed wind turbine and the synchronous unit in the system satisfies the following formula:

[0122] P G0 +P W0 =P L0 (4);

[0123] The output process of the synchronous unit in the system can be simplified as a first-order inertia link as follows:

[0124]

[0125] where T G is the governor action time constant of the synchronous generator, K G is the equivalent unit regulating power of the synchronous generator.

[0126] The doubly-fed wind turbine adopts typical synthetic inertia control to participate in the dynamic trajectory of system frequency in the power system frequency modulation process. Equation (3) and equation (5) are brought into equation (2), and the following differential equation is obtained in combination with the initial condition:

[0127]

[0128] In the formula, f” and f’ are the second-order derivative and first-order derivative of system frequency f respectively.

[0129] Solving equation (6), the time-domain expression of system frequency in the first frequency drop stage can be obtained:

[0130]

[0131] From equation (7) and letting f'(t) = 0, the time t nadir1 of the lowest point of the first frequency drop can be obtained.

[0132]

[0133] Bringing equation (7) into equation (8) and subtracting the frequency reference value, the maximum frequency deviation Δf max1 of the first frequency drop can be obtained as shown in the following formula:

[0134] Δf max1 = a5-a1exp[a2(θ-a4) / a3]cosθ (9).

[0135] Research shows that the droop coefficient K W1 corresponds to the inertia time constant T J of the synchronous generator, the inertia response coefficient K W2 corresponds to the equivalent unit regulating power K G of the synchronous generator. Therefore, for a system with determined parameters, the droop coefficient K W1 of the wind turbine, the inertia response coefficient K W2 and the initial load change ΔP L will affect the maximum frequency deviation Δf max1 of the first frequency drop.

[0136] 2.3 Time-domain analytical analysis of frequency second drop characteristics

[0137] At time t1, the DFIG exits the system frequency support and absorbs power from the grid to restore the rotor speed. At this time, the electromagnetic power of the DFIG is determined by P W0 +ΔP W becomes P m1 -ΔP d In the speed recovery stage, the wind power output is approximately constant, and the system power shortage P caused by the speed recovery of the doubly fed wind turbine can be defined as u for:

[0138] P u =(P L0 +ΔP L )-P G1 -(P m1 -ΔP d )(10),

[0139] Among them, P m1 is the mechanical power of the doubly fed wind turbine at time t1, P G1 is the electromagnetic power output of the synchronous unit at time t1, and f1 is the system frequency at time t1.

[0140] Ignoring the output change rate of the synchronous unit during the speed recovery process, the differential equation with initial conditions shown in formula (11) can be obtained from the derivation process of the doubly fed wind turbine participating in the frequency regulation stage:

[0141]

[0142] Solving equation (11), we can obtain the time domain expression of the system frequency in the speed recovery stage (i.e., the second frequency drop stage):

[0143]

[0144] From formula (12), we can get the time t when the frequency drops to the lowest point for the second time: nadir2 :

[0145]

[0146] Substituting equation (13) into equation (12) and subtracting it from the frequency standard value, we can obtain the maximum value of the frequency secondary drop deviation Δf as shown in the following equation: max2 :

[0147] Δf max2 =1-f1+b5-b1exp[b2(θ1-b4) / b3]cosθ1(14).

[0148] 2.4 System Steady-State Frequency Analysis after the Doubly Fed Wind Turbine Restores to MPPT Mode

[0149] The magnitude of the secondary drop in system frequency is related to the system frequency value f1 at the time when wind power exits frequency regulation (time t1) and the system power shortage P u After the speed recovery process of the doubly fed wind turbine is completed, the doubly fed wind turbine returns to the MPPT mode. Only the synchronous units in the system will participate in the grid frequency regulation. According to the final value theorem, the system steady-state frequency deviation value Δf s for:

[0150]

[0151] Where L[f(t)] is the Laplace transform of frequency, and s is the Laplace operator.

[0152] According to the above formula, the steady-state frequency deviation value Δf of the power grid system after the doubly fed wind turbine returns to the maximum power point tracking mode is s for:

[0153]

[0154] Among them, P W_max is the maximum output power of the doubly fed wind turbine.

[0155] When the system frequency drops to the maximum, the DFIG outputs the maximum power, which can be determined according to formula (16):

[0156] P W_max =K W1 Δf max2 (17);

[0157] According to equations (16) and (17), the expression of the system steady-state frequency deviation after the system frequency recovers to be stable can be determined as follows:

[0158]

[0159] <3. Select the FM control parameters to be optimized>

[0160] After analyzing the frequency drop characteristics of the doubly-fed wind turbine during the frequency regulation process of the power grid system in step S2, appropriate frequency regulation control parameters for optimization can be determined based on the above analysis results in step S3. Specifically, the frequency regulation control parameters of the doubly-fed wind turbine can be used as independent variables, and simulation can be performed using equations (7) to (9), (12) to (14), and (16) obtained in step S2 to analyze the impact of each frequency regulation control parameter on the frequency response index.

[0161] Figure 7 The maximum value of the system frequency drop deviation Δf obtained by simulation in a specific embodiment is shown. max1 and the droop coefficient K of the doubly fed wind turbine W1 , inertial response coefficient KW2 The three-dimensional relationship of the synchronous unit in this embodiment is the inertia time constant T J =13.8s, speed regulator action time constant T G =5s, synchronous unit equivalent unit adjustment power K G =15, the load disturbance ΔP in the system L =0.1pu; Figure 8 Shows the maximum value of the system frequency secondary drop deviation Δf max2 The system frequency f1 and the power shortage P in the system when the double-fed wind turbine exits frequency regulation u The three-dimensional relationship Figure 9 shows the system steady-state frequency deviation Δf s and the droop coefficient K of the doubly fed wind turbine W1 , inertial response coefficient K W2 The three-dimensional relationship between the three Figure 10 The maximum value of the system frequency change rate RoCoF is shown when the doubly fed wind turbine participates in the system frequency support. max and the droop coefficient K of the doubly fed wind turbine W1 , inertial response coefficient K W2 three-dimensional relationship.

[0162] Table 1 below lists the droop coefficient K of the doubly fed wind turbine in this embodiment. W1 , inertial response coefficient K W2 And the relationship between f1 and the frequency regulation related indicators of the power grid system.

[0163] Table 1K W1 , K W2 The relationship between f1 and frequency modulation index

[0164] FM parameters <![CDATA[Δf max1 ]]> <![CDATA[Δf max2 ]]> <![CDATA[Δf s ]]> Frequency change rate K W1 ]]> negative correlation Small impact negative correlation negative correlation <![CDATA[K W2 ]]> negative correlation Positive correlation Unrelated negative correlation <![CDATA[f1]]> Unrelated negative correlation negative correlation Unrelated

[0165] pass Figures 6 to 10 And Table 1 shows that:

[0166] Doubly fed wind turbine droop coefficient K W1 The larger the value, the smaller the maximum value of the frequency drop deviation. However, it will cause the wind turbine to release more shaft kinetic energy, resulting in an increase in the power shortage of the system when exiting frequency regulation. However, the released shaft kinetic energy is related to the frequency deviation, and considering the difference in energy absorption during the speed recovery process under different wind power penetration rates, it can be concluded from the above that the droop coefficient has little effect on the maximum value of the frequency secondary drop deviation. At the same time, K W1 The larger the K is, the smaller the steady-state frequency deviation will be. After the system frequency stabilizes, the frequency will be closer to the system frequency reference value. However, if the K is too large, W1 This will result in a smaller frequency change rate, which will increase the time consumed by the wind turbine to participate in frequency regulation. Therefore, it is necessary to comprehensively consider the above frequency regulation indicators to determine the wind turbine droop coefficient K.W1 The value of .

[0167] Wind turbine inertia response coefficient K W2 The larger the value, the smaller the maximum value of the frequency drop deviation is, and the slower the system frequency changes. However, the total shaft kinetic energy released in the frequency support stage is large, which will also cause the maximum value of the frequency secondary drop deviation to increase. At the same time, appropriately increase K W2 It helps to improve the overall frequency regulation effect, but too large a value may cause the capacity of the converter to exceed the limit and the frequency change rate to slow down. Therefore, it is necessary to comprehensively consider the above indicators to determine the appropriate wind turbine inertia response coefficient K. W2 The value of

[0168] In order to analyze the parameter ΔP d For Δf max2 and dω r The influence of / dt can be combined with equation (10) and equation (12) to obtain the following equation:

[0169]

[0170] From the above formula, we know that ΔP d The larger the Δf max2 The larger the value, the more ΔP d Smaller, yet smaller ΔP d This will make the difference between the electromagnetic power and mechanical power output of the wind turbine smaller, the rotor speed recovery will be slower, and the time required for the wind turbine to return to MPPT mode will be longer. Therefore, selecting a suitable ΔP d It is crucial.

[0171] Through the above analysis, we can find that K W1 , K W2 , f1 (or t1), ΔP d In the process of doubly fed wind turbines participating in system frequency regulation, they affect the frequency regulation index in different degrees and directions. Therefore, in the embodiment of this application, K W1 , K W2 , t1, ΔP d As the frequency regulation parameter of the doubly fed wind turbine that needs to be optimized.

[0172] <4. Frequency Modulation Parameter Optimization Based on Improved Particle Swarm Optimization>

[0173] 4.1 Determine the objective function and constraints for frequency modulation parameter optimization

[0174] Through the above analysis, the goal of the doubly fed wind turbine in participating in the system frequency support is to improve the doubly fed wind turbine's ability to support the system frequency regulation while minimizing the impact of the secondary frequency drop. The frequency regulation parameter setting of the wind turbine can be regarded as an optimization problem. The optimization goal of this optimization problem is to maximize the lowest points of the primary and secondary frequency drops. That is, the objective function of the frequency regulation parameter optimization is shown in formula (20):

[0175] min{max{Δf max1 ,Δf max2}}(20).

[0176] In some embodiments, the constraints for optimizing the frequency modulation parameters are shown in formula (21):

[0177]

[0178] Among them, H W is the doubly fed wind turbine rotor inertia time constant, ω r (t0) is the initial speed of the doubly fed wind turbine, ω min is the lower speed limit of the doubly fed wind turbine, K W1,max K W1 The upper limit of K W2,max K W2 The upper limit of the value.

[0179] The setting of constraints also has a direct impact on the results of FM parameter optimization. For example, in some existing parameter optimization methods, K W1,max , K W2,max It is often selected as a constant based on empirical values. However, as the operating state of the wind turbine changes, its rotor kinetic energy and output power change in real time and are coupled with each other, resulting in a dynamic change in the parameter optimization space. Unreasonable constraints will not only increase the size of the parameter optimization space, but also significantly increase the time for parameter optimization. In addition, if effective constraints are not imposed in combination with the actual operating state of the doubly fed wind turbine, the frequency regulation parameters obtained by optimization will exceed the performance limit of the doubly fed wind turbine, affecting the operational safety of the wind turbine.

[0180] Therefore, in the embodiments of the present application, K W1,max , K W2,max Both are coupling items between the rotor safety and power output safety of the doubly fed wind turbine, and need to be determined in combination with the actual operating state of the doubly fed wind turbine. Specifically, in some embodiments, K W1,max , K W2,max By introducing the kinetic energy calculation factor k from the two aspects of rotor safety and power output safety m and power calculation factor k c To determine, the specific steps include:

[0181] First, k m Expressed as (22):

[0182]

[0183] k m It is used to characterize the proportion of the rotor's available kinetic energy to its maximum available kinetic energy at the current speed of the doubly fed wind turbine. (22) Where E, E max 、E min They represent the actual value, maximum value and minimum value of the rotor kinetic energy of the doubly fed wind turbine respectively.

[0184] Taking into account Formula (22) can be written as formula (23):

[0185]

[0186] Among them, ω r 、ω r_max and ω r_min They represent the actual value, upper limit and lower limit of the rotor speed of the doubly fed wind turbine respectively. The superscript * represents the per-unit value, and its reference value is the synchronous speed ω of the doubly fed wind turbine. rn ,Thus, the primary frequency regulation capability of the doubly-fed wind turbine was quantified from the perspective of wind turbine speed safety.

[0187] Then, k c Expressed as (24):

[0188]

[0189] k c It is used to characterize the ratio of the current remaining capacity of the converter to the maximum active output within the power constraint range of the doubly fed wind turbine at the current speed. (24) In which, P w 、P w_max 、P w_min They represent the actual value, maximum value and minimum value of the active power output of the doubly fed wind turbine respectively. The active power output of the doubly fed wind turbine can be further expressed as the sum of the overspeed load reduction control power of the doubly fed wind turbine and the incremental power it generates for frequency regulation:

[0190]

[0191] (25) where k opt Represents the overspeed load reduction curve coefficient, ΔP e To obtain the frequency modulation active power increment, substitute (25) into (24) and use the rated capacity P of the unit as rated and ω rn Standardize it as the reference value and get:

[0192]

[0193] Equation (26) quantifies the primary frequency regulation capability of the doubly fed wind turbine converter from the perspective of capacity security.

[0194] Furthermore, considering k m With k c , the speed and electromagnetic power of the doubly fed wind turbine are reversely coupled during the frequency regulation process. To ensure that it participates in frequency regulation within a safe range, the smaller value of the two is taken to represent the actual frequency regulation capability of the unit, that is,

[0195] ε=min{k m ,k c},0≤ε≤1(27),

[0196] Among them, ε is the FM participation coefficient, Figure 11 The ω value of the double-fed wind turbine participating in system frequency regulation at different speeds is shown. r -ε curve. Figure 11 As shown, when there is no load interference in the power system, k m With k c Intersecting at point A0, the curve of ε is as follows Figure 11 As shown by the black dotted line; when the frequency changes, k c Curve according to ΔP e The size and sign of k m The curve remains unchanged. When the speed of the double-fed wind turbine is high, k c The curve determines the size of the ε value. The doubly fed wind turbine converter is under a heavy load at this time. The frequency modulation depth of the wind turbine should be reduced as much as possible to avoid a sudden increase in active power and impact on the converter. At low speed, k m The curve determines the size of ε, which will continue to decrease as the fan speed decreases. r When approaching the lower limit, ε approaches zero, causing the unit to no longer participate in the frequency regulation response. Therefore, by adjusting the droop coefficient and inertia response coefficient of the doubly fed wind turbine through ε, the risk of exceeding the unit speed and power limits can be greatly reduced.

[0197] It should be noted that ε is the frequency regulation participation coefficient. When the speed of the doubly fed wind turbine reaches Figure 11 ω in r_maxWhen the wind turbine speed reaches its maximum setting, continued participation in frequency regulation will cause the converter's power to exceed its limit, but ε does not control the unit to exit frequency regulation in a timely manner. Furthermore, the maximum value, A0, in the ε curve is relatively small, failing to fully utilize the wind power's frequency regulation capabilities. Therefore, ε needs to be further optimized: when the unit speed is too low or too high, the wind turbine's frequency regulation capability is weak, and its participation in frequency regulation should be minimized. At moderate speeds, the unit's frequency regulation participation should be maintained at a high value to fully utilize the rotor's kinetic energy and improve the system's frequency stability, which is consistent with the curve characteristics of the S-shaped function.

[0198] Specifically, the LOGISTIC function can be used to optimize it, which is expressed as:

[0199]

[0200] Where K0, K1, K2 and n are all constants.

[0201] Obviously, the K obtained by (28) W1,max and K W2,max It can not only reflect the coupling characteristics between the real-time changes in speed and power output of the doubly fed wind turbine, but also perform appropriate optimization based on a narrow limit range, using it as the constraint range for the droop parameters and inertia response coefficient, which can take into account the wind turbine operation safety and frequency regulation capability, thereby obtaining a more ideal optimization result.

[0202] 4.2 Frequency modulation parameter optimization based on improved particle swarm algorithm

[0203] The core idea of ​​the particle swarm optimization (PSO) algorithm for parameter optimization is to regard the solution space of the problem as a multidimensional superbody and search for the global optimal solution in it. At the same time, particles adjust their states through information exchange and cooperation to move towards the optimal solution. Each particle represents a potential optimal solution to the optimization problem and is characterized by a position vector and a velocity vector. The position vector represents a possible solution to the problem (i.e., a set of K W1 , K W2 , t1, ΔP d The velocity vector determines the direction and distance that the particle moves.

[0204] In this application, an improved particle swarm algorithm is used to perform K W1 , K W2 , t1, ΔP d For optimization, preferably, the improved particle swarm algorithm can be performed by the following steps:

[0205] Set the total number of particles to N, and use X nDenote the position vector of the nth particle, V n Represents its velocity vector, then the position and velocity of the nth particle in the space of dimension m can be expressed as the following formula:

[0206]

[0207] Obviously, when the frequency modulation parameter to be optimized is K W1 , K W2 , t1, ΔP d When m=4, the X of each particle n That is, a set of optional K W1 , K W2 , t1, ΔP d Get the value.

[0208] According to the position vector of each particle, its objective function value is calculated by formula (20) to determine the individual optimal position and the group optimal position of each iterative search. Each particle in the space dynamically adjusts its position vector and velocity vector under the constraint of formula (21) by tracking the previous individual optimal position and group optimal position.

[0209] Specifically, using P n Indicates the individual optimal position of the nth particle, with P f represents the best position of the total population in the previous iteration. The two best positions can be expressed as:

[0210]

[0211] The update formula of the velocity vector and position vector of each particle in space can be expressed as:

[0212]

[0213] Among them, c1 and c2 are learning factors, r1 and r2 are random numbers uniformly distributed in [0,1], w is the inertia weight factor, and k is the current number of iterations.

[0214] The traditional PSO algorithm uses a fixed inertia weight factor and a learning factor. A larger inertia weight factor will prevent the algorithm from falling into the local optimum in the early stage of iteration, which is beneficial to the global search of particles. A smaller inertia weight factor will make the algorithm converge faster in the later stage of iteration, which is convenient for the local search of particles. For the learning factor, a larger c1 and a smaller c2 will be beneficial to the global search of particles in the early stage of iteration, and a smaller c1 and a larger c2 will be beneficial to the local search of particles in the later stage of iteration, thereby making the algorithm converge quickly. For this reason, the improved particle swarm algorithm of this application performs adaptive dynamic adjustment on the inertia weight factor and the learning factor during the iterative optimization process, that is, the factor adaptive mutation process, and its mutation equation can be expressed as:

[0215]

[0216] Among them, w d 、w u is the initial and final value of the inertia weight factor, c 1d 、c 1u is the initial value and final value of the learning factor c1, c 2d 、c 2u is the initial value and final value of the learning factor c2, k max is the total number of iterations.

[0217] <Specific Example 1>

[0218] This embodiment uses Figure 12 The 3-machine 9-node power system model with double-fed wind turbines shown in the figure is simulated, and the frequency regulation parameters of the double-fed wind turbines in the system are optimized by the wind power active support frequency regulation parameter optimization method for suppressing frequency drop provided in this application to verify its effectiveness. Among them, the system rated frequency is 50Hz; SG1 and SG2 are synchronous units, simulating thermal power units, with a rated capacity of 100MW, and an inertia time constant T J =13.8s, the synchronous machine speed regulator adjustment coefficient is 0.015, and the speed regulator action time constant is set to 6.0s; the wind farm contains a certain number of doubly fed wind turbines, each with a rated capacity of 1.5MW, the rated wind speed of the wind turbine is 10m / s, the initial wind speed of the system is set to 9m / s (0.9pu), and the equivalent time constant H W = 5.04s; Load 1 is set to 117MW, with a 20MW parallel load shedding capability; Load 2 is set to 95MW; Load 3 is set to 35MW; and Load 4 is set to 2.5MW. Table 3-2 shows the line lengths and transformer ratios for the system.

[0219] At t0=10s, the system cuts in a load of 20MW at load 1, and the total system load increases, causing the system frequency to drop. The wind turbine participates in frequency support through comprehensive inertia control, improving the frequency stability of the power system.

[0220] Under the condition of 50 wind turbines and a wind power penetration rate of 27.27%, the initial setting of the wind turbine exit frequency regulation time is t1 = 17s. Figure 13 、 Figure 14 K W2 =10, different K W1 Corresponding system frequency response curve and doubly-fed wind turbine speed curve; Figure 15 、 Figure 16 K W1 =10, different K W2Corresponding system frequency response curve and doubly-fed wind turbine speed curve; Figure 17 Shows K W1 =20,K W2 =30, the system frequency response curve corresponding to different t1.

[0221] From the above figures, we can see that in the inertial response coefficient K W2 When the droop coefficient K is large, W1 It can slow down the primary frequency drop and the secondary frequency drop at the same time, but the excessive droop coefficient K W1 It will also cause the wind turbine to release too much rotor kinetic energy during the frequency modulation process, which is not conducive to the speed recovery process of the wind turbine; W1 When a certain value is given, the larger inertia response coefficient K W2 It can effectively slow down the primary frequency drop, but it will also aggravate the secondary frequency drop, making the lowest point of the secondary frequency drop lower than the primary drop, and an excessively large inertia response coefficient will still cause a sharp increase in the rotor kinetic energy released by the wind turbine; when the exit frequency modulation time is too short, the secondary frequency drop will be aggravated. This is because if the exit time is too short, the lower the frequency at the time of exiting frequency modulation, the more serious the secondary frequency drop; at the same time, if the exit frequency modulation time is too long, the more rotor kinetic energy released by the wind turbine will be, which will increase the system power shortage accordingly, and the secondary frequency drop will also be aggravated.

[0222] Taking the system frequency characteristics into consideration, the droop coefficient K W1 The search range is set to [15,40], and the inertial response coefficient K W2 The search range of is set to [15,30]; the search range of the exit frequency modulation time is set to [14s,18s], and the improved particle swarm algorithm described above is used for particle swarm optimization. Figure 18 The algorithm convergence curve when the wind power penetration rate is 27.27% is shown. Table 2 below lists the optimization results of the frequency regulation parameters.

[0223] Table 2 Optimization results of improved particle swarm algorithm when wind power penetration rate is 27.27%

[0224] parameter Optimal location <![CDATA[t1 / s]]> <![CDATA[K W1 ]]> <![CDATA[K W2 ]]> Optimization results 0.1062 15.5 38 28

[0225] Figures 19 to 21 The system frequency response curve, doubly fed wind turbine speed curve and doubly fed wind turbine electromagnetic power curve when the above optimal frequency modulation parameter combination is adopted are shown respectively, and the comparison with other optional frequency modulation parameter combinations is shown. It can be seen that the optimal droop coefficient K obtained by the method provided by this application is W1 , inertial response coefficient K W2and exit frequency modulation time t1, which can better solve the problem of system frequency drop once, and at the same time ensure that the lowest point of the second frequency drop is higher than the lowest point of the first drop. By comparing the simulation results of this group of parameters with the simulation results of its edge parameters, the optimal characteristics of the optimization results of the present application method can be seen. For example, Figure 20 It can be seen that the frequency modulation parameters obtained by the method of this application can ensure that the wind turbine speed is within the appropriate range (0.7pu~1.1pu); Figure 21 It can be seen that the frequency modulation parameters obtained by the method of this application can ensure that the fan has sufficient frequency modulation capability to support the stability of the system frequency.

[0226] The above is a detailed introduction to the specific implementation methods of the present application. For those skilled in the art, several improvements and modifications can be made to the present application without departing from the principles of the present application. These improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A method for optimizing wind power active support frequency modulation parameters for suppressing frequency drops, characterized in that: The following steps are involved: S1, establish a system model of doubly fed wind turbine grid-connected power generation; S2, analyzing the frequency drop characteristics of the doubly-fed wind turbine when participating in grid frequency regulation, wherein the doubly-fed wind turbine participates in grid frequency regulation through a comprehensive inertia control strategy, and the frequency drop characteristics include a primary frequency drop characteristic, a secondary frequency drop characteristic, and a system steady-state frequency characteristic after the doubly-fed wind turbine exits grid frequency regulation; S3, based on the frequency drop characteristics of the double-fed wind turbine when participating in the grid frequency regulation, select the frequency regulation parameters that need to be optimized, the frequency regulation parameters that need to be optimized include K W1 , K W2 , t1 and ΔP d , where K W1 is the droop coefficient, K W2 is the inertial response coefficient, ΔP d is the difference between the mechanical power and electromagnetic power of the doubly fed wind turbine at the time t1 when it exits the grid frequency regulation; S4, using an improved particle swarm algorithm to search for an optimal value of the frequency modulation parameter, wherein the constraint conditions of the improved particle swarm algorithm include a coupling term between rotor safety and power output safety of the doubly fed wind turbine; The coupling item between rotor safety and power output safety of the doubly fed wind turbine includes K W1,max , K W2,max , K W1,max K W1 The upper limit of K W2,max K W2 The upper limit of K is determined by the following steps: W1,max With K W2,max : The kinetic energy calculation factor k of the doubly fed wind turbine at the current speed is determined based on the following formula: m : Among them, ω * r 、ω * r_max and ω * r_min are the actual values ​​of the rotor speed of the doubly fed wind turbine ω r , upper limit value ω r_max and the lower limit ω r_min The per-unit value, the reference value is the synchronous speed ω of the doubly fed wind turbine rn ; The power calculation factor k of the doubly fed wind turbine at the current speed is determined based on the following formula: c : Among them, k opt is the overspeed load reduction curve coefficient, Its per-unit value, ΔP e is the frequency modulation active power increment, is its per-unit value, K is the coefficient of frequency modulation active power increment, P w 、P rated are the actual value and rated value of the active output of the doubly fed wind turbine, The per-unit value is the lower limit of the active power output of the doubly-fed wind turbine; K is determined based on the following formula W1,max With K W2,max : Where ε is the FM participation coefficient, and K0, K1, K2 and n are all constants.

2. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 1, characterized in that: The system model of the doubly-fed wind turbine grid-connected power generation includes a doubly-fed wind turbine, a rotor-side converter, a grid-side converter and a phase-locked loop; The rotor-side converter and the grid-side converter are connected back-to-back via DC bus capacitors; The doubly-fed wind turbine converts captured wind energy into electrical energy, with its rotor side being connected to the grid via a back-to-back rotor-side converter and a grid-side converter, and its stator side being directly connected to the grid; The phase-locked loop is used to synchronize the frequency of the doubly-fed wind turbine with the frequency of the power grid system.

3. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 1, characterized in that: The doubly-fed wind turbine participates in grid frequency regulation through a comprehensive inertia control strategy. Specifically, the doubly-fed wind turbine provides an electromagnetic power increment to the grid based on the following formula: Where ΔP W is the electromagnetic power increment provided by the doubly fed wind turbine to the grid, Δf is the difference between the grid system frequency f and the system frequency reference value f ref The frequency deviation between .

4. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 3 is characterized in that: The frequency primary drop characteristics include the time domain expression of the system frequency of the power grid during the frequency primary drop stage, the lowest point of the system frequency primary drop, and the maximum value of the system frequency primary drop deviation; The time domain expression of the system frequency of the power grid during the first frequency drop phase is as follows: Where ΔP L is the load increment in the power grid system, T J is the inertia time constant of the synchronous unit, T G K is the time constant of the speed regulator of the synchronous unit, G Adjust power for equivalent units of synchronous units; The moment when the system frequency drops to the lowest point is determined by the following formula: Among them, t nadir1 is the moment when the system frequency drops to the lowest point, and t0 is the moment when the doubly fed wind turbine starts to participate in the grid frequency regulation; The maximum value of the system frequency drop deviation is determined by the following formula: Δf max1 =a5-a1 exp[a2(θ-a4) / a3]cosθ, Where Δf max1 It is the maximum deviation of the system frequency when it drops once.

5. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 4, characterized in that: The frequency secondary drop characteristics include the time domain expression of the system frequency of the power grid during the frequency secondary drop stage, the time of the lowest point of the frequency secondary drop, and the maximum value of the frequency secondary drop deviation; The time domain expression of the system frequency of the power grid during the secondary frequency drop stage is as follows: Among them, f1 is the system frequency at the moment t1 when the double-fed wind turbine exits the grid frequency regulation, P u The system power shortage caused by the speed recovery of the doubly fed wind turbine; The moment when the frequency drops to the lowest point for the second time is determined by the following formula: Among them, t nadir2 is the moment when the frequency drops to the lowest point for the second time; The maximum value of the secondary frequency drop deviation is determined by the following formula: Δf max2 =1-f1+b5-b1 exp[b2(θ1-b4) / b3]cosθ1, Where Δf max2 It is the maximum value of the secondary frequency drop deviation.

6. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 5, characterized in that: The system steady-state frequency characteristics after the doubly-fed wind turbine exits the grid frequency regulation are determined by the following formula: Where Δf s is the steady-state frequency deviation value of the power grid system after the doubly fed wind turbine returns to the maximum power point tracking mode, P W0 is the initial value of the electromagnetic power output of the doubly fed wind turbine at time t0.

7. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 6, characterized in that: The objective function of the improved particle swarm optimization algorithm is: min{max{Δf max1 ,Δf max2 }}; The constraints of the improved particle swarm optimization algorithm include: Among them, H W is the doubly fed wind turbine rotor inertia time constant, ω r (t0) is the initial speed of the doubly fed wind turbine, ω min is the lower speed limit of the doubly fed wind turbine, P W (t), P m (t) are the electromagnetic power and mechanical power of the doubly fed wind turbine at time t.

8. The wind power active support frequency modulation parameter optimization method for suppressing frequency drop according to claim 6, characterized in that: The improved particle swarm algorithm uses the following formula to adaptively update the inertia weight factor in the particle swarm algorithm: Among them, w is the inertia weight factor, w d 、w u is the initial value and the final value of w, k is the current number of iterations, k max is the total number of iterations; The improved particle swarm algorithm uses the following formula to adaptively update the learning factor in the particle swarm algorithm: Among them, c1 and c2 are learning factors, c 1d 、c 1u is the initial value and final value of c1, c 2d 、c 2u are the initial and final values ​​of c2.

Images