CN111398860A - Hybrid PMSM drive system inverter open-circuit fault online diagnosis method - Google Patents

Hybrid PMSM drive system inverter open-circuit fault online diagnosis method Download PDF

Info

Publication number
CN111398860A
CN111398860A CN202010231484.6A CN202010231484A CN111398860A CN 111398860 A CN111398860 A CN 111398860A CN 202010231484 A CN202010231484 A CN 202010231484A CN 111398860 A CN111398860 A CN 111398860A
Authority
CN
China
Prior art keywords
sampling
phase current
model
open
training
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010231484.6A
Other languages
Chinese (zh)
Inventor
骆光照
张泽良
陶雪成
陈寿洛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN202010231484.6A priority Critical patent/CN111398860A/en
Publication of CN111398860A publication Critical patent/CN111398860A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/34—Testing dynamo-electric machines
    • G01R31/343—Testing dynamo-electric machines in operation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Medical Informatics (AREA)
  • Evolutionary Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Inverter Devices (AREA)
  • Control Of Ac Motors In General (AREA)

Abstract

The invention relates to an on-line diagnosis method for an open-circuit fault of a hybrid PMSM (permanent magnet synchronous motor) driving system inverter, which can be used for carrying out on-line fault diagnosis and positioning on a plurality of inverter open-circuit faults. Introducing a Luenberger observer to track the three-phase current of the PMSM, so as to obtain the observation residual error of the three-phase current as a fault diagnosis basis; designing a sampling strategy, and acquiring an original sample set for training; in off-line training, firstly introducing a principal component analysis method to preprocess an original sample set to obtain a data dimension reduction model, and then obtaining a classification model through a training support vector machine; and finally, operating the dimension reduction model and the classification model in a driving system to realize online diagnosis and positioning of the fault. The invention does not increase hardware equipment, does not influence drive control, is not influenced by load and model parameter change, does not depend on manually set diagnosis basis through machine learning on actual working conditions, and reduces the possibility of misdiagnosis.

Description

Hybrid PMSM drive system inverter open-circuit fault online diagnosis method
Technical Field
The invention belongs to the field of fault diagnosis of a PMSM (permanent magnet synchronous motor) drive system, and relates to a hybrid online diagnosis method for open-circuit faults of an inverter of the PMSM drive system.
Background
In the fields of modern industrial production and national defense industry, the permanent magnet synchronous motor is widely applied due to the advantages of fast response, high power density, high efficiency and the like. Along with the wider application, the reliability of the permanent magnet synchronous motor driving system in practical application also puts forward higher requirements. In a permanent magnet synchronous motor driving system, an inverter comprises a power electronic device and a driving circuit thereof, is the link with the closest combination of digital control and power output, and is also the weak link with various fault types and frequent faults.
The failure modes of the power devices in the inverter can be divided into an open-circuit failure and a short-circuit failure. The direct short-circuit fault of the bridge arm caused by the wrong driving signal, overvoltage breakdown, thermal breakdown and the like causes great damage to a system and even directly damages an inverter, so that a protection circuit with quick response needs to be designed by hardware aiming at the short-circuit fault, or the short-circuit fault is converted into an open-circuit fault by adopting a mode of additionally installing a thermal fuse. For open-circuit faults caused by conditions of over-high junction temperature device damage, loss of driving signals, poor connection and the like, serious damage to a system cannot be caused immediately, a common system protection mechanism cannot be triggered, and secondary damage to an actuating system can be caused by current imbalance and torque pulsation caused by opening of a certain power tube or certain power tubes. Therefore, research on open-circuit faults of the inverters in various forms becomes a key point of research on the faults of the inverters.
At present, the main diagnostic methods for the open-circuit fault of the inverter include a voltage-current signal-based method, a model-based method, and a data-driven method.
Patent application No. CN201910281011.4 discloses a method for diagnosing open-circuit fault of inverter based on current signal, which includes calculating a per unit value of the sum of absolute values of three-phase currents at a sampling time, comparing the per unit value with a preset threshold value, determining whether open-circuit fault occurs, and locating fault according to the polarity of phase current. The patent application number CN201711287150.5 diagnoses the open-circuit fault through the value range of the current vector under the two-phase rotating coordinate system. The method based on current signal analysis is limited in application in the application occasions where the current amplitude is small or the sampling precision is low, and the preset threshold value serving as a diagnostic standard is greatly influenced by load change.
Patent application nos. CN201811102120.7 and CN201910725088.6 disclose two model-based open-circuit fault diagnosis methods based on a sliding mode observer and a differential current observer, respectively. And obtaining a current observation residual error by constructing a state observer, and then performing evaluation of a residual error signal and decision of fault diagnosis by comparing with a threshold value. Although misjudgment caused by poor signal quality can be better avoided through the judgment of the residual error and the threshold value, the threshold value is improperly set in application depending on a system mathematical model, the actual working condition of an electric drive system cannot be well adapted, and the false alarm rate is higher. In patent application No. CN201310743597.4, the acquisition of the current residual is performed by a hybrid system current observer, but the same problems as described above exist.
Patent application number CN201910404744.2 discloses a three-level inverter open-circuit fault diagnosis method based on an optimized support vector machine, which uses voltage signals at two ends of a clamping diode as the basis of fault diagnosis, and uses an optimized support vector machine to construct a classifier, so as to diagnose and classify open-circuit faults. Patent application No. CN201910668544.8 discloses a method for diagnosing and classifying faults through a BP neural network by using three-phase current signals of an inverter as a basis for fault diagnosis. The data-driven method gets rid of the limitation of a system mathematical model, and diagnosis and positioning can be synchronously carried out by adopting a classifier. However, the data-driven method needs to set the data type of the training input set and the input set size reasonably to take the training cost and the final training result into consideration. On the other hand, data used for learning at present are generally directly obtained by signal sampling, and similar problems still exist in the voltage-current signal-based method.
Disclosure of Invention
Technical problem to be solved
In order to avoid the defects of the prior art, the invention provides the hybrid PMSM drive system inverter open-circuit fault online diagnosis method which can effectively avoid the influence of load and model parameter change, and reduces the possibility of misdiagnosis by machine learning of actual working conditions and without depending on the diagnosis basis set manually.
Technical scheme
An open-circuit fault online diagnosis method for a hybrid PMSM drive system inverter is characterized by comprising the following steps:
step 1: estimating three-phase current of the PMSM by using a state observer to obtain a three-phase current observed value, and subtracting the current observed value from an actual three-phase current value fed back by the PMSM driving system to obtain an observed residual error of the three-phase current;
step 2: after the current observation residual error is obtained, sampling three-phase current observation residual error data to obtain an original sample set for off-line machine learning model training; the sampling method comprises the following steps:
step 2.1: taking the position of a motor rotor when the phase current of the motor A changes from negative half circumference to positive half circumference to zero crossing as a mark, and starting a sampling process of three-phase current observation residual errors when the rotor moves to the position;
step 2.2: calculating electrical cycles according to the motor rotating speed instruction, wherein the time of each electrical cycle is teSetting the time of current sampling period of PMSM drive system as tsAnd then the current observation residual error sampling frequency is Ns＝(te/k)/tsWherein k is a positive integer;
step 2.3: sampling the current observation residual NsSecondly, finishing the sampling process to obtain three-phase current residual data of a 1/k electrical period; after the sampling process is finished, the obtained three-phase current observation residual error samples are respectively set as row vectors
Figure BDA0002429409330000031
Each sample in the original sample set for off-line training is 3 × N in one dimensionsColumn vector of
Figure BDA0002429409330000032
And step 3: after obtaining an original sample set, preprocessing the original sample set in an off-line training to obtain a preprocessed training input set and a model capable of preprocessing a sampling sample, wherein the preprocessing steps are as follows:
step 3.1: let f1,f2,......,fnRespectively n signs of open-circuit failure modes of the inverter, f0For the indication of the health of the inverter, m current residual samples are sampled in each fault mode and in each health state, and the original sample set obtained for offline training is of a size of (3 × N)s) × (n × m)
Figure BDA0002429409330000033
Step 3.2, preprocessing the original sample set by using a principal component analysis method, reserving two groups of eigenvectors with the largest corresponding eigenvalue, and obtaining a vector with the size of (3 × N)s) × 2 matrix
Figure BDA0002429409330000041
Namely, the model is the dimension reduction model;
3.3, using a dimension reduction model to reduce the dimension of each sample in the original sample set to 2 dimensions, wherein the obtained preprocessed training input set is a matrix with the size of 2 × (n × m)
Figure BDA0002429409330000042
And 4, step 4: after a training input set is obtained, a classification model is trained offline by using the training input set to obtain a model capable of classifying sampling samples;
and 5: and after obtaining the dimension reduction model and the classification model, operating the dimension reduction model and the classification model in a control chip of the PMSM drive system, and performing sample sampling, dimension reduction and classification calculation on line to realize diagnosis and positioning of the open-circuit fault of the inverter.
Step 1 the state observer is a lunberg observer.
In the step 1, after the current observation residual is obtained by the state observer, the residual signal is subjected to sliding window average filtering to be used as a basis for fault diagnosis.
In step 4, a directed acyclic graph type support vector machine is used for learning a training input set, the training result is n (n +1)/2 binary classifiers, and the expression form is a × x1+b×x2+ c ═ 0, where x1And x2Two elements of the reduced-dimension sample are obtained, a and b are constants obtained by off-line training, and n classifiers are called each time on-line fault diagnosis and positioning are carried out.
Advantageous effects
The invention provides a hybrid PMSM drive system inverter open-circuit fault online diagnosis method, which has the following beneficial technical effects:
(1) according to the fault diagnosis method, the state observer based on the model is combined with the machine learning model driven by data, the state observer enables the diagnosis process not to be influenced by changes of load and rotating speed, the machine learning model trained by system operation data enables the diagnosis process to eliminate interference of changes of model parameters, meanwhile, the diagnosis basis does not depend on a threshold set manually, and the possibility of misdiagnosis is reduced;
(2) the invention introduces a plurality of classifiers to diagnose and position a plurality of open-circuit faults, so that the fault diagnosis and the positioning are carried out simultaneously, the efficiency of the fault diagnosis and the positioning is improved, and the diagnosis of the open-circuit fault of the inverter can be completed within less than half of the electric cycle;
(3) the sampling strategy provided by the invention judges the time for starting the sampling process according to the corresponding relation between the position of the motor rotor and the phase current phase, and determines the sampling times by combining the running speed of the motor, so that the fault diagnosis method can be applied to a full-speed section, and simultaneously, the time length and the sample number of single sampling can be flexibly adjusted, thereby providing a reliable and efficient data acquisition strategy for online running a machine learning model in a dynamic PMSM driving system.
(4) The fault diagnosis method provided by the invention diagnoses and positions the fault according to the system feedback and the state variable, does not increase hardware equipment and does not influence the design of a driver;
drawings
FIG. 1 is a schematic diagram of an open-circuit fault diagnosis method for an inverter of a PMSM drive system according to the present invention
FIG. 2 is a waveform diagram of the experimental results of the present invention
FIG. 3 is a waveform diagram of the experimental results of the present invention
FIG. 4 is a waveform diagram of the experimental results of the present invention
FIG. 5 is a waveform diagram of the experimental results of the present invention
FIG. 6 is a waveform diagram of the experimental results of the present invention
Detailed Description
The invention will now be further described with reference to the following examples and drawings:
as shown in fig. 1, the method for diagnosing an open-circuit fault of an inverter of a PMSM drive system according to the present invention. The specific implementation is as follows:
estimating three-phase current of the PMSM by using a state observer to obtain a three-phase current observed value, and subtracting the current observed value from an actual three-phase current value in system feedback to obtain an observed residual error of the three-phase current.
The observer for tracking PMSM three-phase current is a Luenberger observer, and the expression is as follows:
Figure BDA0002429409330000061
wherein,
Figure BDA0002429409330000062
，u(t)＝(ud,uq)T，y(t)＝(id,iq)Tk is the feedback coefficient matrix and A, B, C is the coefficient matrix in the PMSM state equation.
As shown in fig. 2, when the health state of the inverter and the open-circuit fault of the upper bridge arm power tube of the phase a occur, the designed luneberg observer is used for changing the three-phase current residual error of the motor. It can be seen that when the current sudden change caused by the open-circuit fault occurs, the three-phase current residual errors rapidly deviate from the vicinity of the zero point during normal operation, and the fault characteristics are shown, so that the requirement on rapidity of fault diagnosis can be met by using the current residual errors as the basis of fault diagnosis.
As shown in fig. 3, the designed luneberg observer is used for changing the residual error of the three-phase current of the motor when the load of the motor changes. Load current IqShowing the variation of the load, the load of the motor is increased from 3n.m to 6n.m in a time of about 0.5 seconds, the load is removed after 6 seconds of operation, and the load is reduced to the initial 3n.m in a very short time. It can be seen that, no matter in the stage of gradually increasing load, the stage of stabilizing load, or the stage of sudden load change, the observed residual of the a-phase current periodically changes in a constant range all the time, and is not affected by the load change. Therefore, the observation residual error of the phase current is used as the basis of fault diagnosis, and the interference of the motor load change on the diagnosis result can be effectively avoided.
And (2) performing sliding window average filtering on the three-phase current observation residual error obtained in the step (1), wherein the filtering process is applied to a sampling process and an online fault diagnosis process so as to reduce the distribution variance among samples in the same type of working state and be beneficial to improving the diagnosis precision of the classifier.
Then, designing a sampling strategy to sample the three-phase current observation residual error, wherein the designed sampling strategy is as follows:
step (2.1) taking the position of a motor rotor when the phase current of the motor A changes from negative half circumference to positive half circumference to zero crossing as a mark, and starting a sampling process of three-phase current observation residual errors when the rotor moves to the position;
step (2.2) calculating electrical cycles according to the motor rotating speed instruction, wherein the time of each electrical cycle is teSetting the time of current sampling period of PMSM drive system as tsAnd then the current observation residual error sampling frequency is Ns＝(te/k)/tsWherein k is a positive integer;
step (2.3) sampling the current observation residual error by NsThen, finishing the sampling process to obtain three-phase current residual error data of 1/k electrical period, and setting the obtained three-phase current observation residual error samples after finishing the sampling processAs a row vector
Figure BDA0002429409330000071
Each sample in the original sample set for off-line training is 3 × N in one dimensionsColumn vector of
Figure BDA0002429409330000072
As shown in fig. 4, in order to sample the three-phase current observation residual data of half an electrical cycle, in fig. 4, from top to bottom, there are acquisition channels one to four of the oscilloscope, channel one is the motor a-phase current after the fault occurs, and channels two to four are A, B, C three-phase current residual samples respectively.
And (3) preprocessing an observation residual original sample set obtained by sampling, wherein the steps are as follows:
(3.1) setting f1,f2,......,fnRespectively n signs of open-circuit failure modes of the inverter, f0The inverter health sign is obtained by sampling m current residual error samples under each fault mode and health state, and the original sample set for offline training is obtained by the method that the size is (3 × N)s) × (n × m)
Figure BDA0002429409330000073
(3.2) introducing a principal component analysis method to preprocess the original sample set, reserving two groups of eigenvectors with the largest corresponding eigenvalue, and obtaining a vector with the size of (3 × N)s) × 2 matrix
Figure BDA0002429409330000075
Namely, the model is the dimension reduction model;
(3.3) reducing the dimension of each sample in the original sample set to 2 dimensions by using a dimension reduction model to obtain a training input set, wherein the training input set is a matrix with the size of 2 × (n × m)
Figure BDA0002429409330000074
Step (4), a directed acyclic graph type support vector machine is introduced to learn a training input set, the training result is n (n +1)/2 binary classifiers, and the expression form is a × x1+b×x2+ c ═ 0, where x1And x2Two elements of the reduced-dimension sample are obtained, a and b are constants obtained by off-line training, and n classifiers are called each time on-line fault diagnosis and positioning are carried out.
And (5) operating the dimension reduction model and the classification model in a control chip of the PMSM drive system to realize diagnosis and positioning of the open-circuit fault of the inverter.
As shown in fig. 5, the calculation results are online calculation results of the classifier at the last node in the directed acyclic support vector machine, and the corresponding relationship between the calculation results and the a-phase current before and after the fault. In fig. 5, the operation result is greater than 1, which indicates that the sampling sample belongs to the healthy state of the inverter, and the operation result is less than-1, which indicates that the sampling sample belongs to the inverter and an open-circuit fault occurs in the upper bridge arm power tube of the phase a.
As can be seen from fig. 5, before the open-circuit fault at T1 occurs, the operation results of the samples on the classifier are all greater than 1, i.e., the diagnosis result is that the inverter is in a healthy state. And after the T1 open circuit fault occurs, the operation results are all smaller than-1, namely the diagnosis result is that the inverter T1 is open.
As shown in fig. 6, in order to test the shortest diagnosis time of the method of the present invention after a plurality of tests, it can be seen that the controller completes the diagnosis of the fault within a time less than a half electrical cycle after the open-circuit fault occurs in the test.
The experimental conditions are that the DC bus voltage is 60V, the pole pair number of the motor is 4 pairs of poles, Ld＝Lq2.4mH, a winding electric group is 0.306 omega, a permanent magnet flux linkage is 0.281Wb, the working rotation speed is 150rpm, the switching frequency of the inverter is 10kHz, and a control chip is TMS320F 28335.

Claims (4)

1. An open-circuit fault online diagnosis method for a hybrid PMSM drive system inverter is characterized by comprising the following steps:
step 1: estimating three-phase current of the PMSM by using a state observer to obtain a three-phase current observed value, and subtracting the current observed value from an actual three-phase current value fed back by the PMSM driving system to obtain an observed residual error of the three-phase current;
step 2: after the current observation residual error is obtained, sampling three-phase current observation residual error data to obtain an original sample set for off-line machine learning model training; the sampling method comprises the following steps:
step 2.1: taking the position of a motor rotor when the phase current of the motor A changes from negative half circumference to positive half circumference to zero crossing as a mark, and starting a sampling process of three-phase current observation residual errors when the rotor moves to the position;
step 2.2: calculating electrical cycles according to the motor rotating speed instruction, wherein the time of each electrical cycle is teSetting the time of current sampling period of PMSM drive system as tsAnd then the current observation residual error sampling frequency is Ns＝(te/k)/tsWherein k is a positive integer;
step 2.3: sampling the current observation residual NsSecondly, finishing the sampling process to obtain three-phase current residual data of a 1/k electrical period; after the sampling process is finished, the obtained three-phase current observation residual error samples are respectively set as row vectors
Figure FDA0002429409320000011
Each sample in the original sample set for off-line training is 3 × N in one dimensionsColumn vector of
Figure FDA0002429409320000012
And step 3: after obtaining an original sample set, preprocessing the original sample set in an off-line training to obtain a preprocessed training input set and a model capable of preprocessing a sampling sample, wherein the preprocessing steps are as follows:
step 3.1: let f1,f2,......,fnRespectively n signs of open-circuit failure modes of the inverter, f0For the inverter health mark, m current residual error samples are respectively sampled under each fault mode and health state to obtain an original sample for off-line trainingSet is one size of (3 × N)s) × (n × m)
Figure FDA0002429409320000013
Step 3.2, preprocessing the original sample set by using a principal component analysis method, reserving two groups of eigenvectors with the largest corresponding eigenvalue, and obtaining a vector with the size of (3 × N)s) × 2 matrix
Figure FDA0002429409320000014
Namely, the model is the dimension reduction model;
3.3, using a dimension reduction model to reduce the dimension of each sample in the original sample set to 2 dimensions, wherein the obtained preprocessed training input set is a matrix with the size of 2 × (n × m)
Figure FDA0002429409320000021
And 4, step 4: after a training input set is obtained, a classification model is trained offline by using the training input set to obtain a model capable of classifying sampling samples;
and 5: and after obtaining the dimension reduction model and the classification model, operating the dimension reduction model and the classification model in a control chip of the PMSM drive system, and performing sample sampling, dimension reduction and classification calculation on line to realize diagnosis and positioning of the open-circuit fault of the inverter.
2. The hybrid PMSM drive system inverter open-circuit fault on-line diagnostic method as recited in claim 1, wherein the step 1 state observer is a Luenberger observer.
3. The hybrid PMSM drive system inverter open-circuit fault on-line diagnosis method according to claim 1, characterized in that after obtaining current observation residual through a state observer in step 1, the residual signal is subjected to sliding window average filtering as a basis for fault diagnosis.
4. A hybrid according to claim 1The method for online diagnosing the open-circuit fault of the PMSM driving system inverter is characterized in that a directed acyclic graph type support vector machine is used for learning a training input set in step 4, the training result is n (n +1)/2 binary classifiers, and the expression form is a × x1+b×x2+ c ═ 0, where x1And x2Two elements of the reduced-dimension sample are obtained, a and b are constants obtained by off-line training, and n classifiers are called each time on-line fault diagnosis and positioning are carried out.
CN202010231484.6A 2020-03-27 2020-03-27 Hybrid PMSM drive system inverter open-circuit fault online diagnosis method Pending CN111398860A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010231484.6A CN111398860A (en) 2020-03-27 2020-03-27 Hybrid PMSM drive system inverter open-circuit fault online diagnosis method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010231484.6A CN111398860A (en) 2020-03-27 2020-03-27 Hybrid PMSM drive system inverter open-circuit fault online diagnosis method

Publications (1)

Publication Number Publication Date
CN111398860A true CN111398860A (en) 2020-07-10

Family

ID=71434653

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010231484.6A Pending CN111398860A (en) 2020-03-27 2020-03-27 Hybrid PMSM drive system inverter open-circuit fault online diagnosis method

Country Status (1)

Country Link
CN (1) CN111398860A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100946A (en) * 2020-08-21 2020-12-18 北京科技大学 Inverter open-circuit fault diagnosis method and device based on fault online simulation
CN112285489A (en) * 2020-10-26 2021-01-29 青岛鼎信通讯股份有限公司 Fault indicator fault positioning method based on feature fusion and model fusion
CN113064073A (en) * 2021-03-12 2021-07-02 合肥恒大江海泵业股份有限公司 Permanent magnet synchronous motor turn-to-turn short circuit fault diagnosis method based on residual current
CN116541772A (en) * 2023-05-05 2023-08-04 兰州理工大学 Cascade H-bridge inverter fault diagnosis method based on multi-source fusion residual error network

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103701394A (en) * 2013-12-30 2014-04-02 哈尔滨工业大学 Current-magnitude-based open-circuit failure online-diagnosis method for power tube of inverter
CN106019058A (en) * 2016-07-07 2016-10-12 西北工业大学 Phase current analysis based three-phase full-bridge inverter open-circuit fault diagnosis method
CN107505133A (en) * 2017-08-10 2017-12-22 滁州学院 The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM
CN111551872A (en) * 2020-02-27 2020-08-18 西北工业大学 Online diagnosis method for open-circuit fault of PMSM (permanent magnet synchronous motor) driving system inverter

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103701394A (en) * 2013-12-30 2014-04-02 哈尔滨工业大学 Current-magnitude-based open-circuit failure online-diagnosis method for power tube of inverter
CN103701394B (en) * 2013-12-30 2016-06-08 哈尔滨工业大学 A kind of online diagnostic method of the open-circuit fault of power tubes of inverter based on the magnitude of current
CN106019058A (en) * 2016-07-07 2016-10-12 西北工业大学 Phase current analysis based three-phase full-bridge inverter open-circuit fault diagnosis method
CN107505133A (en) * 2017-08-10 2017-12-22 滁州学院 The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM
CN107505133B (en) * 2017-08-10 2019-05-28 滁州学院 The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM
CN111551872A (en) * 2020-02-27 2020-08-18 西北工业大学 Online diagnosis method for open-circuit fault of PMSM (permanent magnet synchronous motor) driving system inverter

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100946A (en) * 2020-08-21 2020-12-18 北京科技大学 Inverter open-circuit fault diagnosis method and device based on fault online simulation
CN112100946B (en) * 2020-08-21 2021-05-25 北京科技大学 Inverter open-circuit fault diagnosis method and device based on fault online simulation
CN112285489A (en) * 2020-10-26 2021-01-29 青岛鼎信通讯股份有限公司 Fault indicator fault positioning method based on feature fusion and model fusion
CN112285489B (en) * 2020-10-26 2022-02-22 青岛鼎信通讯股份有限公司 Fault indicator fault positioning method based on feature fusion and model fusion
CN113064073A (en) * 2021-03-12 2021-07-02 合肥恒大江海泵业股份有限公司 Permanent magnet synchronous motor turn-to-turn short circuit fault diagnosis method based on residual current
CN116541772A (en) * 2023-05-05 2023-08-04 兰州理工大学 Cascade H-bridge inverter fault diagnosis method based on multi-source fusion residual error network
CN116541772B (en) * 2023-05-05 2023-10-10 兰州理工大学 Cascade H-bridge inverter fault diagnosis method based on multi-source fusion residual error network

Similar Documents

Publication Publication Date Title
CN111551872B (en) Online diagnosis method for open-circuit fault of PMSM (permanent magnet synchronous motor) driving system inverter
CN111398860A (en) Hybrid PMSM drive system inverter open-circuit fault online diagnosis method
CN113159077B (en) Three-phase inverter fault identification method based on hybrid convolutional neural network
WO2021184820A1 (en) Two-level three-phase voltage source inverter bridge arm open-circuit fault diagnosis method
CN110441643A (en) Inverter power pipe open circuit fault diagnostic method in control system for permanent-magnet synchronous motor
CN101581750A (en) Short circuit phase identification method
CN111458666B (en) Permanent magnet synchronous motor winding open-circuit fault diagnosis method
CN111624514A (en) Method for diagnosing short-circuit and open-circuit faults of switching tube of three-level rectifier
CN109239599B (en) Demagnetization fault diagnosis method for permanent magnet synchronous motor
CN104698397A (en) Fault diagnosis method of multi-level inverter
CN111537914A (en) Open-circuit fault detection method for power tube of three-phase inverter
Muduli et al. Ann based on-board fault diagnostic for induction motor drive in low-cost electric vehicles
CN113534000B (en) New energy automobile driving system inverter and current sensor fault diagnosis method
CN111413646B (en) Real-time detection method for open-circuit fault of bridge arm of three-phase voltage source inverter
CN112688608B (en) Fault diagnosis method for three-phase permanent magnet synchronous motor control system
CN113534035B (en) Tiny fault diagnosis method for current sensor of new energy electric vehicle driving system
CN111679225B (en) Electric drive system open-circuit fault diagnosis method based on fuzzy logic theory
CN113238170A (en) Inverter open-circuit fault online diagnosis method for motor control
CN114172443A (en) Online fault diagnosis method for current sensor of permanent magnet motor driving system
CN108761351B (en) Three-phase rectifier open-circuit fault diagnosis method based on SVPWM sector number
Sun et al. An open-circuit fault diagnosis algorithm based on signal normalization preprocessing for motor drive inverter
Quabeck et al. Detection of broken rotor bars in induction machines using machine learning methods
CN112731193B (en) NPC inverter multi-class fault diagnosis method based on sliding-mode observer
CN113295999B (en) Demagnetizing, winding breaking and turn-to-turn short circuit fault classification method for permanent magnet synchronous motor
CN111983451B (en) Open-circuit fault diagnosis method for brushless direct current motor based on stirrer

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20200710

WD01 Invention patent application deemed withdrawn after publication