Search

CN-122017660-A - Open-circuit fault diagnosis method for voltage source type inverter

CN122017660ACN 122017660 ACN122017660 ACN 122017660ACN-122017660-A

Abstract

The invention provides a voltage source type inverter open-circuit fault diagnosis method. The method comprises the steps of obtaining three-phase stator current and PWM modulated reference voltage signals in each PWM interruption period, determining stator voltage vectors required by a voltage model, obtaining compensation voltage vectors by utilizing deviations of current model flux linkage and voltage model flux linkage, converting the compensation voltage vectors into three-phase compensation voltages through inverse Clark conversion, converting cycle integral average characteristics of a certain phase into phase angle field integral from time domain integral, extracting cycle characteristics of the three-phase compensation voltages in the phase angle field, constructing three-phase absolute cycle characteristics according to the cycle characteristics, comparing the three-phase absolute cycle characteristics with a first fault detection threshold to obtain fault marks, constructing fault variable vectors, mapping the fault variable vectors into 1, 0 or-1 according to a second fault detection threshold, and positioning a fault switching tube according to a table lookup rule. The invention can realize quick and low-calculation-amount on-line diagnosis and positioning under the conditions of speed change, load change and parameter mismatch.

Inventors

  • SUN XIANRUI
  • Du Diexuan
  • DIAO NAIZHE
  • TONG HAO
  • WANG XINYI

Assignees

  • 沈阳化工大学

Dates

Publication Date
20260512
Application Date
20260319

Claims (10)

  1. 1. The open-circuit fault diagnosis method for the voltage source type inverter is used in a speed-sensor-free vector control system, the speed-sensor-free vector control system comprises a current model and a voltage model, the voltage source type inverter is a two-level voltage source type inverter, the two-level voltage source type inverter is a three-phase bridge type inverter circuit, a power supply of the three-phase bridge type inverter circuit is a direct-current power supply V d , an output end of the three-phase bridge type inverter circuit is connected to a three-phase asynchronous motor, the three-phase bridge type inverter circuit is composed of three independent A-phase bridge arms, B-phase bridge arm and C-phase bridge arms, the A-phase bridge arm comprises a No. 1 switching tube and an anti-parallel diode D 1 , a No. 2 switching tube and an anti-parallel diode D 2 , the B-phase bridge arm comprises a No. 3 switching tube and an anti-parallel diode D 3 , a No. 4 switching tube and an anti-parallel diode D 4 , the C-phase bridge arm comprises a No. 5 switching tube and an anti-parallel diode D 5 , the midpoints a, C and C are respectively connected to the three-phase asynchronous motor as alternating-phase output ends, and the three-phase bridge type inverter circuit is characterized in that the three-phase fault diagnosis method comprises that currents of 35 i and 35 i.35 i are respectively: Step S1, in each PWM interruption period, three-phase stator current i a ,i b ,i c and PWM modulated reference voltage signals are obtained, and stator voltage vectors required by a voltage model are determined ; Step S2, establishing a mixed model flux observer to respectively calculate a current model flux and a voltage model flux, and obtaining a compensation voltage vector by utilizing the deviation of the current model flux and the voltage model flux through a proportional-integral link The method comprises the steps of obtaining a voltage model flux linkage, namely a voltage model flux linkage, and further obtaining a high-pass and low-pass weighted relation of a voltage model stator flux linkage compensated by the mixed model flux linkage observer, wherein the mixed model flux linkage observer is a combination of a current model flux linkage observer and the voltage model flux linkage observer, and the current model flux linkage is a current model stator flux linkage and the voltage model flux linkage comprises a voltage model stator flux linkage and a voltage model rotor flux linkage; Step S3, converting the compensation voltage vector by the reverse Clark conversion Conversion to three-phase compensation voltage ; S4, calculating a flux linkage phase angle from the flux linkage component of the rotor of the voltage model And define the flux linkage phase angle Output range is 0 to ; Step S5, utilizing the flux linkage phase angle Sweep from a specified start angle The period integral average characteristic of a certain phase corresponds to the property of a complete fundamental wave period T Conversion from time domain integration to phase angle domain integration and extraction of periodic characteristics of three-phase compensation voltage in phase angle domain ; Step S6, by periodic characteristics Constructing three-phase absolute period features And is in contact with a first failure detection threshold Comparing to obtain fault sign ; S7, constructing fault variable vectors And according to the second fault detection threshold Vector of fault variables Mapping to 1, 0 or-1, and positioning the fault switch tube according to the table look-up rule, wherein the fault variable vector From three-phase fault variables Composition is prepared.
  2. 2. The method according to claim 1, wherein in the step S2, the current model flux is a current model stator flux, and the voltage model flux includes a voltage model stator flux and a voltage model rotor flux; the corresponding expression of the current model stator flux linkage is as follows: (1) In the formula (1), the components are as follows, Estimating data for the current model stator flux linkage; The method comprises the steps of estimating data for a current model rotor flux linkage, wherein L m is excitation inductance parameter data, L r is rotor inductance parameter data, and L s is stator inductance parameter data; Stator current vector data; the expression corresponding to the voltage model stator flux linkage is as follows: (2) In the formula (2), the amino acid sequence of the compound, R s is stator resistance parameter data; Estimating data for a voltage model stator flux linkage; The expression corresponding to the voltage model rotor flux linkage is as follows: (3) in the formula (3), the amino acid sequence of the compound, Estimating data for a voltage model rotor flux linkage; the expression of the compensated voltage model stator flux linkage is: (4) In the formula (4), the amino acid sequence of the compound, Stator flux linkage data for the compensated voltage model; compensating the voltage vector data; The compensation voltage vector The corresponding expression is: (5) in the formula (5), the amino acid sequence of the compound, Is proportionality coefficient data; Is integral coefficient data; And (3) with The difference value of (2) is flux linkage error data; And the expression of the high-pass and low-pass weighted relation of the voltage model stator flux linkage compensated by the mixed model flux linkage observer is as follows: (6) in the formula (6), the amino acid sequence of the compound, Transfer function data for a second-order high-pass filter; is the transfer function data of the second-order low-pass filter.
  3. 3. The method according to claim 1, wherein in step S3, the three-phase compensation voltage The corresponding expression is: (7) in the formula (7), the amino acid sequence of the compound, Respectively are compensation voltages Component data; And respectively three-phase compensation voltage data.
  4. 4. The method according to claim 1, wherein in step S4, the flux phase angle is The corresponding expression is: (8) In the formula (8), the amino acid sequence of the compound, Is flux linkage phase angle data; is the data of the flux linkage component of the rotor of the voltage model, when two input components of atan2 At the same time, 0 will Set to 0.
  5. 5. The method according to claim 1, wherein in step S5, the period integral average characteristic of a certain phase is calculated The corresponding expression is: (9) In the formula (9), the amino acid sequence of the compound, The method comprises the steps of integrating average characteristic data for a period of a certain phase, wherein T is complete fundamental wave period data, T 1 is integration starting time data, p is a phase sequence, and a, B and C respectively correspond to A phase, B phase and C phase; Averaging the periodic integral of a phase The corresponding expression for converting from the time domain integral to the phase angle domain integral is: (10) in the formula (10), the amino acid sequence of the compound, Is flux linkage phase angle data; Angle span data for one revolution of the flux linkage phase angle sweep.
  6. 6. The method according to claim 5, wherein in step S5, the period characteristic is extracted for the three-phase compensation voltage in the phase angle domain The method specifically comprises the following steps: Calculating single-period sampling point number according to the sampling period T s and the complete fundamental wave period T ; Constructing three-phase compensation voltage and flux linkage phase angle Is a discrete sample sequence of (a); In each PWM interruption period, three-phase compensation voltage and flux linkage phase angle Respectively obtaining a sampling value and respectively entering a compensation voltage sampling set and a phase angle sampling set; dividing 0-2 pi equally into n parts to obtain equal flux linkage phase angle step length And by Constructing a constant magnetic chain phase angle discrete phase angle sequence for the initial phase angle; When the phase angle of flux linkage When 2 pi is swept, calculating the equal phase angle sliding average value in each flux linkage phase angle interval; Carrying out second-order phase angle movement summation on the mean value characteristics of the n phase angle intervals to obtain periodic characteristics ; Wherein, the number of single-period sampling points The corresponding expression is: (11) In the formula (11), T s is sampling period data; Sampling point data for a single period; Three-phase compensation voltage and flux linkage phase angle The expression corresponding to the discrete sample sequence is: (12) (13) In the formulas (12) and (13), k is the sampling time; index for sampling points and Taking out (0 of the total weight of the product, ); Compensating the voltage for phase a; compensating the voltage for phase B; Compensating the voltage for phase C; Is the phase angle of magnetic linkage At the moment of% ) Is a sampled value of (1); The expressions of the compensation voltage sampling set and the phase angle sampling set are respectively as follows: (14) (15) In the formulas (14) and (15), Collecting data for compensating voltage samples; collecting data for phase angle sampling; Equal flux linkage phase angle step The calculated expression of (2) is: (16) In the formula (16), n is the phase angle domain equal part; For a fixed increment of the phase angle of each segment of flux linkage, i.e. 0 to The phase angle of each part is equal to the phase angle after the n parts are divided equally; the expression of the equal magnetic chain phase angle discrete phase angle sequence is as follows: (17) in the formula (17), the amino acid sequence of the compound, Phase angle data for the initial flux linkage; Sequence number and for step segment of flux linkage phase angle Taking (0, n-1); Representing slave Start to press step Phase angle points obtained by discrete section by section; the expression of the average value of the phase angle sliding of each flux linkage phase angle interval is as follows: (18) In the formula (18), the amino acid sequence of the compound, Sampling point data falling into the phase angle interval; the mean characteristic data of the phase angle interval is obtained; sampling data for the compensation voltage; Periodic characteristics The calculated expression of (2) is: (19) in the formula (19), the amino acid sequence of the compound, And accumulating the obtained periodic characteristics for the mean value characteristics of the small sections in the phase angle domain.
  7. 7. The method for diagnosing an open circuit fault in a voltage source inverter as recited in claim 6, wherein n is less than To ensure that within each phase angle interval Is not 0 and n is Is in the range of 10 to 15 percent.
  8. 8. The method according to claim 1, characterized in that in step S6, three-phase absolute period characteristic is obtained The expression of (2) is: (20) In the formula (20), the amino acid sequence of the compound, Absolute value data of three-phase periodic characteristics; Fault sign The expression of (2) is: (21) in the formula (21), the amino acid sequence of the amino acid, Is fault sign data; Is the first fault detection threshold data and has a G value in any phase exceeding Time determination Indicating that a switching tube open-circuit fault exists; the absolute periodic characteristic of the A phase compensation voltage is provided; the absolute periodic characteristic of the B phase compensation voltage; Compensating the absolute periodic characteristic of the voltage for the C phase; first failure detection threshold By loading the maximum value of the mean feature under transient conditions The calibration is carried out so that the calibration is carried out, The value range of (2) is larger than And is not smaller than To reduce the risk of false alarms.
  9. 9. The method according to claim 1, wherein in step S7, the fault variable vector is The expression of (2) is: (22) in the formula (22), the amino acid sequence of the compound, Three-phase fault variable data; according to the second fault detection threshold Vector of fault variables The corresponding expression mapped to 1, 0 or-1 is: (23) in the formula (23), the amino acid sequence of the compound, Detecting threshold data for a second fault; Is periodic characteristic data, a second fault detection threshold Is determined by experimental setting and is based on fault variable vectors The value rule of (2) enables the fault variable vector to be 1, 0 or-1.
  10. 10. The method of claim 1, wherein in step S7, the mapping relationship corresponding to the table lookup rule is shown in Table 1, wherein the mapping relationship is shown in Table 1 The system is in one-to-one correspondence with the numbers of the fault switching tubes, and comprises a normal state and switching tube 1 to switching tube 6 fault states, wherein 0 represents no open circuit fault, and 1 or-1 represents open circuit fault; Table 1 mapping relation corresponding to table look-up rule 。

Description

Open-circuit fault diagnosis method for voltage source type inverter Technical Field The invention relates to the technical field of open-circuit fault diagnosis of inverters, in particular to a method for diagnosing open-circuit faults of a voltage source type inverter. Background The voltage source type inverter is widely applied to the fields of motor driving, new energy power generation and the like, and the reliability is crucial. The power switching tube is one of the components which are most prone to faults in the inverter, wherein the open circuit faults do not immediately break down the system, but current distortion and torque fluctuation are caused, and secondary faults of the system can be caused by long-term operation. Existing inverter open circuit fault diagnosis methods can be largely classified into signal characteristic-based methods, model-based methods and data driving-based methods. The method based on the signal characteristics generally extracts the characteristics from waveform distortion direct-current side ripples of three-phase current or voltage-current imbalance, has the advantages of simple and convenient implementation, but the characteristics are often strongly related to running frequency and load change, the threshold value is difficult to set uniformly, and false alarm or missing alarm is easy to occur. The model-based method has a certain interpretability by establishing a motor inverter system model or an observer and extracting features by utilizing residual errors or estimation errors, but has strong dependence on working condition changes before and after model parameter accuracy faults and calculation links in a controller. The data driving method needs a large amount of training data covering all working conditions, and has generalization and interpretation problems in engineering. In the comprehensive view, in the control scene of the sensor without a speed vector of variable frequency operation, how to obtain fault characteristics which are insensitive to frequency change and are convenient for embedded in-loop implementation without adding additional hardware, and further realize quick positioning of single open circuit faults is still a key problem which needs to be solved. In addition, many feature extraction methods based on period integration or moving average generally use a fixed period in a time domain as a window, and in a variable frequency speed regulation scene, the period of a fundamental wave changes with frequency, so that the window length is not fixed, and feature quantities are not comparable or stable under different frequencies. In order to solve the problem, we point out that although the period will vary with the frequency, the sweep of the phase angle from the designated start angle can correspond to a complete period, so that the time domain integral characteristic can be converted into the phase angle domain integral characteristic to adapt to the frequency conversion working condition, and further discrete implementation is proposed to reduce the storage and calculation burden, and is more suitable for online execution in the PWM interruption period of the controller. For the above reasons, we need to propose an on-line diagnosis and positioning method for open-circuit faults of single power switches of voltage source type inverter. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art or related art. The single power switch may be open-circuited when the voltage source inverter is operated for a long period in the motor drive system. The fault can cause phase current distortion and torque ripple, affecting system stability. When the speed-sensor-free vector control system is in variable frequency operation, the fundamental wave period changes along with the frequency change, and the characteristic extraction method based on a fixed time period window is easy to be unstable or incomparable under different frequencies. Therefore, the invention aims to provide a method for diagnosing open-circuit faults of a voltage source inverter, which utilizes compensation voltage obtained in a controller to construct fault characteristics under the condition of not adding external voltage measurement hardware, and converts period integration from a time domain to a phase angle domain so as to adapt to frequency change, thereby realizing quick diagnosis and positioning of open-circuit faults. In order to achieve the above object, the present invention provides a method for diagnosing an open circuit fault of a voltage source inverter. The fault diagnosis method is used in a speed-sensor-free vector control system, wherein the speed-sensor-free vector control system comprises a current model and a voltage model; the voltage source type inverter is a two-level voltage source type inverter; the two-level voltage source type inverter is a three-phase bridge inverter circuit, a power sup