A method for identifying the frequency controller of a full-power wind turbine

Abstract

A method for identifying a frequency controller of a full-power wind turbine generator set relates to the field of power control technology. Aiming at the problem of inaccurate identification results in the prior art, and aiming at electromagnetic transient modeling of full-power wind turbine generator sets, this application proposes a practical frequency controller identification method that uses the unit's measured frequency response characteristic curve morphology to identify the frequency controller structure, and uses the unit's measured frequency response characteristic curve eigenvalues ​​to identify the frequency controller parameters. This application first identifies the frequency controller structure, and identifies the frequency controller parameters based on the frequency controller structure, avoiding the problem of inaccurate identification results caused by using a fixed frequency controller structure for identification in the prior art. The identification results of this application are highly accurate, providing technical support for improving the safe operation level of the power grid and the wind power acceptance capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power control, and in particular to an identification method for a full-power wind turbine frequency modulation controller. Background Art

[0002] The frequency modulation controller is a circuit device used to generate and control frequency modulation signals, and is mainly used in electronic communications, audio conversion and other fields. Its basic principle is to use capacitors and inductors to form an oscillation circuit, and to control the oscillation frequency by changing the values ​​of capacitors and inductors, thereby generating a frequency modulation signal. There are many specific implementation methods, such as VCO (Voltage Controlled Oscillator) circuit and PLL (Phase Locked Loop) circuit.

[0003] In recent years, with the increase in the proportion of new energy and the widespread application of high-voltage direct current transmission systems, power electronic equipment has been connected to the power grid on a large scale, and the structure and operating characteristics of new power systems have changed significantly. However, the identification of frequency controllers is the premise and basis for analyzing the frequency safety and stability of power systems with a high proportion of new energy. The existing identification method of frequency controllers is to perform identification under a fixed frequency controller structure. However, the actual situation is that the structure of the frequency controller is not completely fixed. Therefore, the existing technology leads to the problem of inaccurate identification results. Summary of the invention

[0004] The purpose of the present invention is to propose an identification method for a full-power wind turbine frequency controller in view of the problem of inaccurate identification results in the prior art.

[0005] The technical solution adopted by the present invention to solve the above technical problems is:

[0006] A method for identifying a full-power wind turbine frequency controller comprises the following steps:

[0007] Step 1: Obtain the steady-state active power of the wind turbine in response to the frequency change under the condition of large frequency step disturbance, and select the lowest steady-state active power among the steady-state active power of the wind turbine in response to the frequency change, and then use the lowest steady-state active power as the active threshold P th , when the wind power active power is greater than the active power threshold P th When , execute step 2, the frequency large step disturbance condition is |Δf|≥1Hz, where Δf is the frequency change;

[0008] Step 2: Identify the frequency modulation controller, wherein the identification of the frequency modulation controller includes structural identification of the frequency modulation controller and parameter identification of the frequency modulation controller;

[0009] The specific steps of structural identification of the frequency controller are:

[0010] Step 21: by turning on / off one or more frequency control modes, obtain corresponding frequency and active power data, obtain corresponding frequency response curves according to the frequency and active power data, and construct a typical frequency response curve library using the frequency response curves;

[0011] Step 22: Obtain the frequency and active power data of the wind turbine to be identified, and obtain the curve to be identified based on the frequency and active power data of the wind turbine to be identified, and calculate the Euclidean distance between the curve to be identified and each frequency response characteristic curve in the typical frequency response curve library, select the frequency response characteristic curve with the highest similarity, and obtain the controller structure identification result based on the frequency response characteristic curve with the highest similarity, the controller structure identification result, that is, one or more frequency control modes corresponding to the curve, and its on / off state;

[0012] The specific steps of parameter identification of the frequency modulation controller are:

[0013] Step 1: Use the wind turbine active power of 0.2 pu as the dividing point between low wind speed and medium wind speed, and use the upper limit wind speed of load shedding as the dividing point between medium wind speed and high wind speed;

[0014] Step 2: Obtain the active power output of the wind turbine during steady-state operation at the highest wind speed, and identify the load reduction reserve coefficient η based on the active power output of the wind turbine during steady-state operation at the highest wind speed. del ;

[0015] Step 3: Obtain the active power of the wind turbine under the condition of large frequency step disturbance between medium and high wind speeds, and identify the upper limit of active power regulation dP according to the active power of the wind turbine under the condition of large frequency step change between medium and high wind speeds. max and active power regulation lower limit dP min ;

[0016] Step 4: Obtain the highest frequency f of the wind turbine when the wind speed is between medium and high and the frequency is subject to a large step disturbance. db1 The minimum frequency f of the sum frequency change db2 , where the highest frequency of frequency change is f db1 , that is, the frequency drop condition, the lowest frequency f of the frequency change db2 , that is, the frequency rising condition;

[0017] Step 5: Divide the curve to be identified into five intervals, namely the steady-state interval t0-t1 before the disturbance, the transient interval t1-t2 when the disturbance occurs, the transient steady-state interval t2-t3 when the disturbance is stable, the transient interval t3-t4 when the disturbance is recovered, and the steady-state interval t4-t5 after the disturbance.

[0018] Among them, t0 represents the test start time, t1 represents the disturbance start time, t2 represents the time when the active power reaches stability after the disturbance, t3 represents the time when the frequency begins to recover, t4 represents the time when the active power stabilizes again, and t5 represents the test end time.

[0019] Obtain the average active power of the wind turbine set within t2-t3 under the condition of medium to low wind speed and small frequency step disturbance, and combine it with f db1 and f db2 , and then identify the droop coefficient K 2u and K 2d ,

[0020] Among them, the frequency small step change is 0.1Hz<|Δf|≤0.2Hz;

[0021] Step 6: Obtain the maximum frequency change rate rf of the wind turbine when the frequency changes ramp between medium and high wind speeds db1 The minimum frequency change rate rf db2 , where the maximum frequency change rate rf db1 , that is, the frequency drop condition, the minimum frequency change rate rf db2 , that is, the frequency rising condition;

[0022] Step 7: Obtain the average active power of the wind turbine during t1-t2 under the conditions of medium to low wind speed, medium frequency slope change, and small frequency step disturbance, and combine it with f db1 、f db2 、rf db1 and rf db2 , and then identify the virtual inertia coefficient K 1u and K 1d ;

[0023] Among them, the medium frequency ramp change is 0.3Hz / s<|df / dt|≤0.5Hz / s, df / dt is the frequency ramp change rate;

[0024] Step 8: η obtained from steps 1 to 7 del 、dP max 、dP min 、f db1 、f db2 , K 2u , K 2d 、、rf db1 、rf db2 , K 1u and K 1d , which are the identified parameters of the frequency modulation controller.

[0025] Furthermore, the frequency control method includes: virtual inertia control, droop control, overspeed load reduction control and pitch angle load reduction control.

[0026] Furthermore, the specific steps of step 21 are:

[0027] Step 2: Let virtual inertia control, droop control, overspeed load reduction control and pitch angle load reduction control correspond to flags Flag 1, Flag 2, Flag 3 and Flag 4 respectively. Then, group flags Flag 1, Flag 2, Flag 3 and Flag 4, specifically:

[0028] a: Flag1 = 0, Flag2 = 0;

[0029] b: Flag1 = 1, Flag2 = 0;

[0030] c: Flag1 = 0, Flag2 = 1;

[0031] d: Flag1 = 1, Flag2 = 1;

[0032] e: Flag3 = 0;

[0033] f: Flag3 = 1;

[0034] g: Flag4 = 0;

[0035] h: Flag4 = 1;

[0036] Among them, the flag bit is 0, which means turning off the control, and the flag bit is 1, which means turning on the control;

[0037] Step 212: According to the grouping in step 211, the frequency and active power of the high-voltage side of the wind turbine box transformer under different groups are obtained, and then the frequency response characteristic curve is obtained according to the frequency and active power, and then a typical frequency response curve library is constructed.

[0038] Furthermore, the load reduction reserve coefficient η del It is expressed as:

[0039]

[0040] Among them, P 0max P is the active output of the wind turbine during steady-state operation, i.e. the maximum output after load shedding and standby. n is the rated active power of the wind turbine.

[0041] Furthermore, the active power regulation upper limit dP max and active power regulation lower limit dP min It is expressed as:

[0042]

[0043] Among them, P 1max and P 1min They are the maximum and minimum active powers of the wind turbine during the frequency disturbance, and P0 is the steady-state active output of the wind turbine.

[0044] Furthermore, the maximum frequency f of the frequency change db1 The minimum frequency f of the sum frequency change db2 It is expressed as:

[0045]

[0046] Among them, f db1 * and f db2 * are the highest frequency and the lowest frequency of the response frequency change of the wind turbine under the conditions of medium to high wind speed and large frequency step disturbance, respectively. n is the rated frequency.

[0047] Furthermore, the droop coefficient K 2u and K 2d It is expressed as:

[0048]

[0049] ΔP2=P 23 -P0

[0050] Among them, ΔP2 is the active reference value adjustment value output by the droop controller, K 2u With K 2d are the upward and downward droop coefficients, P 23 It is the average active power of the wind turbine during the period from t2 to t3.

[0051] Furthermore, the virtual inertia coefficient K 1u and K 1d It is expressed as:

[0052]

[0053] ΔP1=P 12 -P0-ΔP2

[0054] Among them, ΔP1 is the active regulation output by the virtual inertia controller, P 12 It is the average active power of the wind turbine during the period from t1 to t2.

[0055] Furthermore, the wind speed scenario of the wind turbine to be identified is taken at intervals of 0.5 m / s.

[0056] Furthermore, the frequency disturbance of the wind turbine to be identified is obtained based on the inertia response and frequency modulation response test conditions specified in GB / T 36994-2018, and the frequency change rate range is ±0.1 Hz / s to ±1 Hz / s, with an interval of 0.05 Hz / s; the frequency disturbance setting value is 48 Hz to 51.5 Hz, with an interval of 0.1 Hz, and a dead zone interval of 0.01 Hz.

[0057] The beneficial effects of the present invention are:

[0058] For electromagnetic transient modeling of full-power wind turbines, this application proposes a practical frequency modulation controller identification method that uses the measured frequency response characteristic curve morphology of the unit to identify the frequency modulation controller structure and uses the characteristic value of the measured frequency response characteristic curve of the unit to identify the frequency modulation controller parameters. This application first identifies the frequency modulation controller structure and identifies the frequency modulation controller parameters based on the frequency modulation controller structure, avoiding the problem of inaccurate identification results caused by the use of a fixed frequency modulation controller structure for identification in the prior art. The identification results of this application are highly accurate, providing technical support for improving the safe operation level of the power grid and the wind power acceptance capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a general model structure diagram of a full-power wind turbine;

[0060] Figure 2 This is a schematic diagram of the load shedding standby mode of the wind turbine generator system;

[0061] Figure 3 It is a schematic diagram of the frequency modulation controller structure and its corresponding curve shape. DETAILED DESCRIPTION

[0062] It should be particularly noted that, in the absence of conflict, the various embodiments disclosed in this application can be combined with each other.

[0063] Specific implementation method 1: refer to Figure 1 Specifically describing this embodiment, a method for identifying a full-power wind turbine frequency controller described in this embodiment is characterized by comprising the following steps:

[0064] Step 1: Obtain the steady-state active power of the wind turbine in response to the frequency change under the condition of large frequency step disturbance, and select the lowest steady-state active power among the steady-state active power of the wind turbine in response to the frequency change, and then use the lowest steady-state active power as the active threshold P th , when the wind power active power is greater than the active power threshold P th When , execute step 2, the frequency large step disturbance condition is |Δf|≥1Hz, where Δf is the frequency change;

[0065] Step 2: Identify the frequency modulation controller, wherein the identification of the frequency modulation controller includes structural identification of the frequency modulation controller and parameter identification of the frequency modulation controller;

[0066] The specific steps of structural identification of the frequency controller are:

[0067] Step 21: by turning on / off one or more frequency control modes, obtain corresponding frequency and active power data, obtain corresponding frequency response curves according to the frequency and active power data, and construct a typical frequency response curve library using the frequency response curves;

[0068] Step 22: Obtain the frequency and active power data of the wind turbine to be identified, and obtain the curve to be identified based on the frequency and active power data of the wind turbine to be identified, and calculate the Euclidean distance between the curve to be identified and each frequency response characteristic curve in the typical frequency response curve library, select the frequency response characteristic curve with the highest similarity, and obtain the controller structure identification result based on the frequency response characteristic curve with the highest similarity, the controller structure identification result, that is, one or more frequency control modes corresponding to the curve, and its on / off state;

[0069] The specific steps of parameter identification of the frequency modulation controller are:

[0070] Step 1: Use the wind turbine active power of 0.2 pu as the dividing point between low wind speed and medium wind speed, and use the upper limit wind speed of load shedding as the dividing point between medium wind speed and high wind speed;

[0071] Step 2: Obtain the active power output of the wind turbine during steady-state operation at the highest wind speed, and identify the load reduction reserve coefficient η based on the active power output of the wind turbine during steady-state operation at the highest wind speed. del ;

[0072] Step 3: Obtain the active power of the wind turbine under the condition of large frequency step disturbance between medium and high wind speeds, and identify the upper limit of active power regulation dP according to the active power of the wind turbine under the condition of large frequency step change between medium and high wind speeds. max and active power regulation lower limit dP min ;

[0073] Step 4: Obtain the highest frequency f of the wind turbine when the wind speed is between medium and high and the frequency is subject to a large step disturbance. db1 The minimum frequency f of the sum frequency change db2 , where the highest frequency of frequency change is f db1 , that is, the frequency drop condition, the lowest frequency f of the frequency change db2 , that is, the frequency rising condition;

[0074] Step 5: Divide the curve to be identified into five intervals, namely the steady-state interval t0-t1 before the disturbance, the transient interval t1-t2 when the disturbance occurs, the transient steady-state interval t2-t3 when the disturbance is stable, the transient interval t3-t4 when the disturbance is recovered, and the steady-state interval t4-t5 after the disturbance.

[0075] Among them, t0 represents the test start time, t1 represents the disturbance start time, t2 represents the time when the active power reaches stability after the disturbance, t3 represents the time when the frequency begins to recover, t4 represents the time when the active power stabilizes again, and t5 represents the test end time.

[0076] Obtain the average active power of the wind turbine set within t2-t3 under the condition of medium to low wind speed and small frequency step disturbance, and combine it with f db1 and f db2 , and then identify the droop coefficient K 2u and K 2d ,

[0077] Among them, the frequency small step change is 0.1Hz<|Δf|≤0.2Hz;

[0078] Step 6: Obtain the maximum frequency change rate rf of the wind turbine when the frequency changes ramp between medium and high wind speeds db1 The minimum frequency change rate rf db2 , where the maximum frequency change rate rf db1 , that is, the frequency drop condition, the minimum frequency change rate rf db2 , that is, the frequency rising condition;

[0079] Step 7: Obtain the average active power of the wind turbine during t1-t2 under the conditions of medium to low wind speed, medium frequency slope change, and small frequency step disturbance, and combine it with f db1 、f db2 、rf db1 and rf db2 , and then identify the virtual inertia coefficient K 1u and K 1d ;

[0080] Among them, the medium frequency ramp change is 0.3Hz / s<|df / dt|≤0.5Hz / s, df / dt is the frequency ramp change rate;

[0081] Step 8: η obtained from steps 1 to 7 del , dP max , dP min 、f db1 、f db2 , K 2u , K2d 、、rf db1 、rf db2 , K 1u and K 1d , which are the identified parameters of the frequency modulation controller.

[0082] Obtain the frequency and active power data of the wind turbine to be identified under different wind speed scenarios and different frequency disturbances. The wind speed scenario is recommended to be taken at intervals of 0.5m / s, covering the entire operating range from cut-in to cut-out wind speed. The frequency disturbance is expanded based on the inertia response and frequency modulation response test conditions specified in GB / T 36994-2018. The recommended frequency change rate range is ±0.1Hz / s to ±1Hz / s, with an interval of 0.05Hz / s; the frequency disturbance setting value is 48Hz to 51.5Hz, with an interval of 0.1Hz, and an interval of 0.01Hz near the dead zone.

[0083] This application is based on the current mainstream MATLAB / Simulink simulation software, and builds a general electromagnetic transient model of a full-power wind turbine; based on the correspondence between the unit's frequency response characteristics and the frequency control strategy / parameters, a practical identification method for a full-power wind turbine frequency controller is proposed, specifically including a practical identification method that uses the unit's measured frequency response characteristic curve morphology to identify the frequency controller structure, and uses the unit's measured frequency response characteristic curve eigenvalues ​​to identify the frequency controller parameters. This method improves the simulation accuracy of a high-proportion new energy power system, and provides technical support for improving the safe operation level of the power grid and the wind power acceptance capacity. The following is a detailed description in conjunction with the specific implementation method.

[0084] Full-power wind turbines capture wind energy through wind turbines, drive permanent magnet synchronous generators to generate electricity, and then are synchronously connected to the grid through "back-to-back" full-power converters and machine-end transformers. The methods of participating in frequency regulation mainly include comprehensive inertia control, power reserve control, and coordinated control between the two types. The general model of full-power wind turbines established in this application includes a wind turbine main circuit, a steady-state controller, and a frequency regulation controller module; the frequency regulation control module includes four specific control methods: virtual inertia control, droop control, overspeed load reduction control, and pitch angle load reduction control, which correspond to flags Flag 1 to Flag 4, respectively. A flag of 0 indicates that the control is disabled, and a flag of 1 indicates that the control is enabled; this application focuses on the identification of the structure and parameters of the frequency regulation controller, and the structure and parameters of the main circuit and steady-state control are known conditions.

[0085] The core idea of ​​the practical identification method of the frequency controller of a full-power wind turbine is to identify the structure of the frequency controller according to the curve shape of the frequency response of the full-power wind turbine, and to identify the parameters of the frequency controller according to the curve amplitude of the frequency response of the full-power wind turbine.

[0086] The practical identification method of the full-power wind turbine frequency controller structure includes the following steps:

[0087] (1) Build a frequency response characteristic curve library, including frequency and active power data;

[0088] (2) segmenting all measured curves in the curve library based on the frequency perturbation process;

[0089] (3) Identify the controller structure according to the correspondence between the curve shape and the frequency controller structure.

[0090] The practical identification method of frequency modulation controller parameters includes the following steps:

[0091] (1) Identify the active power threshold P for frequency modulation th ;

[0092] (2) Identification of load reduction reserve factor η del ;

[0093] (3) Identify the upper and lower limits of active power regulation dP max With dP min ;

[0094] (4) Identify the frequency modulation dead zone f db1 With f db2 ;

[0095] (5) Identification of droop coefficient K 2u With K 2d ;

[0096] (6) Identify the frequency change rate dead zone rf db1 With rf db2 ;

[0097] (7) Identification of virtual inertia coefficient K 1u With K 1d .

[0098] More specifically,

[0099] The test and simulation basis for this application reference:

[0100] GB / T 36994-2018 "Test Procedure for Grid Adaptability of Wind Turbine Generators"

[0101] GB / T 40595-2021 "Technical regulations and test guidelines for primary frequency regulation of grid-connected power supply"

[0102] NB / T 10315-2019 Technical Requirements and Test Procedures for Primary Frequency Regulation of Wind Turbines

[0103] like Figure 1As shown, the full-power wind turbine captures wind energy through the wind turbine, drives the permanent magnet synchronous generator to generate electricity, and then passes through the "back-to-back" full-power converter and the machine-end transformer to synchronize and connect to the grid. The methods of participating in frequency regulation mainly include comprehensive inertia control, power reserve control and coordinated control between the two categories. The general model of the full-power wind turbine established in this application includes the main circuit of the wind turbine, the steady-state controller and the frequency regulation controller module; the frequency regulation control module includes four specific control methods: virtual inertia control, droop control, overspeed load reduction control and pitch angle load reduction control, which correspond to flags Flag 1 to Flag4 respectively. The flag is 0 to disable the control, and the flag is 1 to enable the control; this application focuses on the identification of the structure and parameters of the frequency regulation controller, and the structure and parameters of the main circuit and steady-state control are known conditions.

[0104] The principle of frequency regulation control of full-power wind turbines is as follows:

[0105] The comprehensive inertia control consists of two parts: virtual inertia control and droop control. The former simulates the rotational inertia characteristics of the synchronous generator, uses the rotor kinetic energy stored in the fan to participate in the system frequency modulation control, and responds to the rate of change of the system frequency. The output characteristics can be expressed as:

[0106]

[0107] Where ΔP1 is the active power regulation output by the virtual inertia controller, P0 is the steady-state active power output of the wind turbine, df / dt is the frequency change rate, and K 1u With K 1d 、f db1 With f db2 、rf db1 With rf db2 They are the virtual inertia coefficient, frequency modulation dead zone and frequency change rate dead zone when the frequency decreases and increases respectively, f is the system frequency, Δf is the system frequency deviation, and t is the time.

[0108] The droop control adds an active power reference value proportional to the system frequency deviation to the wind turbine active power reference value to participate in the system frequency regulation. The response characteristics can be expressed as:

[0109]

[0110] Where ΔP2 is the active reference value adjustment value output by the droop controller, K 2u With K 2d Adjust the droop coefficient upward and downward respectively.

[0111] There are two main types of power reserve control: overspeed control and pitch angle load reduction control.

[0112] Pitch angle load reduction control: Figure 2As shown in Figure 1, as the pitch angle β increases, the wind turbine operating point changes from A to B, and the output power decreases. Based on this principle, the pitch angle control can be based on the set standby coefficient η del Realize load reduction operation, and when the system has frequency disturbance, adjust the pitch angle in real time to adjust the active output to achieve primary frequency modulation.

[0113] Overspeed load reduction control: Figure 2 It can be further seen that when the pitch angle is constant, there is a unique speed at each wind speed v that allows the wind turbine to achieve the optimal tip speed ratio λ opt , so that the wind turbine reaches the maximum wind energy utilization state, as shown in formula (3):

[0114]

[0115] Where: v is the wind speed, R is the rotor radius. Controlling the speed deviation of the fan can achieve power reduction and provide power reserve for frequency regulation. If it is achieved by reducing the speed, the stability of the unit will be reduced, and overspeed control is generally used. Figure 2 As shown in the figure, after adopting overspeed control, the wind turbine operating point moves from A to C, and the wind energy utilization rate increases from C to pmax Reduce to C pdel When the system experiences frequency disturbance, primary frequency regulation can be achieved by absorbing and releasing active power.

[0116] The structure of the frequency modulation controller determines the shape of the frequency response curve, and the parameters determine the amplitude of the frequency response. Identify its controller structure, that is, identify the shape of the wind turbine frequency response curve Figure 1 The conduction mode of the flag bit in the structure diagram of the universal frequency controller of the unit is summarized as follows: Figure 3 The corresponding relationship between the combination of different flag bits and the shape of the frequency modulation response curve is shown, and a practical identification method for the structure of the frequency modulation controller is proposed. The identification order and steps are as follows.

[0117] S1. Frequency response characteristic library is constructed. The actual unit or black box model is tested under different wind speeds and different frequency disturbances. Figure 2 A frequency response characteristic curve library containing frequency and active power data. The frequency disturbance is based on the inertia response and frequency modulation response test conditions specified in GB / T36994-2018. The recommended frequency change rate range is ±0.1Hz / s to ±1Hz / s, with an interval of 0.05Hz / s; the frequency disturbance setting value is 48Hz to 51.5Hz, with an interval of 0.1Hz, and the interval near the dead zone is 0.01Hz.

[0118] S2. Frequency response characteristic curve segmentation: All measured curves in the curve library are divided into five intervals: steady state before disturbance (t0-t1), transient state when disturbance occurs (t1-t2), transient steady state when disturbance is stable (t2-t3), transient state when disturbance is recovered (t3-t4) and steady state after disturbance (t4-t5). Among them, t0-t5 are the test start time, disturbance start time, the time when active power reaches stability after disturbance, the time when frequency starts to recover and the time when active power stabilizes again;

[0119] S3, frequency controller structure identification, calculate the Euclidean distance between the curve to be identified and the typical curve, obtain the curve shape with the highest similarity, correspond to the conduction mode of the flag bit, and thus identify the controller structure.

[0120] At this point, a method for identifying the structure of a frequency modulation controller based on the morphology of a frequency response curve has been formed, on this basis the parameters of the frequency modulation controller can be identified.

[0121] The key parameters of the frequency modulation controller include: the active power threshold P th , adjust the frequency change rate up and down dead zone rf db1 With rf db2 , adjust the frequency modulation dead zone f db1 With f db2 , adjust the virtual inertia coefficient K up or down 1u With K 1d , adjust the droop coefficient K up and down 2u With K 2d , load reduction reserve factor η del And active power regulation upper and lower limits dP max With dP min To ensure the economical operation of the fan, dP max It is generally equal to the load reduction reserve factor η in terms of value. del , no need to distinguish,

[0122] The order and steps for identifying the key parameters of the frequency modulation controller are as follows.

[0123] S1. The minimum steady-state active power P of the wind turbine in response to the frequency change under the condition of large frequency step disturbance (1Hz≤|Δf|) th * , which is the active power threshold P for the unit to participate in frequency regulation th .

[0124] S2, the maximum active output of the wind turbine during steady-state operation at the highest wind speed, that is, the maximum output after load reduction and standby P 0max , η del It can be calculated by formula (4):

[0125]

[0126] Where: P n is the rated active power of the wind turbine.

[0127] S3, the active power of the wind turbine at medium and high wind speed (0.4pu≤P0) and large frequency step change, that is, the maximum and minimum active power P of the wind turbine during the frequency disturbance 1max and P 1min , dP max With dP min It can be calculated by formula (5):

[0128]

[0129] S4, when the wind turbine is at medium or high wind speed and the frequency changes greatly, the highest frequency (frequency drop condition) and the lowest frequency (frequency rise condition) of the response frequency change, that is, f db1 * and f db2 * , f db1 With f db2 It can be calculated by formula (6):

[0130]

[0131] S5. Extract the average active power of the wind turbine in the interval t2-t3 under the conditions of medium and low wind speed (0.3pu≤P0<0.4pu) and small frequency step change (0.1Hz≤|Δf|<0.2Hz), that is, the characteristic value P 23 , in order to effectively avoid the influence of pitch angle action delay and active power regulation limit on active power response. Then, K is calculated by combining equations (2) and (7) using the least squares method. 2u With K 2d .

[0132] ΔP2=P 23 -P0 (7)

[0133] S6, maximum frequency change rate of the wind turbine in response to frequency change when the wind speed is medium or high and the frequency ramp changes (frequency drop condition) rf db1 * and minimum frequency change rate (frequency rising condition) rf db2 * , which is the frequency change rate dead zone rf db1 With rf db2 ;

[0134] S7, extract the average active power of the wind turbine in the interval t1-t2 under the conditions of medium and low wind speed, medium frequency slope change (0.3Hz / s<|df / dt|≤0.5Hz / s) and small frequency change amplitude (0.1Hz≤|Δf|<0.2Hz), that is, the characteristic value P 12 , combined with equations (1) and (8), the least squares method can be used to calculate K 1u With K 1d Under such test conditions, the active power regulation of the wind turbine is the superposition of the virtual inertia control and droop control commands, so the command value of the virtual inertia control is:

[0135] ΔP1=P 12 -P0-ΔP2 (8)

[0136] Among them, ΔP2 can be calculated by combining the identification results of S4 and S5 with the frequency data of the interval t1-t2.

[0137] It should be noted that the specific implementation is only an explanation and description of the technical solution of the present invention, and cannot be used to limit the scope of protection of the rights. Any partial changes made according to the claims and description of the present invention should still fall within the scope of protection of the present invention.

Claims

1. A method for identifying a full-power wind turbine frequency controller, characterized in that The following steps are involved: Step 1: Obtain the steady-state active power of the wind turbine in response to the frequency change under the condition of large frequency step disturbance, and select the lowest steady-state active power among the steady-state active power of the wind turbine in response to the frequency change, and then use the lowest steady-state active power as the active threshold P th , when the wind power active power is greater than the active power threshold P th When , execute step 2, the frequency large step disturbance condition is |Δf|≥1Hz, where Δf is the frequency change; Step 2: Identify the frequency modulation controller, wherein the identification of the frequency modulation controller includes structural identification of the frequency modulation controller and parameter identification of the frequency modulation controller; The specific steps of structural identification of the frequency controller are: Step 21: by turning on / off one or more frequency control modes, obtain corresponding frequency and active power data, obtain corresponding frequency response curves according to the frequency and active power data, and construct a typical frequency response curve library using the frequency response curves; Step 22: Obtain the frequency and active power data of the wind turbine to be identified, and obtain the curve to be identified based on the frequency and active power data of the wind turbine to be identified, and calculate the Euclidean distance between the curve to be identified and each frequency response characteristic curve in the typical frequency response curve library, select the frequency response characteristic curve with the highest similarity, and obtain the controller structure identification result based on the frequency response characteristic curve with the highest similarity, the controller structure identification result, that is, one or more frequency control modes corresponding to the curve, and its on / off state; The specific steps of parameter identification of the frequency modulation controller are: Step 1: Use the wind turbine active power of 0.2 pu as the dividing point between low wind speed and medium wind speed, and use the upper limit wind speed of load shedding as the dividing point between medium wind speed and high wind speed; Step 2: Obtain the active power output of the wind turbine during steady-state operation at the highest wind speed, and identify the load reduction reserve coefficient η based on the active power output of the wind turbine during steady-state operation at the highest wind speed. del ; Step 3: Obtain the active power of the wind turbine under the condition of large frequency step disturbance between medium and high wind speeds, and identify the upper limit of active power regulation dP according to the active power of the wind turbine under the condition of large frequency step change between medium and high wind speeds. max and active power regulation lower limit dP min ; Step 4: Obtain the highest frequency f of the wind turbine when the wind speed is between medium and high and the frequency is subject to a large step disturbance. db1 The minimum frequency f of the sum frequency change db2 , where the highest frequency of frequency change is f db1 , that is, the frequency drop condition, the lowest frequency f of the frequency change db2 , that is, the frequency rising condition; Step 5: Divide the curve to be identified into five intervals, namely the steady-state interval t0-t1 before the disturbance, the transient interval t1-t2 when the disturbance occurs, the transient steady-state interval t2-t3 when the disturbance is stable, the transient interval t3-t4 when the disturbance is recovered, and the steady-state interval t4-t5 after the disturbance. Among them, t0 represents the test start time, t1 represents the disturbance start time, t2 represents the time when the active power reaches stability after the disturbance, t3 represents the time when the frequency begins to recover, t4 represents the time when the active power stabilizes again, and t5 represents the test end time. Obtain the average active power of the wind turbine set within t2-t3 under the condition of medium to low wind speed and small frequency step disturbance, and combine it with f db1 and f db2 , and then identify the droop coefficient K 2u and K 2d , Among them, the frequency small step change is 0.1Hz<|Δf|≤0.2Hz; Step 6: Obtain the maximum frequency change rate rf of the wind turbine when the frequency changes ramp between medium and high wind speeds db1 The minimum frequency change rate rf db2 , where the maximum frequency change rate rf db1 , that is, the frequency drop condition, the minimum frequency change rate rf db2 , that is, the frequency rising condition; Step 7: Obtain the average active power of the wind turbine during t1-t2 under the conditions of medium to low wind speed, medium frequency slope change, and small frequency step disturbance, and combine it with f db1 、f db2 、rf db1 and rf db2 , and then identify the virtual inertia coefficient K 1u and K 1d ; Among them, the medium frequency ramp change is 0.3Hz / s<|df / dt|≤0.5Hz / s, df / dt is the frequency ramp change rate; Step 8: η obtained from steps 1 to 7 del , dP max , dP min 、f db1 、f db2 , K 2u , K 2d 、、rf db1 、rf db2 , K 1u and K 1d , which are the identified parameters of the frequency modulation controller.

2. The method for identifying a full-power wind turbine frequency controller according to claim 1, characterized in that The frequency control methods include: virtual inertia control, droop control, overspeed load reduction control and pitch angle load reduction control.

3. A method for identifying a full-power wind turbine frequency controller according to claim 2, characterized in that The specific steps of step 21 are: Step 2: Let virtual inertia control, droop control, overspeed load reduction control and pitch angle load reduction control correspond to flags Flag 1, Flag 2, Flag 3 and Flag 4 respectively. Then, group flags Flag 1, Flag 2, Flag 3 and Flag 4, specifically: a: Flag1 = 0, Flag2 = 0; b: Flag1 = 1, Flag2 = 0; c: Flag1 = 0, Flag2 = 1; d: Flag1 = 1, Flag2 = 1; e: Flag3 = 0; f: Flag3 = 1; g: Flag4 = 0; h: Flag4 = 1; Among them, the flag bit is 0, which means turning off the control, and the flag bit is 1, which means turning on the control; Step 212: According to the grouping in step 211, the frequency and active power of the high-voltage side of the wind turbine box transformer under different groups are obtained, and then the frequency response characteristic curve is obtained according to the frequency and active power, and then a typical frequency response curve library is constructed.

4. A method for identifying a full-power wind turbine frequency controller according to claim 3, characterized in that The load shedding reserve factor η del It is expressed as: Among them, P 0max P is the active output of the wind turbine during steady-state operation, i.e. the maximum output after load shedding and standby. n is the rated active power of the wind turbine.

5. A method for identifying a full-power wind turbine frequency controller according to claim 4, characterized in that The active power regulation upper limit dP max and active power regulation lower limit dP min It is expressed as: Among them, P 1max and P 1min They are the maximum and minimum active powers of the wind turbine during the frequency disturbance, and P0 is the steady-state active output of the wind turbine.

6. A method for identifying a full-power wind turbine frequency controller according to claim 5, characterized in that The highest frequency f of the frequency change db1 The minimum frequency f of the sum frequency change db2 It is expressed as: Among them, f db1 * and f db2 * are the highest frequency and the lowest frequency of the response frequency change of the wind turbine under the conditions of medium to high wind speed and large frequency step disturbance, respectively. n is the rated frequency.

7. A method for identifying a full-power wind turbine frequency controller according to claim 6, characterized in that The droop coefficient K 2u and K 2d It is expressed as: ΔP2=P 23 -P0 Among them, ΔP2 is the active reference value adjustment value output by the droop controller, K 2u With K 2d are the upward and downward droop coefficients, P 23 It is the average active power of the wind turbine during the period from t2 to t3.

8. A method for identifying a full-power wind turbine frequency controller according to claim 7, characterized in that The virtual inertia coefficient K 1u and K 1d It is expressed as: ΔP1=P 12 -P0-ΔP2 Among them, ΔP1 is the active regulation output by the virtual inertia controller, P 12 It is the average active power of the wind turbine during the period from t1 to t2.

9. A method for identifying a full-power wind turbine frequency controller according to claim 8, characterized in that The wind speed scenario of the wind turbine to be identified is taken at intervals of 0.5 m / s.

10. A method for identifying a full-power wind turbine frequency controller according to claim 9, characterized in that The frequency disturbance of the wind turbine to be identified is obtained based on the inertia response and frequency modulation response test conditions specified in GB / T 36994-2018, and the frequency change rate range is ±0.1 Hz / s to ±1 Hz / s, with an interval of 0.05 Hz / s; the frequency disturbance setting value is 48 Hz to 51.5 Hz, with an interval of 0.1 Hz, and a dead zone interval of 0.01 Hz.

Images