Converter fault detection method and computer device

Abstract

This invention discloses a converter fault detection method and a computer device. The converter fault detection method includes: obtaining the output control vector and output feedback vector of the target parameter in the converter subsystem under test; calculating the difference between the output control vector and the output feedback vector to obtain a characteristic deviation vector; and locating the fault in the subsystem under test based on the characteristic deviation vector; wherein the output control vector and the output feedback vector are both determined by the turbine operating data associated with the subsystem under test. Using the technical solution in this invention, the type of converter fault can be automatically located, improving the efficiency of converter fault analysis.

Description

[0001] This application is a divisional application based on the invention with application number 201910568484.2, application date June 27, 2019, applicant Beijing Goldwind Science & Technology Wind Power Equipment Co., Ltd., entitled "Inverter Fault Detection Method and Device, Computer Equipment". Technical Field

[0002] This invention relates to the field of wind power generation technology, and in particular to a converter fault detection method and computer equipment. Background Technology

[0003] During operation, the converter of a wind turbine generator set will inevitably report faults due to abnormalities. Currently, fault alarms can only indicate that a problem has occurred in the operation of the converter, but cannot pinpoint the location and cause of the problem. Users need to analyze the fault information to locate the fault.

[0004] However, this method of fault location, which requires user participation in analysis, is highly dependent on the user's skills and experience. Among all converter fault types, non-control faults such as electrical and cooling faults are relatively easy to analyze, while control faults (such as voltage and current faults) often involve complex operating principles, making analysis very difficult. Generally, only industry experts can quickly arrive at conclusions. However, the vast majority of maintenance personnel are far from reaching expert levels. This leads to low fault analysis efficiency, long downtime due to faults, impact on overall power generation, and damage to customer interests. Summary of the Invention

[0005] This invention provides a converter fault detection method and computer equipment that can automatically locate the type of converter fault and improve the efficiency of converter fault analysis.

[0006] In a first aspect, embodiments of the present invention provide a converter fault detection method, the method comprising: Obtain the output control vector and output feedback vector of the target parameters in the converter subsystem under test; The difference between the output control vector and the output feedback vector is calculated to obtain the characteristic deviation vector; Fault location is performed on the subsystem under test based on the characteristic deviation vector; The output control vector and the output feedback vector are both determined by the wind turbine operating data associated with the subsystem under test.

[0007] In one possible implementation of the first aspect, the step of fault location of the subsystem under test based on the feature deviation vector includes: if the value of at least one element in the feature deviation vector is not equal to 0, then it is determined that the subsystem under test has an internal fault; if the value of all elements in the feature deviation vector is 0, then it is determined that the subsystem under test has an external fault.

[0008] In one possible implementation of the first aspect, the target parameter is voltage; the subsystem under test is a grid-side subsystem; the output control vector is determined by the DC bus voltage and the grid-side PWM signal; and the output feedback vector is determined by the grid-side voltage and the grid-side current.

[0009] In one possible implementation of the first aspect, the expression for the characteristic deviation vector Re of the network-side subsystem is: Re Where P is a preset feature matrix, ea, eb, ec are the three-phase voltages on the grid side, ia, ib, ic are the three-phase currents on the grid side, udc is the DC bus voltage, sa, sb, sc are the three-phase PWM signals on the grid side, R is the grid-side line resistance, and L is the grid-side reactance.

[0010] In one possible implementation of the first aspect, The steps for fault location of the subsystem under test based on the characteristic deviation vector include: If r1=0, r2=0, and r3=0, then it is determined that an external fault has occurred in the network-side subsystem. If r1≠0, r2=0, and r3≠0, then it is determined that a phase a branch fault has occurred in the network-side subsystem. If r1≠0, r2≠0, and r3=0, then it is determined that a phase b branch fault has occurred in the network-side subsystem. If r1=0, r2≠0, and r3≠0, then it is determined that a c-phase branch fault has occurred in the network-side subsystem.

[0011] If r1≠0, r2≠0, and r3≠0, then it is determined that the DC bus voltage sensor has failed or that at least two phases of the three-phase branch have failed simultaneously.

[0012] In one possible implementation of the first aspect, the target parameter is voltage; the subsystem under test is the machine-side subsystem; the output control vector is determined by the generator back EMF and the machine-side current; and the output feedback vector is determined by the machine-side voltage.

[0013] In one possible implementation of the first aspect, the expression for the characteristic deviation vector Rg of the machine-side subsystem is: Where Es is the generator back EMF, isd is the d-axis component of the generator-side current, isq is the q-axis component of the generator-side current, usd is the d-axis component of the generator-side voltage, usq is the q-axis component of the generator-side voltage, and Rs is the generator-side line resistance. Lsd is the electrical angular velocity, Lsq is the d-axis component of the machine-side inductance, and Lsq is the q-axis component of the machine-side inductance.

[0014] In one possible implementation of the first aspect, the step of fault location of the subsystem under test based on the characteristic deviation vector includes: If r4=0 and r5=0, then it is determined that an external fault has occurred in the machine-side subsystem; If at least one of r4 and r5 is not 0, then an internal fault has occurred in the machine-side subsystem. The internal fault of the machine-side subsystem includes at least one of the following faults: a fault in the line between the machine-side subsystem and the generator, a fault in the machine-side current sensor circuit, and a fault in the machine-side drive circuit.

[0015] In one possible implementation of the first aspect, the subsystem under test is a grid-side subsystem or a machine-side subsystem; after determining that an external fault has occurred in the subsystem under test, the method further includes: determining the type of the external fault; if the current value of the grid-side voltage or machine-side voltage exceeds a limit threshold, determining that the grid-side voltage or machine-side voltage has experienced a value over-limit fault; if the harmonic amplitude of the grid-side voltage or machine-side voltage after filtering the fundamental frequency exceeds a limit harmonic amplitude, determining that the grid-side voltage or machine-side voltage has experienced a harmonic over-limit fault; if the current time of the grid-side voltage or machine-side voltage and the previous time... If the phase difference between two points in time exceeds the preset phase difference range, a phase change fault is determined to have occurred in the grid-side voltage or the machine-side voltage. If the current frequency of the grid-side voltage or the machine-side voltage exceeds the preset frequency range, a frequency over-limit fault is determined to have occurred in the grid-side voltage or the machine-side voltage. If the absolute value of the difference between the grid-side power and the machine-side power is greater than a preset threshold, a fault is determined to have occurred in the DC bus voltage. If both the grid-side voltage and the DC bus voltage are normal, a fault is determined to have occurred in the grid-side PWM signal. If both the machine-side voltage and the DC bus voltage are normal, a fault is determined to have occurred in the machine-side PWM signal.

[0016] In one possible implementation of the first aspect, the target parameter is current; the subsystem under test is a DC bus subsystem; the output control vector is determined by the machine-side current and the machine-side PWM signal; and the output feedback vector is determined by the DC bus voltage, the grid-side current, and the grid-side PWM signal.

[0017] In one possible implementation of the first aspect, the expression for the characteristic deviation vector Rd of the DC bus subsystem is: Rd Among them, s ga s gb s gc This is the machine-side PWM signal. , , For the three-phase current on the machine side, , , This is the grid-side three-phase PWM signal. , , UDC represents the three-phase current on the grid side, and udc represents the DC bus voltage. This is the DC bus capacitor.

[0018] In one possible implementation of the first aspect, the step of fault location of the subsystem under test based on the characteristic deviation vector includes: if Rd = 0, then it is determined that an external fault has occurred in the DC bus subsystem, the external fault including at least one of the following faults: DC bus voltage fault, machine-side current fault, and grid-side current fault; if Rd ≠ 0, then it is determined that an internal fault has occurred in the DC bus subsystem, the internal fault of the DC bus subsystem including at least one of the following faults: DC bus capacitor damage fault and PWM modulation error fault.

[0019] In a second aspect, embodiments of the present invention provide a converter fault detection device, the device comprising: The target parameter vector acquisition module is used to obtain the output control vector and output feedback vector of the target parameters in the converter test subsystem; The feature deviation vector calculation module is used to calculate the difference between the output control vector and the output feedback vector to obtain the feature deviation vector; The fault location module is used to locate faults in the subsystem under test based on the feature deviation vector. The output control vector and the output feedback vector are both determined by the wind turbine operating data associated with the subsystem under test.

[0020] In one possible implementation of the second aspect, the device is located in the main controller or converter controller of the wind turbine generator set.

[0021] Thirdly, embodiments of the present invention provide a computer device having a program stored thereon, which, when executed by a processor, implements the converter fault detection method as described above.

[0022] In this embodiment of the invention, the difference between the output control vector and the output feedback vector of the target parameter in the subsystem under test is used as a feature deviation vector to characterize and quantify the signal fluctuations caused by the fault in the factor system, thereby realizing the fault location of the converter.

[0023] Compared with existing technologies that require user participation in analysis to locate faults, the embodiments of the present invention do not rely on the user's ability and experience. Instead, they can process the wind turbine operating data associated with the subsystem under test to obtain the output control vector and output feedback vector of the target parameter in the faulty subsystem. The characteristic deviation vector obtained by subtracting the two is used to characterize and quantify the signal fluctuations caused by the system fault, thereby automatically locating the type of subsystem fault and improving the analysis efficiency of converter faults. Attached Figure Description

[0024] The invention can be better understood from the following description of specific embodiments of the invention in conjunction with the accompanying drawings, wherein the same or similar reference numerals denote the same or similar features.

[0025] Figure 1 A schematic diagram of the grid connection structure of a wind turbine generator set designed for an embodiment of the present invention.

[0026] Figure 2 This is a flowchart illustrating a converter fault detection method according to an embodiment of the present invention. Figure 3 A flowchart illustrating a converter fault detection method provided in another embodiment of the present invention; Figure 4 This is a schematic diagram of signals associated with faults in the network-side subsystem provided in an embodiment of the present invention; Figure 5 A logic block diagram for fault diagnosis based on the network-side subsystem provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of signals associated with faults in the engine-side subsystem provided in an embodiment of the present invention; Figure 7 A logic block diagram for fault diagnosis based on the machine-side subsystem provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the converter fault detection device provided in an embodiment of the present invention. Detailed Implementation

[0027] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. Numerous specific details are set forth in the following detailed description in order to provide a thorough understanding of the invention.

[0028] Figure 1 This is a schematic diagram of the grid connection structure of a wind turbine generator set according to an embodiment of the present invention.

[0029] like Figure 1 As shown, a converter 100 is installed between the wind turbine generator set and the power grid.

[0030] The converter 100 comprises, from the generator side to the grid side, a generator-side subsystem 101, a DC bus subsystem 102, and a grid-side subsystem 103. The generator-side subsystem 101 includes a full-bridge structure composed of six IGBTs, used to rectify the three-phase AC power generated by the wind turbine generator. The DC bus subsystem 102 includes a braking unit (not shown) and a DC bus capacitor C. The braking unit dissipates active power to maintain a stable voltage across the DC bus capacitor C. The grid-side subsystem 103 includes a full-bridge structure composed of six IGBTs, used to convert the rectified DC power back into three-phase AC power and integrate it into the power grid.

[0031] This invention provides a converter fault detection method and computer equipment that can automatically locate the type of converter fault without the need for industry experts to participate in the analysis, thereby improving the efficiency of converter fault analysis.

[0032] Figure 2 This is a schematic flowchart of a converter fault detection method provided in an embodiment of the present invention. Figure 2 As shown, the converter fault detection method includes steps 201 to 203.

[0033] In step 201, the output control vector and output feedback vector of the target parameters in the converter under test subsystem are obtained.

[0034] Figure 1 The three subsystems under test of the converter shown are: machine-side subsystem 101, DC bus subsystem 102, and grid-side subsystem 103.

[0035] Taking the grid-side subsystem 103 as an example, the target parameter can be voltage. The output control vector of voltage in the grid-side subsystem 103 can be determined by the wind turbine operating parameters associated with the grid-side subsystem 103: DC bus voltage and grid-side PWM signal. The output feedback vector of voltage can be determined by the wind turbine operating parameters associated with the grid-side subsystem 103: grid-side voltage and grid-side current.

[0036] Taking the subsystem under test as the machine-side subsystem 101 as an example, the target parameter can be voltage. The output control vector of voltage in the machine-side subsystem 101 can be determined by the wind turbine operating parameters associated with the machine-side subsystem 101: the back EMF of the generator and the machine-side current. The output feedback vector of voltage can be determined by the wind turbine operating parameters associated with the machine-side subsystem 101: the machine-side voltage.

[0037] Taking the DC bus subsystem 102 as an example, the target parameter can be voltage. The output control vector of the DC bus subsystem 102 current can be determined by the wind turbine operating parameters associated with the DC bus subsystem 102: the turbine-side current and the turbine-side PWM signal. The output feedback vector of the current can be determined by the wind turbine operating parameters associated with the turbine-side subsystem 101: the DC bus voltage, the grid-side current, and the grid-side PWM signal.

[0038] In step 202, the difference between the output control vector and the output feedback vector is calculated to obtain the characteristic deviation vector.

[0039] In step 203, the fault location of the subsystem under test is performed based on the characteristic deviation vector.

[0040] Because all signals of each subsystem of the converter are in a stable state during normal operation (DC signal is stable, AC signal is relatively stable), the converter or the controller of each subsystem will record the signals or actions in real time during operation. However, when any subsystem of the converter fails, the DC or AC signal associated with that subsystem will fluctuate.

[0041] In this embodiment of the invention, the difference between the output control vector and the output feedback vector of the target parameter in the subsystem under test is used as a feature deviation vector to characterize and quantify the signal fluctuations caused by the fault in the factor system, thereby realizing the fault location of the converter.

[0042] Compared with existing technologies that require user participation in analysis to locate faults, the embodiments of the present invention do not rely on the user's ability and experience. Instead, they can process the wind turbine operating data associated with the subsystem under test to obtain the output control vector and output feedback vector of the target parameter in the faulty subsystem. The characteristic deviation vector obtained by subtracting the two is used to characterize and quantify the signal fluctuations caused by the system fault, thereby automatically locating the type of subsystem fault and improving the analysis efficiency of converter faults.

[0043] Figure 3 This is a flowchart illustrating a converter fault detection method according to another embodiment of the present invention. Figure 3 and Figure 2 The difference is that, Figure 2 Step 203 can be further refined as follows: Figure 3 Steps 2031 and 2032 in the process.

[0044] In step 2031, if the value of at least one element in the characteristic deviation vector is not equal to 0, then it is determined that the subsystem under test has an internal fault.

[0045] Internal faults refer to malfunctions in the components or circuits of the subsystem itself, including open circuits, short circuits, sensor circuit faults, and drive circuit faults.

[0046] In step 2032, if the value of all elements in the characteristic deviation vector is 0, then it is determined that an external fault has occurred in the subsystem under test.

[0047] External faults refer to faults in external signals associated with the subsystem, including: grid-side voltage or machine-side voltage faults, DC bus voltage faults, and grid-side or PWM signal faults.

[0048] In this embodiment of the invention, the number of elements contained in the feature deviation vector is related to the coordinate system used for calculation. If the abc coordinate system is used, the number of elements contained in the feature deviation vector is 3. If the dq coordinate system is used, the number of elements contained in the feature deviation vector is 2.

[0049] In this embodiment of the invention, the feature deviation vector can characterize and quantify signal fluctuations caused by system failures. If all elements in the feature deviation vector are 0, it indicates that the components or circuits of the subsystem itself are not faulty, and it can be determined that the subsystem has experienced an external failure, i.e., the external signal associated with the subsystem has failed. Conversely, if at least one element in the feature deviation vector is not equal to 0, it indicates that the components or circuits of the subsystem itself have failed, i.e., the subsystem has experienced an internal failure.

[0050] The following section uses various subsystems in a converter as examples to provide a detailed explanation of the fault location method in this embodiment of the invention.

[0051] Figure 4 This is a schematic diagram of signals associated with faults in the network-side subsystem provided in an embodiment of the present invention.

[0052] The following is combined with Figure 4 The control principle of the grid-side subsystem 103 is explained below: DC bus voltage sensor 401, grid-side current sensor 403 and grid-side voltage sensor 404 collect main circuit signals and transmit them to grid-side controller 405. After processing the signals, grid-side controller 405 generates PWM signals and sends them to grid-side driver 402. Grid-side driver 402 drives IGBT to work. The operation of IGBT causes changes in the signals in the main circuit. The changed signals cause the controller to adjust the PWM, and so on.

[0053] During normal operation, all signals are in a stable state, the DC signal is stable, and the AC signal is relatively stable. During operation, the grid-side controller 405 will record the signals or actions in real time and input these data into the characteristic deviation vector Re calculation model of the grid-side subsystem 103.

[0054] In one example, the expression for the feature bias vector Re is: Re (1) Where P is a preset feature matrix, ea, eb, ec are the three-phase voltages on the grid side (e_abc), ia, ib, ic are the three-phase currents on the grid side (i_abc), udc is the DC bus voltage (u_dc), sa, sb, sc are the three-phase PWM signals on the grid side, R is the grid side line resistance, and L is the grid side reactance.

[0055] In equation (1), the grid-side three-phase voltage (e_abc) and grid-side three-phase current (i_abc) form the output feedback vector of the grid-side subsystem 103, and the DC bus voltage (u_dc) and grid-side three-phase PWM signal form the output control vector of the grid-side subsystem 103.

[0056] According to equation (1), when the network-side subsystem 103 is operating normally, the output feedback vector and the output control vector should be equal, that is, the value of each element in the characteristic deviation vector Re is 0; when the network-side subsystem 103 has an internal fault, the output feedback vector and the output control vector are no longer equal, and the value of each element in the characteristic deviation vector Re will not all be 0.

[0057] In one example, the expression for the feature matrix P is: (2) The fault location method of the network-side subsystem 103 is explained below in conjunction with equations (1) and (2): If r1=0, r2=0, r3=0, it means that the output control vector and output feedback vector of the grid-side subsystem are equal, that is, the output value and feedback value of the voltage are consistent, indicating that the grid-side subsystem is operating normally, and it can be determined that the grid-side subsystem has an external fault. Since the A-phase signal is related to the equations corresponding to r1 and r3, but not to the equation corresponding to r2, if r1≠0, r2=0, and r3≠0, then it can be determined that a phase a branch fault has occurred in the grid-side subsystem. Since the B-phase signal is related to the equations corresponding to r1 and r2, but not to the equation corresponding to r3, if r1≠0, r2≠0, and r3=0, then it can be determined that a B-phase branch fault has occurred in the grid-side subsystem. Since the B-phase signal is related to the equations corresponding to r2 and r3, but not to the equation corresponding to r1, if r1=0, r2≠0, and r3≠0, then it can be determined that a c-phase branch fault has occurred in the grid-side subsystem.

[0058] If r1≠0, r2≠0, and r3≠0, it indicates that a fault has occurred that simultaneously affects the three-phase branches. This can be determined that the DC bus voltage sensor has failed or that at least two of the three-phase branches have failed simultaneously.

[0059] In one example, when the sum of the grid-side voltage and the grid-side current is zero, equation (1) can be simplified to the following form: Re (3) In one example, the expression for equation (3) based on the dq coordinate system is: Re (4) Where ed is the d-axis component of the grid-side voltage, eq is the q-axis component of the grid-side voltage, id is the d-axis component of the grid-side current, iq is the q-axis component of the grid-side voltage, sd is the d-axis component of the grid-side PWM signal, and sq is the q-axis component of the grid-side PWM signal. ω represents the angular velocity of the grid-side voltage.

[0060] Furthermore, if an external fault occurs in the network-side subsystem 103, the type of external fault can be diagnosed by analyzing signal characteristics. The specific diagnostic method is as follows: a1. If the current value of the grid-side voltage exceeds the limit threshold, then a grid-side voltage over-limit fault is determined. For example, if the limit threshold is 650V, and the current value exceeds 650V, then a grid-side voltage over-limit fault has occurred.

[0061] b1. If the harmonic amplitude of the grid-side voltage after filtering the fundamental frequency exceeds the limit harmonic amplitude, then the grid-side voltage is determined to have a harmonic over-limit fault.

[0062] When the grid-side voltage is normal, FFT (Fast Fourier Transform) analysis of the grid-side voltage only shows the fundamental component. However, if the grid-side voltage contains harmonics, the FFT analysis results will show harmonic components other than the fundamental component, and the amplitude of the harmonics will not be zero. Therefore, when the harmonic amplitude exceeds the limit harmonic amplitude, it can be determined that the grid-side voltage has a harmonic over-limit fault.

[0063] c1. If the phase difference between the current moment and the previous moment of the grid-side voltage exceeds the preset phase difference range, then it is determined that a phase change fault has occurred in the grid-side voltage.

[0064] When the grid-side voltage is normal, the phase information obtained by the phase-locked loop is uniformly increasing, that is, the phase difference between the current moment and the previous moment is a known constant. However, once a sudden change occurs, the phase will jump, and the phase difference will not be equal to the constant. Therefore, by comparing the phase difference with the constant, it is possible to determine whether a phase sudden change fault has occurred in the grid-side voltage.

[0065] d1. If the current frequency of the grid-side voltage exceeds the preset frequency range, it is determined that the grid-side voltage has experienced a frequency over-limit fault.

[0066] When the grid-side voltage is normal, the frequency of the grid-side voltage is within a known frequency range. The frequency value can be calculated through a phase-locked loop. When the frequency value exceeds this frequency range, it can be determined that the grid-side voltage has experienced a frequency over-limit fault.

[0067] e1. If the absolute value of the difference between the grid-side power and the generator-side power is greater than the preset threshold, that is, the grid-side power and the generator-side power are mismatched, then it is determined that the DC bus voltage has failed.

[0068] f1. If both the grid-side voltage and the DC bus voltage are normal, the grid-side PWM signal can be determined to be faulty by the process of elimination.

[0069] As described above, the embodiments of the present invention can first determine whether the network-side subsystem has an internal or external fault based on the feature deviation vector, and then further locate the internal and external fault types, thereby realizing the step-by-step location of the fault cause of the network-side subsystem, thus avoiding the problem of inaccurate location caused by the confusion of internal and external causes.

[0070] To facilitate understanding by those skilled in the art, the specific implementation of the fault location method of this invention will be described below using the network-side subsystem 103 as an example.

[0071] Figure 5 A logic block diagram of fault diagnosis based on the network-side subsystem provided in this embodiment of the invention is shown below. Figure 5 As shown, the fault diagnosis process of the network-side subsystem 103 is as follows: S1. Input the signal into the grid-side fault diagnosis system; S2. Calculate the voltage output control vector based on the DC bus voltage and the grid-side PWM signal, and calculate the voltage output feedback vector based on the grid-side voltage and the grid-side current (see equation (1)). S3. Subtract the voltage output control vector from the output feedback vector to generate the characteristic deviation vector (see equation (1)). S4. Analyze the feature deviation vector to achieve fault location, such as determining whether the fault is internal or external. S5. Further analyze and determine the cause of the fault, such as whether it is an internal or external fault.

[0072] Figure 6 This is a schematic diagram of signals associated with faults in the engine-side subsystem provided in an embodiment of the present invention.

[0073] The following is combined with Figure 6 Explanation of the control principle of the machine-side subsystem 101: The rotor position sensor 406, the machine-side current sensor 407, and the DC bus voltage sensor 401 collect the main circuit signals and transmit them to the machine-side controller 409. After processing the signals, the machine-side controller 409 generates a PWM signal and sends it to the machine-side driver 408. The machine-side driver 408 drives the IGBT to work. The movement of the IGBT will cause changes in the signals in the main circuit. The changed signals will cause the controller to adjust the PWM, and so on.

[0074] During normal operation, the DC signal is stable and the AC signal is relatively stable. During operation, the machine-side controller 409 will record the signals or actions in real time and input these data into the characteristic deviation vector Rg calculation model of the machine-side subsystem 101.

[0075] In one example, the expression for the characteristic deviation vector Rg of the machine-side subsystem is: (5) (6) (7) in, It is the back electromotive force of the generator. It is the rotor's electrical angular velocity. It is the rotor flux linkage, and Theta is the rotor position angle. For a given generator, It is a known constant. Angle information can be obtained by the motor rotor position sensor 406. And then Differentiating, we get: isd is the d-axis component of the machine-side current, isq is the q-axis component of the machine-side current, usd is the d-axis component of the machine-side voltage, usq is the q-axis component of the machine-side voltage, and Rs is the machine-side line resistance. Lsd is the electrical angular velocity, Lsq is the d-axis component of the machine-side inductance, and Lsq is the q-axis component of the machine-side inductance.

[0076] In equation (5), the generator back EMF Es, the d-axis component isd of the generator-side current, and the q-axis component isq form the output control vector of the generator-side subsystem 101, while the d-axis component usd and the q-axis component usq of the generator-side voltage form the output feedback vector of the generator-side subsystem 101. According to equation (5), when the machine-side subsystem 101 is operating normally, the output feedback vector and the output control vector should be equal, that is, the value of each element of the characteristic deviation vector Rg is 0; when the machine-side subsystem 101 has an internal fault, the output feedback vector and the output control vector are no longer equal, and the value of each element of the characteristic deviation vector Rg will not all be 0.

[0077] The fault location method for the engine-side subsystem 101 is explained below in conjunction with equation (5): If r4=0 and r5=0, then it is determined that an external fault has occurred in the machine-side subsystem 101; If at least one of r4 and r5 is not 0, then an internal fault is determined to have occurred in the machine-side subsystem 101.

[0078] Among them, internal faults of the machine-side subsystem include at least one of the following faults: a fault in the line between the machine-side subsystem and the generator (such as a short circuit or short circuit), a fault in the machine-side current sensor circuit, and a fault in the machine-side drive circuit.

[0079] Furthermore, if an external fault occurs in the machine-side subsystem 101, the type of external fault can be diagnosed by analyzing signal characteristics. The specific diagnostic method is as follows: a2. If the current value of the generator-side voltage exceeds the limit threshold, then an over-limit fault has occurred in the generator-side voltage. For example, if the limit threshold is 650V, and the current value exceeds 650V, then an over-limit fault has occurred in the grid-side voltage.

[0080] b2. If the harmonic amplitude of the machine-side voltage after filtering the fundamental frequency exceeds the limit harmonic amplitude, then the machine-side voltage is determined to have a harmonic over-limit fault.

[0081] When the machine-side voltage is normal, FFT (Fast Fourier Transform) analysis of the machine-side voltage only shows the fundamental component. However, if the machine-side voltage contains harmonics, the FFT analysis results will show harmonic components other than the fundamental component, and the amplitude of the harmonics will not be zero. Therefore, when the harmonic amplitude exceeds the limit harmonic amplitude, it can be determined that the machine-side voltage has a harmonic over-limit fault.

[0082] c2. If the phase difference between the current moment and the previous moment of the machine-side voltage exceeds the preset phase difference range, then it is determined that a phase change fault has occurred in the machine-side voltage.

[0083] When the machine-side voltage is normal, the phase information obtained by the phase-locked loop is uniformly increasing, that is, the phase difference between the current moment and the previous moment is a known constant. However, once a sudden change occurs, the phase will jump, and the phase difference will not be equal to the constant. Therefore, by comparing the phase difference with the constant, it is possible to determine whether a phase sudden change fault has occurred in the machine-side voltage.

[0084] d2. If the current frequency of the machine-side voltage exceeds the preset frequency range, then the machine-side voltage is determined to have an over-frequency fault.

[0085] When the generator-side voltage is normal, the generator-side voltage frequency is within a known frequency range. The frequency value can be calculated using a phase-locked loop. When the frequency value exceeds this frequency range, it can be determined that the generator-side voltage has experienced a frequency over-limit fault.

[0086] e2. If the absolute value of the difference between the grid-side power and the generator-side power is greater than the preset threshold, that is, the grid-side power and the generator-side power are mismatched, then it is determined that the DC bus voltage has failed.

[0087] f2. If both the generator side voltage and the DC bus voltage are normal, the grid side PWM signal can be determined to be faulty by the process of elimination.

[0088] Figure 7 This is a logic block diagram for fault diagnosis based on the machine-side subsystem provided in an embodiment of the present invention. (See diagram below.) Figure 7 As shown, the fault diagnosis process of the engine-side subsystem 101 is as follows: S1. Input the signal into the machine-side fault diagnosis system; S2. Calculate the output control vector of the voltage based on the DC bus voltage and the machine-side PWM signal; calculate the rotor angular velocity based on the rotor position angle; calculate the back EMF of the generator based on the rotor flux linkage and the rotational speed angular velocity; and calculate the output feedback vector of the voltage based on the fault point and the machine-side current (see equations (5)-(7)). S3. Subtract the voltage output control vector from the output feedback vector to generate the characteristic deviation vector (see equation (5)); S4. Analyze the feature deviation vector to achieve fault location, such as determining whether the fault is internal or external. S5. Further analyze and determine the cause of the fault, such as whether it is an internal or external fault.

[0089] Similarly, the expression for the characteristic deviation vector Rd of the DC bus subsystem 102 is: Rd (8) Among them, s ga s gb s gc This is the machine-side PWM signal. , , For the three-phase current on the machine side, , , This is the grid-side three-phase PWM signal. , , UDC represents the three-phase current on the grid side, and udc represents the DC bus voltage. This is the DC bus capacitor.

[0090] In equation (8), the machine-side three-phase PWM signal (s ga s gb s gc ) and machine-side three-phase current ( , , ) forms the output control vector of the DC bus subsystem 102, and the grid-side three-phase PWM signal ( , , ), grid-side three-phase current ( , , The DC bus voltage (udc) forms the output feedback vector of the DC bus subsystem 102.

[0091] According to equation (8), when the DC bus subsystem 102 is operating normally, the output feedback vector and the output control vector should be equal, that is, the value of each element in the characteristic deviation vector Rd is 0; when the DC bus subsystem 102 has an internal fault, the output feedback vector and the output control vector are no longer equal, and the value of each element in the characteristic deviation vector Rd will not all be 0.

[0092] The fault location method for DC bus subsystem 102 is explained below in conjunction with equation (8): If Rd = 0, then an external fault has occurred in the DC bus subsystem 102. The external fault of the DC bus subsystem 102 may be at least one of the following faults: DC bus voltage fault, machine-side current fault, and grid-side current fault.

[0093] If Rd≠0, then it is determined that an internal fault has occurred in the DC bus subsystem. The internal fault of the DC bus subsystem 102 may be at least one of the following faults: DC bus capacitor damage fault and PWM modulation error fault.

[0094] Figure 8 This is a schematic diagram of the converter fault detection device provided in an embodiment of the present invention. Figure 2 The explanations and descriptions in the text can be applied to this embodiment. For example... Figure 8 As shown, the converter fault detection device includes: a target parameter vector acquisition module 801 (which has the function corresponding to step 201), a feature deviation vector calculation module 802 (which has the function corresponding to step 202), and a fault location module 803 (which has the function corresponding to step 203).

[0095] The target parameter vector acquisition module 801 is used to obtain the output control vector and output feedback vector of the target parameters in the converter subsystem under test. Both the output control vector and the output feedback vector are determined by the wind turbine operating data associated with the subsystem under test.

[0096] The feature deviation vector calculation module 802 is used to calculate the difference between the output control vector and the output feedback vector to obtain the feature deviation vector.

[0097] The fault location module 803 is used to locate faults in the subsystem under test based on the feature deviation vector.

[0098] As described above, in order to achieve fault location of the converter, the embodiments of the present invention do not rely on the user's ability and experience. Instead, the target parameter vector acquisition module 801 can process the wind turbine operation data associated with the subsystem under test to obtain the output control vector and output feedback vector of the target parameter in the faulty subsystem. The characteristic deviation vector obtained by the characteristic deviation vector calculation module 802 is used to characterize and quantify the signal fluctuation caused by the system fault, thereby enabling the fault location module 803 to automatically locate the type of subsystem fault and improve the analysis efficiency of converter faults.

[0099] It should be noted that the converter fault detection device in the embodiments of the present invention can be set in the main controller or converter controller of the wind turbine generator set, so that no hardware changes are required. It can also be a logic device with independent computing function, which is not limited here.

[0100] This invention also provides a computer device on which a program is stored, and when the program is executed by a processor, it implements the converter fault detection method as described above.

[0101] It should be clarified that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. For the device embodiments, relevant parts can be referred to the description section of the method embodiments. The embodiments of the present invention are not limited to the specific steps and structures described above and shown in the figures. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the embodiments of the present invention. Furthermore, for the sake of brevity, detailed descriptions of known methods and techniques are omitted here.

[0102] The functional blocks shown in the above structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this embodiment are programs or code segments used to perform the required tasks. The programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0103] The embodiments of the present invention may be implemented in other specific forms without departing from their spirit and essential characteristics. For example, the algorithm described in a particular embodiment may be modified, while the system architecture does not depart from the basic spirit of the embodiments of the present invention. Therefore, the present embodiments are to be regarded as exemplary rather than limiting in all respects, and the scope of the embodiments of the present invention is defined by the appended claims rather than the foregoing description, and all changes falling within the meaning of the claims and their equivalents are thus included within the scope of the embodiments of the present invention.

Claims

1. A converter fault detection method, characterized in that, include: Obtain the output control vector and output feedback vector of the target parameters in the converter subsystem under test; The difference between the output control vector and the output feedback vector is calculated to obtain the characteristic deviation vector; The fault location of the subsystem under test is performed based on the characteristic deviation vector. The output control vector and the output feedback vector are both determined by the wind turbine operating data associated with the subsystem under test; When the target parameter is voltage and the subsystem under test is a grid-side subsystem, the output control vector is determined by the DC bus voltage and the grid-side pulse width modulation (PWM) signal. The output feedback vector is determined by the grid-side voltage and the grid-side current; The expression for the characteristic deviation vector Re of the network-side subsystem is: Re Where P is a preset feature matrix, ea, eb, ec are the three-phase voltages on the grid side, ia, ib, ic are the three-phase currents on the grid side, udc is the DC bus voltage, sa, sb, sc are the three-phase PWM signals on the grid side, R is the grid side line resistance, and L is the grid side reactance. ； The step of locating the fault in the subsystem under test based on the feature deviation vector includes: If r1=0, r2=0, and r3=0, then it is determined that an external fault has occurred in the network-side subsystem. If r1≠0, r2=0, and r3≠0, then it is determined that a phase a branch fault has occurred in the network-side subsystem. If r1≠0, r2≠0, and r3=0, then it is determined that a phase b branch fault has occurred in the network-side subsystem. If r1=0, r2≠0, and r3≠0, then it is determined that a c-phase branch fault has occurred in the network-side subsystem. If r1≠0, r2≠0, r3≠0, then it is determined that the DC bus voltage sensor has failed or at least two phases of the three-phase branch have failed simultaneously. When the target parameter is voltage and the subsystem under test is a machine-side subsystem, the output control vector is determined by the generator's back EMF and the machine-side current. The output feedback vector is determined by the machine-side voltage; The expression for the characteristic deviation vector Rg of the machine-side subsystem is: Where Es is the generator back EMF, isd is the d-axis component of the generator-side current, isq is the q-axis component of the generator-side current, usd is the d-axis component of the generator-side voltage, usq is the q-axis component of the generator-side voltage, and Rs is the generator-side line resistance. Lsd is the electrical angular velocity, Lsq is the d-axis component of the machine-side inductance, and Lsq is the q-axis component of the machine-side inductance. The step of locating the fault in the subsystem under test based on the feature deviation vector includes: If r4=0 and r5=0, then it is determined that an external fault has occurred in the machine-side subsystem; If at least one of r4 and r5 is not 0, then it is determined that the machine-side subsystem has an internal fault. The internal fault of the machine-side subsystem includes at least one of the following faults: a fault in the line between the machine-side subsystem and the generator, a fault in the machine-side current sensor circuit, and a fault in the machine-side drive circuit. When the target parameter is current and the subsystem under test is a DC bus subsystem, the output control vector is determined by the machine-side current and the machine-side PWM signal. The output feedback vector is determined by the DC bus voltage, grid-side current, and grid-side PWM signal. The expression for the characteristic deviation vector Rd of the DC bus subsystem is: Rd Among them, s ga s gb s gc This is the machine-side PWM signal. , , For the three-phase current on the machine side, , , This is the grid-side three-phase PWM signal. , , UDC represents the three-phase current on the grid side, and udc represents the DC bus voltage. For DC bus capacitors; The step of locating the fault in the subsystem under test based on the feature deviation vector includes: If Rd = 0, then it is determined that an external fault has occurred in the DC bus subsystem. The external fault includes at least one of the following faults: DC bus voltage fault, machine-side current fault, and grid-side current fault. If Rd≠0, then it is determined that an internal fault has occurred in the DC bus subsystem. The internal fault of the DC bus subsystem includes at least one of the following faults: DC bus capacitor damage fault and PWM modulation error fault.

2. A computer device having a program stored thereon, wherein, When the program is executed by the processor, it implements the converter fault detection method as described in claim 1.

Images