Fault ride-through control method for doubly-fed motor based on transient impedance and power angle compensation

Abstract

The application belongs to the technical field of wind power generator control, and particularly relates to a doubly-fed motor fault ride-through control method based on transient impedance and power angle compensation. The method comprises the following steps: obtaining a first electrical quantity representing the size of rotor current; judging whether the first electrical quantity exceeds a preset safety threshold; when the safety threshold is exceeded, generating a transient virtual impedance according to an out-of-limit deviation; generating a compensation torque according to a virtual active power dissipation power corresponding to the transient virtual impedance; introducing the compensation torque into a virtual synchronous generator swing equation to compensate for the power balance deviation generated in the current limiting process; and generating a driving signal of a rotor-side converter according to the virtual synchronous generator output, the transient virtual impedance and positive and negative sequence voltage and current components. The application can inhibit the overcurrent of the doubly-fed motor during fault ride-through, and reduce the power angle deviation and instability risk in virtual synchronous control.

Description

Technical Field

[0001] This invention relates to the field of wind turbine control technology, and in particular to a fault ride-through control method for a doubly fed induction generator based on transient impedance and power angle compensation. Background Technology

[0002] As the proportion of new energy sources in the power grid continues to increase, the power system is placing higher demands on the grid support capabilities of new energy generating units. Traditional doubly-fed wind turbines are gradually shifting from grid-following control to grid-building control. Grid-building control simulates the external characteristics of synchronous generators, making the doubly-fed motor exhibit voltage source characteristics at the grid connection point, thus enabling it to support weak grid voltage and frequency. However, during transient processes such as voltage dips or surges in the grid, existing grid-building doubly-fed motors have significant drawbacks: First, the risk of overcurrent is extremely high. Grid-building motors exhibit voltage source characteristics, and when the grid voltage changes abruptly, the voltage difference between the virtual electromotive force inside the motor and the grid-side voltage increases dramatically, leading to a large transient overcurrent on the rotor side, seriously threatening the safety of the rotor-side converter; second, the contradiction between current limiting and stability is prominent. Introducing virtual impedance current limiting will generate additional virtual voltage drop and virtual power dissipation, causing power imbalance in the swing equation, leading to power angle oscillations or even loss of synchronization and grid disconnection; third, there is a lack of rigorous asymmetric fault response strategies, failing to distinguish the differences in impedance requirements between symmetrical and asymmetric faults, and lacking targeted suppression of negative sequence components. Even when using virtual impedance to limit the current of a grid-connected doubly fed generator, existing technologies typically only focus on suppressing the current amplitude, without feeding back the equivalent power dissipation caused by the virtual impedance to the virtual synchronous generator swing equation for power angle compensation. This results in a coupling conflict between current limiting control and grid stability control. Summary of the Invention

[0003] This invention proposes a fault ride-through control method for doubly-fed induction generators based on transient impedance and power angle compensation, aiming to solve the technical problem in the existing grid-controlled doubly-fed induction generator fault ride-through that cannot resolve the contradiction between "overcurrent vulnerability" and "loss limiting instability".

[0004] This invention provides a fault ride-through control method for a doubly-fed induction generator based on transient impedance and power angle compensation, comprising: Obtain the first electrical quantity characterizing the magnitude of the rotor current of the doubly-fed motor; Determine whether the first electrical quantity exceeds a preset safety threshold; If the threshold is exceeded, a transient virtual impedance for limiting rotor overcurrent is adaptively constructed based on the deviation between the first electrical quantity and the safety threshold. Based on the virtual power dissipation caused by the transient virtual impedance, a compensation torque is generated to compensate for the power imbalance of the virtual synchronous generator. The compensation torque is introduced into the swing equation of the virtual synchronous generator to correct the power balance relationship in the swing equation; Based on the output of the virtual synchronous generator and the transient virtual impedance, a drive signal for controlling the rotor-side converter is generated.

[0005] Furthermore, obtaining the first electrical quantity characterizing the magnitude of the doubly-fed motor rotor current includes: acquiring the three-phase rotor current and extracting the amplitude of the rotor positive-sequence current using a dual second-order generalized integrator. ; The adaptive construction of the transient virtual impedance includes: based on the amplitude With the security threshold Integrate the out-of-limit error to generate a transient virtual resistance. and transient virtual reactance The expression is: ; ; All electrical variables, control parameters, and impedance commands in the aforementioned adaptive reconfiguration control law have been per-unit (pu) based on the doubly-fed motor capacity reference. If the output impedance needs to be converted to actual ohm units, it must be multiplied by the reference impedance. Integral gain and The corresponding physical dimensions are This essentially characterizes the transient time-domain response speed of the virtual impedance injection loop when a transient overcurrent occurs. Furthermore, to ensure the completeness of the aforementioned transient impedance reconstruction strategy in both mathematical description and actual control closed loop, a mathematically defined "integrator with conditional control" is introduced. That is, the integral operator is not an unconstrained pure integral, but rather a conditional integral operator defined in a nonlinear bounded space with piecewise time-domain reset and nonlinear anti-saturation limiting constraints. Its mathematical boundary conditions are defined as follows: 1) Limiting constraint: Output impedance is limited to physical safety boundaries. Within the range, the integral accumulation value is forcibly cleared to zero when it is less than 0, and integration stops when it is greater than the maximum design safety impedance; 2) Transient reset constraint: When the system determines that the fault ride-through has ended and the rotor positive sequence current has returned to a steady state below the safe threshold (i.e. Duration exceeding the set threshold When this occurs, the integrator executes a desaturation and zeroing mechanism, driving... and Reset to zero, thereby restoring the unit to steady-state grid operation.

[0006] Furthermore, the generation of the compensation torque for compensating the power imbalance of the virtual synchronous generator includes: acquiring the stator three-phase current, extracting the stator positive-sequence current component through a dual second-order generalized integrator, and obtaining the stator positive-sequence current amplitude. Based on the virtual resistance component in the transient virtual impedance The equivalent virtual resistance on the stator side is calculated. According to the stator positive sequence current amplitude and the equivalent virtual resistance on the stator side Calculate virtual power dissipation According to the virtual power dissipation power Virtual angular velocity of virtual synchronous generator Calculate the compensation torque : ; in, The virtual power dissipation power corresponding to the transient virtual impedance, This represents the virtual angular velocity of the virtual synchronous generator.

[0007] In this embodiment, although the transient virtual impedance is generated by the rotor positive sequence current over-limit error, its control effect is ultimately manifested as the additional voltage drop relationship between the equivalent output voltage and current at the stator grid connection point. To facilitate power balance compensation in the virtual synchronous generator swing equation, the virtual resistance component in the transient virtual impedance can be converted into the equivalent virtual resistance on the stator side. .in, The stator and rotor voltage equations, flux coupling relationship, and per-unit reference used in the doubly-fed motor can be determined based on the doubly-fed motor's stator and rotor voltage equations, flux coupling relationship, and the per-unit reference used. The conversion yields results under a unified capacity and voltage standard. Can be taken as and The corresponding per-unit equivalent value.

[0008] During fault ride-through, when the stator positive sequence current flows through the equivalent virtual resistance on the stator side, it will generate equivalent virtual power dissipation. The virtual active power dissipation causes a shift in the active power balance relationship in the virtual synchronous generator swing equation. To compensate for this shift, this embodiment is based on... With virtual angular velocity Calculate the compensation torque: ; in, It can be calculated based on the stator positive sequence current amplitude and the stator-side equivalent virtual resistance. By incorporating the compensation torque into the swing equation, the equivalent power deficit caused by the transient virtual impedance current limiting process can be compensated, thereby suppressing the transient deviation of virtual angular velocity and power angle.

[0009] Furthermore, incorporating the compensation torque into the swing equation of the virtual synchronous generator includes: adding the compensation torque as a feedforward subtraction term to the electromagnetic torque term in the swing equation, resulting in the corrected swing equation: ; in For virtual rotational inertia, For virtual mechanical torque, For actual electromagnetic torque, For the compensation torque, This is the virtual damping coefficient. The rated angular velocity of the power grid. The present invention does not simply counteract the power angle deviation by increasing the restoring torque, but rather converts the equivalent power dissipation caused by the virtual impedance into a compensating torque and introduces this compensating torque into the swing equation to reduce the net acceleration torque and power balance deviation during the current limiting process. This suppresses virtual angular velocity fluctuations, keeping the power angle near the pre-fault steady-state power angle, thereby reducing the risk of power angle overshoot, oscillation, or loss of synchronization during current limiting ride-through of the grid-connected doubly-fed induction generator.

[0010] Furthermore, generating a drive signal based on the output of the virtual synchronous generator and the transient virtual impedance includes: converting the steady-state internal potential vector output by the virtual synchronous generator... Subtract the transient virtual impedance With stator positive sequence current vector The product voltage drop generates the stator-side equivalent positive-sequence voltage reference vector: ; Then, based on the stator and rotor voltage equations, magnetic flux coupling relationship, and rotor electrical angle position of the doubly-fed motor, the equivalent positive-sequence voltage reference vector on the stator side is converted into a positive-sequence rotor voltage reference vector; The drive signal is generated based on the positive sequence rotor voltage reference vector.

[0011] Furthermore, the method also includes asymmetric fault handling: The stator three-phase voltage and three-phase current are collected, and the negative sequence component of the stator voltage is separated by a dual second-order generalized integrator. and stator current negative sequence component ; Based on the preset negative sequence virtual impedance Generate the stator-side equivalent negative sequence voltage reference vector Then, based on the stator and rotor voltage equations, magnetic flux coupling relationship, and rotor electric angle position of the doubly fed motor, the equivalent negative sequence voltage reference vector on the stator side is converted into a negative sequence rotor voltage reference vector. The negative sequence rotor voltage reference vector is superimposed with the positive sequence rotor voltage reference vector to obtain the rotor-side total voltage reference vector, and the drive signal is generated based on the rotor-side total voltage reference vector; Wherein, the negative sequence virtual impedance The size can be set to 1 to 3 times the rated impedance of the doubly fed motor, preferably 2 times.

[0012] Furthermore, prior to the asymmetric fault handling, the method further includes: separating the negative sequence component of the stator voltage using a dual second-order generalized integrator. Calculate its amplitude | |;When| When the value exceeds a preset threshold, an asymmetrical fault is determined, and the steps of generating an equivalent negative sequence voltage reference vector on the stator side and calculating a negative sequence rotor voltage reference vector are executed.

[0013] Furthermore, the generation of the drive signal for controlling the rotor-side converter includes: Based on the collected rotor mechanical position angle sum of extreme logarithms Calculate the rotor electrical angular position ; Based on the virtual internal potential phase of the virtual synchronous generator With respect to the rotor electrical angular position Calculate the slip angle ; The combined total rotor voltage reference vector is transformed to a stationary coordinate system by using the slip angle in the inverse coordinate transformation. The driving signal is generated by space vector pulse width modulation.

[0014] Furthermore, before generating the drive signal based on the output of the virtual synchronous generator and the transient virtual impedance, the process further includes: The stator three-phase voltage and three-phase current are collected, and the positive-sequence component of the stator voltage is separated by a dual second-order generalized integrator. and stator current positive sequence component ; The actual electromagnetic active power is calculated using the positive sequence components of the stator voltage and stator current. and reactive power To filter out second harmonic fluctuations under asymmetric faults; The active power and reactive power By inputting the active-frequency loop and reactive-voltage loop of the virtual synchronous generator, the magnitude of the steady-state internal potential vector is calculated. and phase .

[0015] Furthermore, when it is determined that the first electrical quantity does not exceed the safety threshold, the transient virtual impedance is set to zero, and the compensation torque is set to zero; in the absence of an asymmetric fault, the drive signal is generated based on the steady-state output of the virtual synchronous generator; in the presence of an asymmetric fault, the drive signal is also superimposed with a negative sequence voltage reference quantity generated based on the negative sequence component.

[0016] The technical advantages of the fault ride-through control method for a doubly-fed induction generator (DFIG) based on transient impedance and power angle compensation disclosed in this invention are as follows: By reconstructing the adaptive transient impedance based on the degree of rotor current exceeding the limit, the overcurrent on the rotor side is suppressed during grid voltage dips, surges, or asymmetrical faults, keeping the rotor current near the safe range allowed by the converter; at the same time, by generating a compensation torque based on the power dissipation of the virtual impedance and introducing this compensation torque into the swing equation of the virtual synchronous generator, the power imbalance caused by current limiting is compensated, and the transient offset of virtual angular velocity and power angle is reduced, thereby improving the stability and recovery capability of the grid-connected DFIG during fault ride-through. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the fault ride-through control method for a doubly fed motor based on transient impedance and power angle compensation proposed in an embodiment of the present invention. Figure 2 for Figure 1 The logic block diagram of the voltage ride-through coordinated control system within the dashed box; Figure 3 A schematic flowchart of a fault ride-through control method for a doubly fed motor based on transient impedance and power angle compensation provided in an embodiment of the present invention. Figure 4 The control signal flow diagram provided in the embodiments of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] With the increasing penetration of new energy sources, grid-connected doubly-fed induction generators (DFIGs) possess grid support capabilities due to their simulated synchronous generator characteristics. However, they face severe overcurrent risks during transient processes such as grid voltage dips or surges. Existing current-limiting methods that introduce virtual impedance disrupt the power balance of the virtual synchronous generator (VSG), leading to power angle oscillations or even loss of synchronization. Therefore, this invention provides a fault ride-through control method for DFIGs based on transient impedance and power angle compensation. The applicable system architecture includes a DFIG, a rotor-side converter, a grid-side converter, and a voltage ride-through coordinated control system. The stator of the DFIG is directly connected to the grid, and the rotor is connected to the DC bus via the rotor-side converter. The grid-side converter connects the DC bus to the grid. The voltage ride-through coordinated control system receives signals from sensors and outputs gate control signals to the rotor-side converter. (Reference) Figures 1 to 4 As shown, the specific steps include: S1, Data Acquisition, real-time acquisition of the following physical quantities of the doubly-fed motor: stator three-phase voltage Stator three-phase current Rotor three-phase current and the rotor mechanical position angle obtained through position sensors. (Used to calculate rotor electrical angular position) ).

[0020] As a preferred option, the sampling frequency is no less than 10kHz to ensure a rapid response during power grid fault transients.

[0021] This step provides the raw data foundation for subsequent sequence component separation, power calculation, fault diagnosis, and drive signal generation.

[0022] S2, Sequence Component Separation and Preprocessing: The three-phase instantaneous signals acquired in step S1 are input into a sequence component separation and coordinate transformation module based on a dual second-order generalized integrator (DSOGI). This module extracts the positive and negative sequence instantaneous components and their amplitudes of the stator and rotor voltages and currents using quadrature phase shifting and instantaneous symmetrical component methods. Specifically, the underlying filtering algorithm and technical implementation methods for separating positive and negative sequence components using DSOGI are as follows: Step S2.1, Coordinate Transformation. First, the acquired three-phase AC signals (taking stator voltage as an example) are mapped to a two-phase stationary coordinate system using the Clarke transformation matrix to obtain... Axial components and .

[0023] Step S2.2: Second-order generalized integral and orthogonal signal generation. Then... and When input in parallel to two identical second-order generalized integrators, it can achieve the fundamental resonant frequency of the power grid. The input signal undergoes zero steady-state error tracking and bandpass filtering to remove high-frequency harmonics and DC bias, and generates two sets of mutually orthogonal outputs for each signal. Its core closed-loop characteristic transfer function is expressed as: Bandpass in-phase output transfer function: ; Low-pass quadrature output transfer function: ; In the formula, For the input raw signal (corresponding) or ); The fundamental component that is in phase with the original signal after filtering (corresponding to) or ); Lagging the original signal The orthogonal fundamental wave component of the phase; The damping coefficient is used to determine the filter bandwidth and dynamic response speed; The fundamental angular frequency of the power grid is tracked in real time by the phase-locked loop.

[0024] Step S2.3, instantaneous symmetric component method separation. In obtaining... shaft and After considering the in-phase and quadrature components of the axis, a phase-shifting operator is introduced. That is, in the time domain, it manifests as a lag generated by DSOGI. Orthogonal components Based on the instantaneous symmetric component method, a positive / negative order separation matrix is ​​constructed in the stationary coordinate system: Positive sequence voltage vector The computation matrix is: ; Negative sequence voltage vector The computation matrix is: ; Step S2.4: Vector synthesis and amplitude calculation. Finally, the positive-sequence component of the stator voltage is output through the above steps. With negative order components Similarly, the stator current and rotor current are also separated into their respective sequence components through independent DSOGI channels. Based on this, the vector magnitude required for the subsequent impedance reconstruction algorithm is directly calculated using the Pythagorean theorem. For example, the formula for calculating the rotor positive sequence current magnitude is: ; Specifically, the output includes: the positive sequence component of the stator voltage. stator voltage negative sequence component Stator current positive sequence component negative sequence component of stator current , positive sequence component of rotor current The rotor positive sequence current amplitude was calculated. Stator positive sequence current amplitude and stator positive sequence voltage amplitude .

[0025] As a preferred option, the DSOGI module also incorporates an adaptive frequency phase-locked loop to accurately separate the positive and negative sequence components even when the grid frequency shifts.

[0026] It effectively filters out the second harmonic fluctuations under asymmetric faults, providing a smooth and clean feedback for subsequent network control and avoiding pulsation interference in power calculation.

[0027] S3, Steady-state network control – Virtual synchronous generator solution.

[0028] Using the stator voltage positive sequence component separated in step S2 and stator current positive sequence component Calculate the actual electromagnetic active power and reactive power Only the positive-sequence component is used to eliminate second-harmonic ripples under asymmetric faults.

[0029] Active power and reactive power The active-frequency loop and reactive-voltage loop of the virtual synchronous generator are input separately. Internally, they contain key parameter tuning mechanisms to prevent multi-timescale coupling instability. The physical and mathematical model of the control loop, the basis for selecting key control parameters, per-unit processing, and dimensional constraints are recorded as follows: In the reactive-voltage loop (excitation control), the system uses the voltage reference value. With stator positive sequence voltage amplitude The deviation, and the reactive power reference value With actual reactive power The deviation is used to generate the steady-state internal potential amplitude of the virtual synchronous generator through the integral controller. : ; In the formula, both sides of the equation and all deviations have been normalized (dimensionless) under a unified standard. To ensure the dimensionality of the integrator, the physical dimension of the integral gain is the reciprocal of time. The specific boundary constraints are as follows: 1) Voltage droop integral gain This characterizes the depth of reactive current feedforward support provided by the generating unit to grid-side voltage dips. Its lower limit can be determined based on the slope requirements for dynamic reactive power injection during low voltage ride-through (LVRT) as specified in the "Technical Regulations for Wind Farm Connection to Power Systems"; its upper limit is limited by the maximum transient current thermal margin of the rotor-side converter (RSC). Based on both hardware and guideline boundaries, preferably... The value range is constrained to 0.1~0.5. , 2) Reactive power integral gain This characterizes the closed-loop response speed of the reactive power control loop. Based on the time-scale separation principle of the impedance model, to prevent cross-coupling resonance between the outer loop reactive power dynamics and the inner loop electromagnetic transients, the outer loop cutoff frequency is preferably configured to be a fraction of the inner loop bandwidth. The following, following the reverse decoupling criterion of frequency domain, preferably, The value range is set to 10~50 .

[0030] In the active-frequency loop, the actual electromagnetic torque is first calculated. ,in The current virtual angular velocity (initial value equal to the grid's rated angular velocity) The detailed calculation of this loop will be explained in step S6 in conjunction with the compensation torque.

[0031] The beneficial effect of this step is that, through VSG control, the doubly fed motor exhibits voltage source characteristics in steady state, actively supporting the grid voltage and frequency.

[0032] S4, Transient current limiting adaptive judgment and positive sequence transient impedance reconstruction.

[0033] Real-time calculation of rotor positive sequence current amplitude (From step S2), and determine whether it exceeds the set hardware security limit threshold. That is, to judge Is >0 true?

[0034] Scenario A (Fault State, Current Limiting Required): If the threshold is exceeded, it indicates a severe fault in the power grid, such as a deep dip or surge, posing an overcurrent hazard. The system immediately activates the transient impedance reconstruction module. This module adaptively generates a positive-sequence transient complex impedance based on the over-limit error through integral calculation. Includes virtual resistance and virtual reactance : ; ; ; All electrical variables, control parameters, and impedance commands in the aforementioned adaptive reconfiguration control law have been per-unit (pu) based on the doubly-fed motor capacity reference. If the output impedance needs to be converted to actual ohm units, it must be multiplied by the reference impedance. Integral gain and The corresponding physical dimensions are In its physical essence, it characterizes the transient time-domain response speed of the virtual impedance injection loop when a transient overcurrent occurs. The value ranges from 150 to 200. , Values ​​range from 300 to 400 Furthermore, to ensure the completeness of the aforementioned transient impedance reconstruction strategy in both mathematical description and actual control closed loop, a mathematically defined "integrator with conditional control" is introduced. This means that the integral operator is not an unconstrained pure integral, but rather a conditional integral operator defined in a nonlinear bounded space with piecewise time-domain reset and nonlinear anti-saturation limiting constraints. Its mathematical boundary conditions are defined as follows: 1) Limiting constraint: Output impedance is limited to physical safety boundaries. Within the range, the integral accumulation value is forcibly cleared to zero when it is less than 0, and integration stops when it is greater than the maximum design safety impedance; 2) Transient reset constraint: When the system determines that the fault ride-through has ended and the rotor positive sequence current has returned to a steady state below the safe threshold (i.e. Duration exceeding the set threshold When this occurs, the integrator executes a desaturation and zeroing mechanism, driving... and Reset to zero, thereby restoring the unit to steady-state grid operation.

[0035] Case B (Steady state or minor fault, no current limiting required): If the threshold is not exceeded, the system will achieve positive sequence transient impedance. The value is assigned to 0, and the subsequent compensation torque... It is also assigned the value 0 (see step S5).

[0036] It adopts a closed-loop adaptive method based on current over-limit error, replacing the traditional open-loop judgment based on the depth of grid voltage drop. It can cope with multiple complex operating conditions such as deep drop, shallow drop and high voltage surge, and the impedance value is automatically adjusted according to the degree of over-limit, avoiding over-limiting or under-limiting of current.

[0037] S5, Virtual dissipation calculation and compensation torque generation: The system extracts the power dissipation component from the positive-sequence transient impedance reconstructed in step S4, i.e., the virtual resistance component. Based on the stator and rotor voltage equations, flux coupling relationship, and per-unit standard of the doubly-fed motor, it is converted into the equivalent virtual resistance on the stator side. ,when When =0, =0. Combined with the stator positive sequence current amplitude obtained in step S2. Calculate the virtual power dissipation corresponding to the transient virtual impedance. In one alternative implementation, when When representing the effective value of the positive-sequence phase current, the virtual energy dissipation power can be expressed as: ; In another alternative implementation, when When representing the magnitude of the positive-sequence space vector in a two-phase stationary coordinate system, the virtual energy dissipation power can be expressed as: ; The above two expressions are selected based on the definition of the stator positive sequence current amplitude: when the controller uses the effective value of the phase current as the amplitude definition, the first expression is used; when the controller uses the spatial vector amplitude in a two-phase stationary coordinate system as the amplitude definition, the second expression is used.

[0038] The system based on the virtual power dissipation power and the current virtual angular velocity Calculate the equivalent compensation torque corresponding to the virtual power dissipation generated in the virtual loop due to the current limiting operation. : ; As an optional implementation method, when the calculated Exceeding the preset maximum compensation limit At that time, amplitude limiting is applied to reduce the risk of oscillation caused by overcompensation. The settings can be adjusted based on the rated torque, rotor-side converter current margin, and controller sampling period. For example, the settings can be 20% to 50% of the rated torque.

[0039] This invention quantitatively transforms the power dissipation caused by virtual impedance into a compensating torque that can be injected into the swing equation, providing a numerical basis for subsequent power angle compensation. This is one of the key features that distinguishes this invention from traditional current limiting methods.

[0040] S6, the swing equation of a virtual synchronous generator with power angle compensation and correction.

[0041] The system will calculate the compensation torque in step S5. The swing equation of the virtual synchronous generator is introduced as a feedforward reduction term. The modified active-frequency swing equation is as follows: ; In the formula, to ensure the dimensional validity of the differential dynamic equations, all electrical quantities have been per-unit (pu) normalized based on the rated capacity and rated angular velocity of the doubly-fed motor. This is a virtual mechanical torque, provided by the upper-level scheduling or constant power reference; This refers to the actual electromagnetic torque. The compensation torque; The rated angular velocity of the power grid. For virtual angular velocity; time Maintaining the natural time-domain scale, the unit is seconds. Based on the above dimensional isomorphism criterion, the tuning basis for key control parameters is as follows: 1) Virtual moment of inertia This parameter characterizes the system's ability to withstand frequency disturbances during transient active power steps, i.e., it limits the rate of frequency change. In a per-unit system, this parameter is equivalent to a virtual inertial time constant, with its physical dimension being seconds. Its physical reference is mapped to the typical inertial constant of a traditional megawatt-class synchronous generator set. (usually 2~6) According to the per-unit kinetic energy equivalence rule To ensure the virtual frequency fluctuation during transient fault travel It is suppressed at the safe threshold allowed by grid connection (e.g., 2.0). Within ) , preferably, virtual moment of inertia Set to 4~12 .

[0042] 2) Virtual damping coefficient The damping coefficient determines the rate of decay of the low-frequency power angle oscillation of the system after a large disturbance; it is a dimensionless pure number. The damping coefficient is not determined empirically, but rather based on the second-order small-signal characteristic equation of the active power closed loop of the VSG network (i.e., ,in The equivalent synchronization coefficient is calculated by reverse reasoning. To ensure that the system's transient oscillation eigenvalues ​​are distributed in the optimal attenuation region of the left half-plane, the system's damping ratio... The preferred configuration is near the critical damping band. Substituting the optimal damping ratio boundary into the inverse solution of the characteristic equation, and considering the actual converter control delay margin, preferably, the virtual damping coefficient... The value range is set to 20~100.

[0043] Solving this differential equation yields the compensated virtual angular velocity. Further integration generates a virtual internal potential phase: ; Compensating torque This invention is used to compensate for the equivalent power deficit caused by transient virtual impedance connection, reduce the power and torque imbalance in the swing equation, suppress virtual angular velocity fluctuations, and thus reduce the risk of power angle deviation and unit synchronism failure. This is the core technical means of the invention to improve the contradiction between "limited flow and instability".

[0044] S7, dual-sequence voltage command synthesis, the system synthesizes reference voltage vectors in the positive sequence and negative sequence branches respectively.

[0045] Positive sequence branch: The system uses the positive sequence internal potential vector of the virtual synchronous generator obtained in step S3. The positive-sequence transient virtual impedance obtained in step S4 and stator positive sequence current vector First, the equivalent positive sequence voltage reference vector on the stator side is generated: ; Then, based on the stator and rotor voltage equations, flux linkage coupling relationship, and rotor electrical angle position of the doubly-fed generator, the equivalent positive-sequence voltage reference vector on the stator side is converted to the rotor side to obtain the positive-sequence rotor voltage reference vector. This conversion process can be achieved through coordinate transformation in the stator flux linkage oriented coordinate system, the synchronous rotating coordinate system, or the two-phase stationary coordinate system. Thus, the positive-sequence branch maintains both the grid-type voltage source characteristics of the virtual synchronous generator and achieves rotor-side current limiting control through transient virtual impedance voltage drop.

[0046] Negative sequence branch (asymmetric fault handling): The system monitors the amplitude of the negative sequence component of the stator voltage separated in step S2 in real time. | and determine whether it is greater than a preset threshold value. (For example, take 0.05 pu). If | |≤ This indicates that the power grid is in a normal symmetrical operating state or that a three-phase symmetrical fault has occurred. The system does not require negative sequence compensation; the total rotor voltage reference vector can be directly set. .

[0047] If | | This indicates that an asymmetrical fault, such as a single-phase or two-phase short circuit, has occurred in the power grid, and the system immediately activates the negative-sequence suppression branch. A large negative-sequence virtual impedance is injected into this branch. As a preferred option, The value is preset to twice the rated impedance of the doubly-fed motor. Then, the stator-side equivalent negative sequence voltage reference vector is generated: ; Based on the stator and rotor voltage equations, magnetic flux coupling relationship, and rotor electrical angle position of the doubly fed motor, the equivalent negative sequence voltage reference vector on the stator side is converted into a negative sequence rotor voltage reference vector; By superimposing the positive-sequence and negative-sequence reference vectors, the total rotor voltage reference vector is obtained: ; The beneficial effects of this step are as follows: the positive-sequence branch is used to achieve "grid support + adaptive current limiting + power angle compensation", while the negative-sequence branch is used to achieve "asymmetric voltage feedforward + high impedance vibration suppression". This avoids the mutual interference of traditional hybrid control, helps reduce the oscillation risk of the control system under asymmetric faults such as single-phase and two-phase short circuits, and improves the smoothness of control commands. Taking the negative-sequence virtual impedance as twice the rated impedance is a preferred implementation method, which can achieve a better balance between suppressing negative-sequence current and reducing the risk of over-modulation.

[0048] S8, Pulse modulation and drive signal generation: The system utilizes the virtual internal potential phase obtained in step S6. and the rotor mechanical position angle acquired in step S1 The converted rotor electrical angular position Calculate the slip angle ; The slip angle is used to adjust the total rotor voltage reference vector obtained in step S7. A reverse coordinate transformation (from a rotating coordinate system to a stationary coordinate system) is performed. The transformed two-phase stationary coordinate system voltage vectors are then processed by a space vector pulse width modulation (SVPWM) module to generate six pulse control signals for driving the power switches of the rotor-side converter.

[0049] As a preferred option, the switching frequency of the SVPWM module is set to 2 kHz to 10 kHz to balance switching losses and current harmonics.

[0050] The voltage command calculated by the control algorithm is converted into an executable switching signal to achieve precise control of the rotor-side converter.

[0051] S9 controls the cycle and steady-state exit mechanism.

[0052] Steps S1 to S8 above complete one control cycle. The system then enters the next sampling cycle and repeats the above process to achieve adaptive cooperative ride-through control under all time periods and all voltage conditions (including symmetrical faults, asymmetrical faults, voltage dips, and voltage surges).

[0053] Specifically, when step S4 determines that the rotor positive-sequence current amplitude does not exceed the safety threshold (i.e., the system is in steady state or has a minor fault), the positive-sequence transient impedance... and compensation torque All values ​​are zero. At this point, the system maintains normal network support function with minimal control overhead, avoiding control losses caused by unnecessary impedance injection.

[0054] Achieving a smooth switch between steady-state and transient control modes ensures both effective protection and stable support during fault periods, as well as high-efficiency operation under steady-state conditions.

[0055] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A fault ride-through control method for a doubly-fed induction generator based on transient impedance and power angle compensation, characterized in that, include: Obtain the first electrical quantity characterizing the magnitude of the rotor current of the doubly-fed motor; Determine whether the first electrical quantity exceeds a preset safety threshold; When the first electrical quantity exceeds the safety threshold, a transient virtual impedance for limiting rotor overcurrent is adaptively generated based on the over-limit deviation between the first electrical quantity and the safety threshold. The virtual power dissipation corresponding to the transient virtual impedance is calculated based on the virtual resistance component and the stator positive sequence current component in the transient virtual impedance. The compensation torque is generated based on the virtual power dissipation, and the compensation torque is introduced into the swing equation of the virtual synchronous generator to compensate for the power balance shift caused by the transient virtual impedance current limiting process. The voltage reference quantity of the rotor-side converter is generated based on the internal electromotive force output by the virtual synchronous generator, the transient virtual impedance, and the stator positive sequence current component. A drive signal for controlling the rotor-side converter is generated based on the voltage reference value.

2. The method according to claim 1, characterized in that, The acquisition of the first electrical quantity characterizing the magnitude of the doubly-fed motor rotor current includes: collecting the three-phase rotor current and extracting the amplitude of the rotor positive-sequence current using a dual second-order generalized integrator. ; The adaptive generation of transient virtual impedance for limiting rotor overcurrent includes: based on the amplitude With the security threshold Integrate the out-of-limit error to generate a transient virtual resistance. and transient virtual reactance The expression is: ； ； The transient virtual impedance is: ； The integration operation is a conditional integration operation with finite amplitude constraints and reset constraints, and the transient virtual resistance and the transient virtual reactance are limited to a preset impedance range; when the fault crossing ends and Duration exceeding the set threshold At this time, the transient virtual resistance and the transient virtual reactance are reset to zero; and All are integral gains.

3. The method according to claim 1, characterized in that, The step of generating compensation torque based on the virtual power dissipation includes: acquiring the stator three-phase current, extracting the stator positive-sequence current component through a dual second-order generalized integrator, and obtaining the stator positive-sequence current amplitude. Based on the virtual resistance component in the transient virtual impedance The equivalent virtual resistance on the stator side is calculated. According to the stator positive sequence current amplitude Equivalent virtual resistance on the stator side Calculate virtual power dissipation According to the virtual power dissipation power Virtual angular velocity of virtual synchronous generator Calculate the compensation torque The expression is: ； in, The virtual power dissipation power corresponding to the transient virtual impedance, This represents the virtual angular velocity of the virtual synchronous generator.

4. The method according to claim 1, characterized in that, Introducing the compensation torque into the swing equation of the virtual synchronous generator includes: adding the compensation torque as a feedforward subtraction term to the electromagnetic torque term in the swing equation, resulting in the corrected swing equation: ； in For virtual rotational inertia, For virtual mechanical torque, For actual electromagnetic torque, For the compensation torque, This is the virtual damping coefficient. The rated angular velocity of the power grid. This is the virtual angular velocity.

5. The method according to claim 1, characterized in that, The voltage reference quantity of the rotor-side converter is generated based on the internal electromotive force output by the virtual synchronous generator, the transient virtual impedance, and the stator positive sequence current component. Generating drive signals for controlling the rotor-side converter based on the voltage reference includes: converting the steady-state internal potential vector output by the virtual synchronous generator... Subtract the transient virtual impedance With stator positive sequence current vector The product voltage drop yields the equivalent positive-sequence voltage reference vector on the stator side: ； Based on the stator and rotor voltage equations, magnetic flux coupling relationship, and rotor electrical angle position of the doubly fed motor, the equivalent positive sequence voltage reference vector on the stator side is converted into a positive sequence rotor voltage reference vector; The drive signal is generated based on the positive sequence rotor voltage reference vector.

6. The method according to claim 1, characterized in that, The method also includes asymmetric fault handling: The stator three-phase voltage and three-phase current are collected, and the negative sequence component of the stator voltage is separated by a dual second-order generalized integrator. and stator current negative sequence component ; Based on the preset negative sequence virtual impedance Generate the stator-side equivalent negative sequence voltage reference vector: ； Based on the stator and rotor voltage equations, magnetic flux coupling relationship, and rotor electrical angle position of the doubly fed motor, the equivalent negative sequence voltage reference vector on the stator side is converted into a negative sequence rotor voltage reference vector; The negative sequence rotor voltage reference vector is superimposed with the positive sequence rotor voltage reference vector to obtain the rotor-side total voltage reference vector, and the drive signal is generated based on the rotor-side total voltage reference vector; Wherein, the negative sequence virtual impedance The size is 1 to 3 times the rated impedance of the doubly fed motor.

7. The method according to claim 6, characterized in that, Prior to the asymmetric fault handling, the process also includes: separating the negative sequence component of the stator voltage using a dual second-order generalized integrator. Calculate its amplitude | |;When| When the value exceeds a preset threshold, an asymmetrical fault is determined, and the steps of generating an equivalent negative sequence voltage reference vector on the stator side and calculating a negative sequence rotor voltage reference vector are executed.

8. The method according to claim 1, characterized in that, The generation of the drive signal for controlling the rotor-side converter includes: Based on the collected rotor mechanical position angle and the number of pole pairs of a doubly fed motor Calculate the rotor's electrical angular position: Based on the virtual internal potential phase of the virtual synchronous generator With respect to the rotor electrical angular position Calculate slip angle ; The rotor-side total voltage reference vector is transformed inversely using the slip angle to the stationary coordinate system; The driving signal is generated by space vector pulse width modulation.

9. The method according to claim 1, characterized in that, Before generating the voltage reference value for the rotor-side converter based on the internal electromotive force output by the virtual synchronous generator, the transient virtual impedance, and the stator positive sequence current component, the process further includes: The stator three-phase voltage and three-phase current are collected, and the positive-sequence component of the stator voltage is separated by a dual second-order generalized integrator. and stator current positive sequence component ; The actual electromagnetic active power is calculated using the positive sequence components of the stator voltage and stator current. and reactive power To filter out second harmonic fluctuations under asymmetric faults; The active power and reactive power By inputting the active-frequency loop and reactive-voltage loop of the virtual synchronous generator, the magnitude of the steady-state internal potential vector is calculated. and phase .

10. The method according to claim 1, characterized in that, When it is determined that the first electrical quantity does not exceed the safety threshold, the transient virtual impedance is set to zero and the compensation torque is set to zero; when no asymmetric fault is detected, the drive signal is generated based on the positive sequence rotor voltage reference vector; when an asymmetric fault is detected, the drive signal is generated based on the rotor-side total voltage reference vector obtained by superimposing the positive sequence rotor voltage reference vector and the negative sequence rotor voltage reference vector.

Images