Wind turbine generator low voltage ride through dominant parameter analysis method based on trajectory sensitivity

Abstract

The invention discloses a wind turbine generator low-voltage ride-through dominant parameter analysis method based on trajectory sensitivity, and relates to the technical field of low-voltage ride-through, and the method comprises the steps: screening parameters related in a reactive power control process according to voltage sensitivity characteristics; calculating the trajectory sensitivity of the parameters, and extracting the frequency characteristic of the trajectory sensitivity based on a discrete Fourier transform method, namely the error characteristic caused by the difference between the model parameters and the actual parameters; the method comprises the following steps: setting an average amplitude index and an amplitude ratio index of parameters, evaluating the influence capability of the parameters, screening the parameters based on the average amplitude index to obtain a candidate parameter set, setting the number of dominant parameters, sorting the candidate parameters from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameters, and parameters with serial numbers smaller than the number of the dominant parameters are selected as dominant control parameters, so that the influence of each parameter on the bus voltage can be quantitatively evaluated, and the dominant control parameters of the wind turbine generator in the low voltage ride-through period are accurately identified.

Description

Technical Field

[0001] The present invention relates to the technical field of low voltage ride through, and in particular to a method for analyzing dominant parameters of low voltage ride through of a wind turbine generator set based on trajectory sensitivity. Background Art

[0002] Wind power generation is one of the fastest growing forms of renewable energy. In recent years, the capacity of wind power generation has continued to grow rapidly, while the scale of wind farms has also continued to expand. As wind power penetration in the power grid increases, low voltage ride-through (LVRT) has become a key technology in wind power research. The new grid code stipulates that when the voltage at the point of common coupling (PCC) drops during a grid fault, grid-connected wind turbines (WT) must remain connected to the grid and can even provide reactive power to the grid to support voltage recovery during low voltage periods.

[0003] In terms of grid-connected control strategies, many studies have been conducted on the optimization of wind turbine control parameters, but most of the optimization strategies involved are based on manual experience or a large-scale optimization of all grid-related control parameters. They fail to clearly and quantitatively evaluate the impact of each parameter on the bus voltage, and are unable to determine the dominant parameters.

[0004] Therefore, how to quantitatively evaluate the impact of various parameters on the bus voltage and accurately identify the dominant control parameters of wind turbines during low voltage ride-through has become an urgent problem to be solved in the field of low voltage ride-through technology. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a dominant parameter analysis method for low voltage ride-through control of wind turbines based on trajectory sensitivity. The dynamic characteristics between control parameters and voltage changes can be extracted according to the frequency domain characteristics of trajectory sensitivity, and the frequency domain feature information contained in trajectory sensitivity can be fully explored to provide theoretical guidance for the interpretability of dominant parameters. In addition, multiple sensitive frequency points appearing in the spectrum are considered, and the average amplitude and amplitude ratio are used as the main evaluation indicators to establish an identification process for the dominant parameter set, so that the identification results have higher accuracy.

[0006] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0007] The present invention is a method for analyzing dominant parameters of low voltage ride through control of a wind turbine generator set based on trajectory sensitivity, comprising the following steps:

[0008] The parameters involved in the reactive power control process are preliminarily screened out based on the voltage sensitivity characteristics;

[0009] Calculating the trajectory sensitivity of the parameter, and extracting the frequency characteristics of the trajectory sensitivity based on a discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameter and the actual parameter;

[0010] The average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter are set to evaluate the influence of the parameter, the parameter is screened based on the average amplitude index to obtain a candidate parameter set, the number of dominant parameters is set, and the candidate parameters are sorted from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and the parameter with a sequence number less than the number of dominant parameters is selected as the dominant control parameter.

[0011] A further improvement of the present invention is that the parameters involved in the reactive power control process are initially screened out according to the voltage sensitivity characteristics, specifically including:

[0012] During the low voltage period, the reactive power priority strategy is used. When the grid voltage drops below the normal value, it is necessary to limit the active current and expand the range of reactive current. The reactive power priority strategy is as follows:

[0013] i gqref =K·I N ·(U N -U)

[0014] In the formula, i gqref represents the reactive current reference value; K represents the reactive current reference value control coefficient; U represents the instantaneous terminal voltage; U N Indicates the nominal voltage of the wind turbine; I N Indicates the nominal current of the wind turbine;

[0015] When the wind turbine operates in fault ride-through mode, the control equation of the improved grid-side converter is as follows:

[0016]

[0017] In the formula, K 1_p Indicates the proportional control coefficient of active power control; K 1_i Indicates the integral control coefficient of active power control; K 2_p Indicates the proportional control coefficient of reactive power control; K 2_i Indicates the integral control coefficient of reactive power control; u ed represents the voltage control component on the d-axis; u eq represents the voltage control component on the q axis; i gdref Indicates the active current reference value; i gqref Represents the reactive current reference value; i gq Indicates the reactive current output value; i gd Indicates the active current output value; ω g Indicates angular frequency; L g Indicates the inductance of the filter circuit; u gd represents the voltage output component on the d-axis; u gqrepresents the voltage output component on the q axis; t represents time;

[0018] Different control parameters are set in the fault ride-through control of the direct-drive fan at the delivery end to simulate different output characteristics of the direct-drive fan at the delivery end. The expressions of the active current and reactive current control instructions during the low voltage ride-through process are as follows:

[0019] The expression of active current control instruction during low voltage ride-through is:

[0020] I p =K 1_Ip u g +K 2_Ip I p0 +I pset

[0021] In the formula, I p Indicates the active current control instruction during the low voltage ride-through process; K 1_Ip , K 2_Ip Indicates the low voltage ride through active current calculation coefficient; I pset Indicates the low voltage ride through active current setting value; I p0 Indicates the initial active current; u g Indicates the grid connection point voltage;

[0022] The expression of reactive current control instruction in the low voltage ride-through process is:

[0023] I q =K 1_Iq (0.9-u g )+K 2_Iq I q0 +I qset

[0024] In the formula, I q Indicates the reactive current control instruction during the low voltage ride-through process; K 1_Iq , K 2_Iq Indicates the low voltage ride-through reactive current calculation coefficient; I qset Indicates the low voltage ride-through reactive current setting value; I q0 Indicates the initial reactive current; u g Indicates the grid connection point voltage;

[0025] Assume that the bus voltage at the grid connection point is E, the grid-side voltage of the wind turbine is V, and the wind turbine output power is P+jQ. The grid-side voltage characteristics are analyzed based on the wind turbine grid-connected equivalent model:

[0026]

[0027] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance; θ represents the grid-side voltage phase of the wind turbine; sin represents the sine function; cos represents the cosine function;

[0028] The above formula can be simplified to:

[0029] V 4 -(2QX+E 2 )V 2 +(P 2 +Q 2 )X 2 =0

[0030] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance;

[0031] The active power and reactive power of the wind turbine are both decoupled and controlled. The wind turbine grid-side power grid obtains the partial derivative of the active power and reactive power of the wind turbine:

[0032]

[0033] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance;

[0034] Under normal working conditions, In the case of fault conditions,

[0035] In a steady state under normal or fault conditions, the reactive characteristics of the wind turbine during a low voltage fault and recovery process have a greater impact on the grid-side voltage of the wind turbine than the active characteristics do;

[0036] According to the voltage sensitivity characteristics, the parameters involved in the reactive power control process are preliminarily screened out, namely, the proportional control parameter K of the reactive power control. 2_p , Integral control parameter K of reactive power control 2_i 、Low voltage ride through reactive current calculation coefficient K 1_Iq , K 2_Iq .

[0037] A further improvement of the present invention is that the expression of the dynamic behavior of the wind turbine grid-connected sending-end power grid system after being disturbed is as follows:

[0038]

[0039] In the formula, t represents time; x(t) represents the system state variable with respect to time t; y(t) represents the system algebraic variable with respect to time t; α represents the selected parameter; represents the integral of the system state variables; f represents the equivalent function of the integral of the system state variables with respect to the state variables, algebraic variables and equivalent parameters; g represents the equivalent function of the error variable of the system with respect to the state variables, algebraic variables and equivalent parameters;

[0040] The trajectory sensitivity equation is as follows:

[0041]

[0042] In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α It represents the sensitivity of parameter α to the system algebraic variable y(t); represents the integral of sensitivity; f represents the equivalent function of the integral of the system state variable with respect to the state variable, algebraic variable and equivalent parameter; g represents the equivalent function of the system error variable with respect to the state variable, algebraic variable and equivalent parameter;

[0043] Simplify and calculate the trajectory sensitivity:

[0044]

[0045] In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α represents the sensitivity of parameter α to the system algebraic variable y(t); Δα is the parameter perturbation; t represents time.

[0046] A further improvement of the present invention is that the trajectory sensitivity of the parameters is calculated, and the frequency characteristics of the trajectory sensitivity are extracted based on the discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameters and the actual parameters, specifically including:

[0047] The expression of the time domain continuous curve f(t) is:

[0048]

[0049] Where F(t) represents the time domain continuous curve; a0 represents the constant offset; a c 、b c represents the Fourier coefficient of the cth harmonic; T represents the total duration; cos represents the cosine function; sin represents the sine function; t represents time; Σ represents the summation symbol;

[0050] The time domain continuous curve F(t) is sampled to obtain a time domain discrete finite length sample sequence A(n). The transformed frequency domain sample sequence B(k) contains the series related to the time domain discrete finite length sample sequence A(n). Each point in the frequency domain sample sequence B(k) contains the information of all time domain points. The expression of discrete Fourier transform is as follows:

[0051]

[0052] Wherein, N represents the length of the discrete finite-length sample sequence A(n) in the time domain; B(k) represents the frequency domain sample sequence corresponding to A(n); A(n) represents the discrete finite-length sample sequence in the time domain; Σ represents the summation symbol; k represents the discrete sequence number; n represents a natural number; j represents an imaginary unit;

[0053] The error Δo(n) is used to represent the difference between the measured data and the simulation results, as shown below:

[0054] Δo(n)=o r (n)-o s (n)

[0055] Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; r (n) represents measured data; o s (n) represents the simulation result;

[0056] Based on the error characteristics caused by the difference between the model parameters and the actual parameters, the error Δo(n) is expressed as the accumulation of trajectory sensitivity integrals, as shown below:

[0057]

[0058] Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; S w Representation and parameter α w The relevant trajectory sensitivity; Indicates the actual parameter value; represents the model parameter value; Y represents the total number of parameters; n represents a natural number;

[0059] The trajectory sensitivity of each parameter in the error sequence is transformed in the frequency domain according to the expression of discrete Fourier transform, and the sensitive frequency points appearing in the spectrum of the trajectory sensitivity are taken as analysis targets. The amplitude of the sensitive frequency points represents the influence of the parameters on the sensitive frequency points, and there are multiple sensitive frequency points in the spectrum.

[0060] A further improvement of the present invention is that: the average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter are set, the influence of the parameter is evaluated, the parameter is screened based on the average amplitude index to obtain a candidate parameter set, the number of dominant parameters is set, and the candidate parameters are sorted from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and the parameter with a sequence number less than the number of dominant parameters is selected as the dominant control parameter, specifically including:

[0061] For the sensitive frequency points of the spectrum, the average amplitude index is used Indicates the average influence of the parameter on the sensitive frequency point. The average amplitude index The expression is as follows:

[0062]

[0063] In the formula, represents the average amplitude index; Z represents the total number of sensitive frequency points; M l Indicates the amplitude of the lth sensitive frequency point;

[0064] The expression of the amplitude ratio index R is as follows:

[0065]

[0066] In the formula, R represents the amplitude ratio index; Indicates the average amplitude index; max indicates the maximum value operation; M indicates the amplitude index; l indicates the serial number of the sensitive frequency point; Z indicates the total number of sensitive frequency points;

[0067] The average amplitude index of the trajectory sensitivity of the parameter is used as the first screening condition. If one or more of the parameters α are not less than the threshold value H 1,l If the sensitive frequency amplitude and the average amplitude index is greater than the threshold H2, the parameter α is used as a candidate parameter, as shown in the following formula:

[0068]

[0069] In the formula, Ω c represents the candidate parameter set; M α,l represents the amplitude of the lth sensitive frequency point of parameter α; H 1,l Indicates the lth threshold corresponding to the amplitude of the sensitive frequency point; represents the average amplitude index of parameter α; H2 represents the threshold corresponding to the average amplitude index; Z represents the total number of sensitive frequency points;

[0070] The difference ΔR between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter is calculated according to the candidate parameter set, as shown in the following formula:

[0071] △R=|R α -R e |

[0072] Where ΔR represents the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of parameter α; R α The trajectory sensitivity amplitude ratio index of parameter α; R e The error amplitude ratio index representing parameter α;

[0073] According to the operating characteristics and working conditions of the power grid system at the sending end of the wind turbine, the number of dominant parameters is set to N. d , and sort the candidate parameters from small to large according to the difference ΔR, extracting the top N d The candidate parameters generate the dominant parameter set Ω d .

[0074] A further improvement of the present invention is that: the threshold H is determined based on the average amplitude index 1,l and H2:

[0075]

[0076] In the formula, g1 and g2 represent weight coefficients; H 1,l represents the lth threshold corresponding to the amplitude of the sensitive frequency point; H2 represents the threshold corresponding to the average amplitude index; M α,l Represents the amplitude of the lth sensitive frequency point of parameter α; represents the average amplitude index of parameter α; It means that for any parameter α, the expression takes the maximum value.

[0077] The beneficial effects of the present invention are:

[0078] The present invention proposes a method for analyzing the dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity. By constructing a wind power DC transmission model, the traditional control strategy and low voltage ride through control strategy of the grid-side converter are introduced in detail, and the reactive control parameters affecting the bus voltage are preliminarily screened out.

[0079] The present invention uses the frequency domain characteristics of trajectory sensitivity to extract the dynamic characteristics between control parameters and voltage changes, fully taps the frequency domain characteristic information contained in trajectory sensitivity, provides theoretical guidance for the interpretability of dominant parameters, and considers multiple sensitive frequency points appearing in the spectrum, takes average amplitude and amplitude ratio as the main evaluation indicators, and establishes the identification process of the dominant parameter set, so that its identification results have higher accuracy. By optimizing the dominant parameters, the performance of the system can be significantly improved, the impact of external interference on the system can be reduced, and at the same time, the system's dependence on specific parameters can be reduced, and the reliability of the system can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A schematic flow chart of a method for analyzing dominant parameters of low voltage ride through control of a wind turbine generator system based on trajectory sensitivity provided in an embodiment of the present invention;

[0081] Figure 2 A schematic diagram of a control strategy for an improved grid-side converter provided by an embodiment of the present invention;

[0082] Figure 3 A schematic diagram of a wind turbine grid-connected equivalent model provided by an embodiment of the present invention;

[0083] Figure 4 A schematic diagram of the amplitude ratio index of each parameter provided in an embodiment of the present invention;

[0084] Figure 5 The low-throughput reactive current control coefficient K provided in the embodiment of the present invention 1_Iq Schematic diagram of transient voltage waveform during change;

[0085] Figure 6 The low-throughput reactive current control coefficient K provided in the embodiment of the present invention 2_Iq Schematic diagram of transient voltage waveform during change;

[0086] Figure 7 The proportional control parameter K provided in the embodiment of the present invention is 2_p Schematic diagram of transient voltage waveform during change;

[0087] Figure 8 A schematic diagram of a transient voltage waveform when the integral control parameter K2_i provided in an embodiment of the present invention changes. DETAILED DESCRIPTION

[0088] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0089] See also Figure 1 The method for analyzing the dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity in this embodiment includes the following operations:

[0090] S1, preliminarily screen out the parameters involved in the reactive power control process according to the voltage sensitivity characteristics;

[0091] S2, calculating the trajectory sensitivity of the parameter, and extracting the frequency characteristics of the trajectory sensitivity based on a discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameter and the actual parameter;

[0092] S3, setting the average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter, evaluating the influence of the parameter, screening the parameter based on the average amplitude index to obtain a candidate parameter set, setting the number of dominant parameters, sorting the candidate parameters from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and selecting the parameter with a sequence number less than the number of dominant parameters as the dominant control parameter.

[0093] The parameters involved in the reactive power control process are initially screened out according to the voltage sensitivity characteristics, specifically including:

[0094] In order to support the grid voltage, a reactive power priority strategy is used during the low voltage duration. When the grid voltage drops below the normal value, it is necessary to limit the active current and expand the range of reactive current. The reactive power priority strategy is as follows:

[0095] i gqref =K·I N ·(U N -U)

[0096] In the formula, i gqref represents the reactive current reference value; K represents the reactive current reference value control coefficient; U represents the instantaneous terminal voltage; U N Indicates the nominal voltage of the wind turbine; I N Indicates the nominal current of the wind turbine;

[0097] The capacity of the grid-side converter can fully meet the needs of reactive power control. Through the new control strategy, the permanent magnet synchronous generator can have the capabilities of low voltage ride-through (LVRT) and grid voltage support at the same time.

[0098] See also Figure 2 , when the wind turbine operates in fault ride-through (FRT) mode, the control equation of the improved grid side converter (GSC) is as follows:

[0099]

[0100] In the formula, K 1_p Indicates the proportional control coefficient of active power control; K 1_i Indicates the integral control coefficient of active power control; K 2_p Indicates the proportional control coefficient of reactive power control; K 2_i Indicates the integral control coefficient of reactive power control; u ed represents the voltage control component on the d-axis; ueq represents the voltage control component on the q axis; i gdref Indicates the active current reference value; i gqref Represents the reactive current reference value; i gq Indicates the reactive current output value; i gd Indicates the active current output value; ω g Indicates angular frequency; L g Indicates the inductance of the filter circuit; u gd represents the voltage output component on the d-axis; u gq represents the voltage output component on the q axis; t represents time;

[0101] At this time, different control parameters are set in the fault ride-through control of the direct-drive fan at the delivery end to simulate the different output characteristics of the direct-drive fan at the delivery end. The expressions of the active current and reactive current control instructions during the low voltage ride-through process are as follows:

[0102] The expression of active current control instruction during low voltage ride-through is:

[0103] I p =K 1_Ip u g +K 2_Ip I p0 +I pset

[0104] In the formula, I p Indicates the active current control instruction during the low voltage ride-through process; K 1_Ip , K 2_Ip Indicates the low voltage ride through active current calculation coefficient; I pset Indicates the low voltage ride through active current setting value; I p0 Indicates the initial active current; u g Indicates the grid connection point voltage;

[0105] The expression of reactive current control instruction in the low voltage ride-through process is:

[0106] I q =K 1_Iq (0.9-u g )+K 2_Iq I q0 +I qset

[0107] In the formula, I q Indicates the reactive current control instruction during the low voltage ride-through process; K 1_Iq , K 2_Iq Indicates the low voltage ride-through reactive current calculation coefficient; I qset Indicates the low voltage ride-through reactive current setting value; I q0 Indicates the initial reactive current; ug Indicates the grid connection point voltage;

[0108] See also Figure 3 , assuming that the bus voltage at the grid connection point is E, the grid-side voltage of the wind turbine is V, and the wind turbine output power is P+jQ, the grid-side voltage characteristics are analyzed based on the wind turbine grid-connected equivalent model:

[0109]

[0110] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance; θ represents the grid-side voltage phase of the wind turbine; sin represents the sine function; cos represents the cosine function;

[0111] The above formula can be simplified to:

[0112] V 4 -(2QX+E 2 )V 2 +(P 2 +Q 2 )X 2 =0

[0113] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance;

[0114] The active power and reactive power of the wind turbine are both decoupled and controlled. The wind turbine grid-side power grid obtains the partial derivative of the active power and reactive power of the wind turbine:

[0115]

[0116] In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance;

[0117] Under normal working conditions, In the case of fault conditions,

[0118] In a steady state under normal or fault conditions, the reactive characteristics of the wind turbine during a low voltage fault and recovery process have a greater impact on the grid-side voltage of the wind turbine than the active characteristics do;

[0119] From the above data, it can be seen that, whether in a normal or faulty steady state, the reactive characteristics of the fan during low voltage fault and recovery have a greater impact on the grid voltage, while the active characteristics have a smaller impact on the grid voltage. Therefore, the present invention subsequently focuses on the analysis of parameters involved in the reactive control process, and preliminarily screens out the parameters involved in the reactive control process based on the voltage sensitivity characteristics, namely, the proportional control parameter K of the reactive control. 2_p , Integral control parameter K of reactive power control 2_i 、Low voltage ride through reactive current calculation coefficient K 1_Iq , K 2_Iq .

[0120] By analyzing the sensitivity of system parameter trajectories, the trajectory sensitivity can be obtained, and then the response of the system to small disturbances under specific operating conditions can be evaluated, so that the stability of the power system can be effectively monitored and controlled.

[0121] The dynamic behavior of the system after being disturbed can be described by the differential equation and algebraic equation of the system. The expression of the dynamic behavior of the wind turbine grid-connected sending-end power grid system after being disturbed is as follows:

[0122]

[0123] In the formula, t represents time; x(t) represents the system state variable with respect to time t; y(t) represents the system algebraic variable with respect to time t; α represents the selected parameter; represents the integral of the system state variables; f represents the equivalent function of the integral of the system state variables with respect to the state variables, algebraic variables and equivalent parameters; g represents the equivalent function of the error variable of the system with respect to the state variables, algebraic variables and equivalent parameters;

[0124] According to the definition of trajectory sensitivity, the trajectory sensitivity equation can be obtained as follows:

[0125]

[0126] In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α It represents the sensitivity of parameter α to the system algebraic variable y(t); represents the integral of sensitivity; f represents the equivalent function of the integral of the system state variable with respect to the state variable, algebraic variable and equivalent parameter; g represents the equivalent function of the system error variable with respect to the state variable, algebraic variable and equivalent parameter;

[0127] In order to solve the complex calculation problem caused by the simultaneous equations of the traditional numerical method, the trajectory sensitivity is simplified and calculated:

[0128]

[0129] In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α represents the sensitivity of parameter α to the system algebraic variable y(t); Δα is the parameter perturbation; t represents time.

[0130] The calculating of the trajectory sensitivity of the parameter and extracting the frequency characteristics of the trajectory sensitivity based on the discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameter and the actual parameter, specifically includes:

[0131] Since trajectory sensitivity is different from static sensitivity, it involves the trajectory of the system state variables changing over time, which can reveal the behavioral characteristics of the system in a non-steady state. Therefore, trajectory sensitivity pays more attention to the sensitivity of the state variables of the system to changes in parameters or initial conditions during the entire dynamic process. The present invention uses frequency domain analysis methods to extract dynamic features in trajectory sensitivity and simulation errors, establishes the relationship between the two, and provides important theoretical support for the accurate identification of dominant parameters.

[0132] Discrete Fourier Transform (DFT) can convert discrete signals in time or space into discrete signals in the frequency domain, and reveal the spectral characteristics of the signal by decomposing the signal into a combination of sine and cosine waves of different frequencies.

[0133] The expression of the time domain continuous curve f(t) is:

[0134]

[0135] Where F(t) represents the time domain continuous curve; a0 represents the constant offset; a c 、b c represents the Fourier coefficient of the cth harmonic; T represents the total duration; cos represents the cosine function; sin represents the sine function; t represents time; Σ represents the summation symbol;

[0136] The time domain continuous curve F(t) is sampled to obtain a time domain discrete finite length sample sequence A(n). The transformed frequency domain sample sequence B(k) contains the series related to the time domain discrete finite length sample sequence A(n). Each point in the frequency domain sample sequence B(k) contains the information of all time domain points. The expression of discrete Fourier transform is as follows:

[0137]

[0138] Wherein, N represents the length of the discrete finite-length sample sequence A(n) in the time domain; B(k) represents the frequency domain sample sequence corresponding to A(n); A(n) represents the discrete finite-length sample sequence in the time domain; Σ represents the summation symbol; k represents the discrete sequence number; n represents a natural number; j represents an imaginary unit;

[0139] The error Δo(n) is used to represent the difference between the measured data and the simulation results, as shown below:

[0140] Δo(n)=o r (n)-o s (n)

[0141] Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; r (n) represents measured data; o s (n) represents the simulation result;

[0142] Based on the error characteristics caused by the difference between the model parameters and the actual parameters, the error Δo(n) is expressed as the accumulation of trajectory sensitivity integrals, as shown below:

[0143]

[0144] Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; S w Representation and parameter α w The relevant trajectory sensitivity; Indicates the actual parameter value; represents the model parameter value; Y represents the total number of parameters; n represents a natural number;

[0145] The trajectory sensitivity of each parameter in the error sequence is transformed in the frequency domain according to the expression of discrete Fourier transform, and the sensitive frequency points appearing in the spectrum of the trajectory sensitivity are taken as analysis targets. The amplitude of the sensitive frequency points represents the influence of the parameters on the sensitive frequency points, and there are multiple sensitive frequency points in the spectrum.

[0146] The setting of the average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter, evaluating the influence of the parameter, screening the parameter based on the average amplitude index to obtain a candidate parameter set, setting the number of dominant parameters, sorting the candidate parameters from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and selecting the parameter with a sequence number less than the number of dominant parameters as the dominant control parameter, specifically includes:

[0147] Since there are multiple sensitive frequency points in the sensitivity and error spectrum, and their amplitudes may be different, each parameter has a different influence ability, and thus the present invention defines an average amplitude index and an amplitude ratio index.

[0148] For the sensitive frequency points of the spectrum, the average amplitude index is used Indicates the average influence of the parameter on the sensitive frequency point. The average amplitude index The expression is as follows:

[0149]

[0150] In the formula, represents the average amplitude index; Z represents the total number of sensitive frequency points; M l Indicates the amplitude of the lth sensitive frequency point;

[0151] The expression of the amplitude ratio index R is as follows:

[0152]

[0153] In the formula, R represents the amplitude ratio index; Indicates the average amplitude index; max indicates the maximum value operation; M indicates the amplitude index; l indicates the serial number of the sensitive frequency point; Z indicates the total number of sensitive frequency points;

[0154] When the parameter trajectory sensitivity amplitude ratio index and the error amplitude ratio index are similar, it means that the two have the same trend. Therefore, if the parameter is adjusted, the indicators can be adjusted synchronously.

[0155] The dominant parameter refers to the parameter that has the greatest impact on system performance, process results or model output, and plays a decisive role among many parameters. The present invention uses the average amplitude index of the trajectory sensitivity of the parameter as the first screening condition. If the parameter α has at least one value that is not less than the threshold value H 1,l If the sensitive frequency amplitude and the average amplitude index is greater than the threshold H2, the parameter α is used as a candidate parameter, as shown in the following formula:

[0156]

[0157] In the formula, Ω c represents the candidate parameter set; M α,l represents the amplitude of the lth sensitive frequency point of parameter α; H 1,l Indicates the lth threshold corresponding to the amplitude of the sensitive frequency point; represents the average amplitude index of parameter α; H2 represents the threshold corresponding to the average amplitude index; Z represents the total number of sensitive frequency points;

[0158] The difference ΔR between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter is calculated according to the candidate parameter set, as shown in the following formula:

[0159] △R=|R α -Re |

[0160] Where ΔR represents the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of parameter α; R α The trajectory sensitivity amplitude ratio index of parameter α; R e The error amplitude ratio index representing parameter α;

[0161] According to the operating characteristics and working conditions of the power grid system at the sending end of the wind turbine, the number of dominant parameters is set to N. d , and sort the candidate parameters from small to large according to the difference ΔR, extracting the top N d The candidate parameters generate the dominant parameter set Ω d .

[0162] The selection of threshold is a key step in data processing, which is directly related to the performance of the algorithm and the accuracy of the results. Selecting an appropriate threshold is crucial to improving the system sensitivity, specificity, accuracy and reducing the false positive rate.

[0163] Threshold H 1,l The purpose of setting is to screen out the parameters that have a significant impact on transient voltage. In the preliminary data analysis, the target variable may show different correlations with different parameters, which may show weak correlations or even no correlation. By setting a suitable threshold H 1,l , parameters with weak correlation or not statistically significant can be excluded. The setting of threshold H2 is based on the threshold H 1,l On the basis of screening, parameters that do not contribute much to the improvement of model performance are further excluded.

[0164] Determine the threshold H based on the average amplitude index 1,l and H2:

[0165]

[0166] In the formula, g1 and g2 represent weight coefficients; H 1,l represents the lth threshold corresponding to the amplitude of the sensitive frequency point; H2 represents the threshold corresponding to the average amplitude index; M α,l Represents the amplitude of the lth sensitive frequency point of parameter α; represents the average amplitude index of parameter α; It means that for any parameter α, the expression takes the maximum value.

[0167] Among them, the values ​​of weight coefficients g1 and g2 need to comprehensively consider the system operation status, disturbance conditions and trajectory sensitivity characteristics of the parameters.

[0168] In order to verify the effect of the present invention, the advancement of the proposed wind turbine low voltage ride-through control parameter analysis method based on trajectory sensitivity is described through the following specific examples:

[0169] A wind power DC transmission system simulation system was built on the electromagnetic transient simulation platform PSCAD / EMTDC, and modeling was performed based on the classic design parameters of wind turbines. According to the theoretical analysis of the above dominant parameters, the difference between the trajectory sensitivity and the error amplitude ratio index was used as the screening condition. Figure 4 , the difference values ​​of the parameters in the candidate parameter set are ranked from small to large as follows: K 1_Iq , K 2_Iq , K 2_p Therefore, the dominant parameter in the wind power low voltage ride-through control process considered in the present invention is K 1_Iq . By optimizing the dominant parameters, the system performance can be significantly improved, the impact of external interference on the system can be reduced, and the system's dependence on specific parameters can be reduced, thereby improving the system's reliability. However, the high sensitivity of the parameters also brings corresponding risks. If they are not adjusted properly, the grid-connected transient voltage may deteriorate sharply, and even affect the stable operation of the system. Therefore, understanding and controlling the dominant parameters is extremely important to ensure the safe, efficient and stable operation of the system.

[0170] In order to verify the control performance of the dominant parameters, the proportional control parameter K 2_p , integral control parameter K 2_i , low-through reactive current control coefficient K 1_Iq , K 2_Iq , increase the amplitude by 20% (scenario 2) and 50% (scenario 3) on the traditional value (scenario 1), add a fault link at the simulation time of 1.0s-1.2s, and compare the transient voltage simulation waveform under the wind power transmission model.

[0171] See also Figure 5-Figure 8 , when each control parameter changes with the same amplitude, the low-pass reactive current control coefficient K 1_Iq The impact is the greatest, causing the transient voltage to rise from 0.667pu to 0.802pu and 0.909pu. Compared with the low-through reactive current control coefficient K 1_Iq , proportional control parameter K 2_p , integral control parameter K 2_i , the effects were increased by 27%, 191%, and 554% respectively. Therefore, the parameter K 1_Iq As the dominant parameter in the low voltage ride-through process, it is consistent with the above theoretical analysis, verifying the correctness of the strategy proposed in the present invention.

[0172] Through the introduction of the above embodiments, the present invention uses a method for analyzing the dominant parameters of low voltage ride-through control of wind turbines based on trajectory sensitivity, and constructs a wind power DC transmission model to introduce in detail the traditional control strategy and low voltage ride-through control strategy of the grid-side converter, and preliminarily screens out the reactive control parameters that affect the bus voltage. The present invention uses the frequency domain characteristics of trajectory sensitivity to extract the dynamic characteristics between the control parameters and the voltage change, fully explores the frequency domain characteristic information contained in the trajectory sensitivity, provides theoretical guidance for the interpretability of the dominant parameters, and considers multiple sensitive frequency points appearing in the spectrum. With the average amplitude and amplitude ratio as the main evaluation indicators, an identification process for the dominant parameter set is established, so that the identification results have higher accuracy. By optimizing the dominant parameters, the performance of the system can be significantly improved, the impact of external interference on the system can be reduced, and at the same time, the system's dependence on specific parameters can be reduced, thereby improving the reliability of the system.

[0173] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have the same meaning as in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined similarly here.

[0174] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for analyzing dominant parameters of low voltage ride-through control of wind turbines based on trajectory sensitivity, characterized in that: The steps include: The parameters involved in the reactive power control process are preliminarily screened out based on the voltage sensitivity characteristics; Calculating the trajectory sensitivity of the parameter, and extracting the frequency characteristics of the trajectory sensitivity based on a discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameter and the actual parameter; The average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter are set to evaluate the influence of the parameter, the parameter is screened based on the average amplitude index to obtain a candidate parameter set, the number of dominant parameters is set, and the candidate parameters are sorted from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and the parameter with a sequence number less than the number of dominant parameters is selected as the dominant control parameter.

2. The method for analyzing dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity according to claim 1 is characterized in that: The parameters involved in the reactive power control process are initially screened out according to the voltage sensitivity characteristics, specifically including: During the low voltage period, the reactive power priority strategy is used. When the grid voltage drops below the normal value, the active current is limited and the range of reactive current is expanded. The reactive power priority strategy is as follows: and gqref =K·I N ·(IN N -IN) In the formula, i gqref represents the reactive current reference value; K represents the reactive current reference value control coefficient; U represents the instantaneous terminal voltage; U N Indicates the nominal voltage of the wind turbine; I N Indicates the nominal current of the wind turbine; When the wind turbine operates in fault ride-through mode, the control equation of the improved grid-side converter is as follows: In the formula, K 1_p Indicates the proportional control coefficient of active power control; K 1_i represents the integral control coefficient of active power control; K2_p represents the proportional control coefficient of reactive power control; K2_i represents the integral control coefficient of reactive power control; u ed represents the voltage control component on the d-axis; u eq represents the voltage control component on the q axis; i gdref Indicates the active current reference value; i gqref Represents the reactive current reference value; i gq Indicates the reactive current output value; i gd Indicates the active current output value; ω g Indicates angular frequency; L g Indicates the inductance of the filter circuit; u gd represents the voltage output component on the d-axis; u gq represents the voltage output component on the q axis; t represents time; Different control parameters are set in the fault ride-through control of the direct-drive fan at the delivery end to simulate different output characteristics of the direct-drive fan at the delivery end. The expressions of the active current and reactive current control instructions during the low voltage ride-through process are as follows: The expression of active current control instruction during low voltage ride-through is: I p =K 1_Ip u g +K 2_Ip I p0 +I pset In the formula, I p Indicates the active current control instruction during the low voltage ride-through process; K 1_Ip , K 2_Ip Indicates the low voltage ride through active current calculation coefficient; I pset Indicates the low voltage ride through active current setting value; I p0 Indicates the initial active current; u g Indicates the grid connection point voltage; The expression of reactive current control instruction in the low voltage ride-through process is: I q =K 1_Iq (0.9-u g )+K 2_Iq I q0 +I qset In the formula, I q Indicates the reactive current control instruction during the low voltage ride-through process; K 1_Iq , K 2_Iq Indicates the low voltage ride-through reactive current calculation coefficient; I qset Indicates the low voltage ride-through reactive current setting value; I q0 Indicates the initial reactive current; u g Indicates the grid connection point voltage; Assume that the bus voltage at the grid connection point is E, the grid-side voltage of the wind turbine is V, and the wind turbine output power is P+jQ. The grid-side voltage characteristics are analyzed based on the wind turbine grid-connected equivalent model: In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance; θ represents the grid-side voltage phase of the wind turbine; sin represents the sine function; cos represents the cosine function; The above formula can be simplified to: In 4 -(2QX+E 2 )V 2 +(P 2 +Q 2 )X 2 =0 In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance; The active power and reactive power of the wind turbine are both decoupled and controlled. The wind turbine grid-side power grid obtains the partial derivative of the active power and reactive power of the wind turbine: In the formula, P represents the active power of the wind turbine; Q represents the reactive power of the wind turbine; E represents the bus voltage at the grid-connected location; V represents the grid-side voltage of the wind turbine; X represents the total grid-connected impedance; Under normal working conditions, In the case of fault conditions, In a steady state under normal or fault conditions, the reactive characteristics of the wind turbine during a low voltage fault and recovery process have a greater impact on the grid-side voltage of the wind turbine than the active characteristics do; According to the voltage sensitivity characteristics, the parameters involved in the reactive power control process are preliminarily screened out, namely, the proportional control parameter K of the reactive power control. 2_p , Integral control parameter K of reactive power control 2_i 、Low voltage ride through reactive current calculation coefficient K 1_Iq , K 2_Iq .

3. The method for analyzing dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity according to claim 2 is characterized in that: The expression of the dynamic behavior of the wind turbine grid-connected sending-end power grid system after being disturbed is as follows: In the formula, t represents time; x(t) represents the system state variable with respect to time t; y(t) represents the system algebraic variable with respect to time t; α represents the selected parameter; represents the integral of the system state variables; f represents the equivalent function of the integral of the system state variables with respect to the state variables, algebraic variables and equivalent parameters; g represents the equivalent function of the error variable of the system with respect to the state variables, algebraic variables and equivalent parameters; The trajectory sensitivity equation is as follows: In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α It represents the sensitivity of parameter α to the system algebraic variable y(t); represents the integral of sensitivity; f represents the equivalent function of the integral of the system state variable with respect to the state variable, algebraic variable and equivalent parameter; g represents the equivalent function of the system error variable with respect to the state variable, algebraic variable and equivalent parameter; Simplify and calculate the trajectory sensitivity: In the formula, x α represents the sensitivity of parameter α to the system state variable x(t); y α represents the sensitivity of parameter α to the system algebraic variable y(t); Δα is the parameter perturbation; t represents time.

4. The method for analyzing dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity according to claim 1 is characterized in that: The calculating of the trajectory sensitivity of the parameter and extracting the frequency characteristics of the trajectory sensitivity based on the discrete Fourier transform method, that is, the error characteristics caused by the difference between the model parameter and the actual parameter, specifically includes: The expression of the time domain continuous curve f(t) is: Where F(t) represents the time domain continuous curve; a0 represents the constant offset; a c , b c represents the Fourier coefficient of the cth harmonic; T represents the total duration; cos represents the cosine function; sin represents the sine function; t represents time; Σ represents the summation symbol; The time domain continuous curve F(t) is sampled to obtain a time domain discrete finite length sample sequence A(n). The transformed frequency domain sample sequence B(k) contains the series related to the time domain discrete finite length sample sequence A(n). Each point in the frequency domain sample sequence B(k) contains the information of all time domain points. The expression of discrete Fourier transform is as follows: Wherein, N represents the length of the discrete finite-length sample sequence A(n) in the time domain; B(k) represents the frequency domain sample sequence corresponding to A(n); A(n) represents the discrete finite-length sample sequence in the time domain; Σ represents the summation symbol; k represents the discrete sequence number; n represents a natural number; j represents an imaginary unit; The error Δo(n) is used to represent the difference between the measured data and the simulation results, as shown below: Δo(n)=o r (n)-o s (n) Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; r (n) represents measured data; o s (n) represents the simulation result; Based on the error characteristics caused by the difference between the model parameters and the actual parameters, the error Δo(n) is expressed as the accumulation of trajectory sensitivity integrals, as shown below: Where Δo(n) represents the error between the measured data and the simulation results, i.e., the difference characteristic; S w Representation and parameter α w The relevant trajectory sensitivity; Indicates the actual parameter value; represents the model parameter value; Y represents the total number of parameters; n represents a natural number; The trajectory sensitivity of each parameter in the error sequence is transformed in the frequency domain according to the expression of discrete Fourier transform, and the sensitive frequency points appearing in the spectrum of the trajectory sensitivity are taken as analysis targets. The amplitude of the sensitive frequency points represents the influence of the parameters on the sensitive frequency points, and there are multiple sensitive frequency points in the spectrum.

5. The method for analyzing dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity according to claim 1 is characterized in that: The setting of the average amplitude index and the amplitude ratio index of the trajectory sensitivity of the parameter, evaluating the influence of the parameter, screening the parameter based on the average amplitude index to obtain a candidate parameter set, setting the number of dominant parameters, sorting the candidate parameters from small to large according to the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter, and selecting the parameter with a sequence number less than the number of dominant parameters as the dominant control parameter, specifically includes: For the sensitive frequency points of the spectrum, the average amplitude index is used Indicates the average influence of the parameter on the sensitive frequency point. The average amplitude index The expression is as follows: In the formula, represents the average amplitude index; Z represents the total number of sensitive frequency points; M l Indicates the amplitude of the lth sensitive frequency point; The expression of the amplitude ratio index R is as follows: In the formula, R represents the amplitude ratio index; Indicates the average amplitude index; max indicates the maximum value operation; M indicates the amplitude index; l indicates the serial number of the sensitive frequency point; Z indicates the total number of sensitive frequency points; The average amplitude index of the trajectory sensitivity of the parameter is used as the first screening condition. If one or more of the parameters α are not less than the threshold value H 1,l If the sensitive frequency amplitude and the average amplitude index is greater than the threshold H2, the parameter α is used as a candidate parameter, as shown in the following formula: In the formula, Ω c represents the candidate parameter set; M α,l represents the amplitude of the lth sensitive frequency point of parameter α; H 1,l Indicates the lth threshold corresponding to the amplitude of the sensitive frequency point; represents the average amplitude index of parameter α; H2 represents the threshold corresponding to the average amplitude index; Z represents the total number of sensitive frequency points; The difference ΔR between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of the parameter is calculated according to the candidate parameter set, as shown in the following formula: ΔR=|R α -R e | Where ΔR represents the difference between the trajectory sensitivity amplitude ratio index and the error amplitude ratio index of parameter α; R α The trajectory sensitivity amplitude ratio index of parameter α; R e The error amplitude ratio index representing parameter α; According to the operating characteristics and working conditions of the power grid system at the sending end of the wind turbine, the number of dominant parameters is set to N. d , and sort the candidate parameters from small to large according to the difference ΔR, extracting the top N d The candidate parameters generate the dominant parameter set Ω d .

6. The method for analyzing dominant parameters of low voltage ride through control of wind turbines based on trajectory sensitivity according to claim 5 is characterized in that: Determine the threshold H based on the average amplitude index 1,l and H2: In the formula, g1 and g2 represent weight coefficients; H 1,l represents the lth threshold corresponding to the amplitude of the sensitive frequency point; H2 represents the threshold corresponding to the average amplitude index; M α,l Represents the amplitude of the lth sensitive frequency point of parameter α; represents the average amplitude index of parameter α; It means that for any parameter α, the expression takes the maximum value.

Images