CN108614940A - A kind of permanent magnet synchronous motor performance degradation assessment method and system - Google Patents
A kind of permanent magnet synchronous motor performance degradation assessment method and system Download PDFInfo
- Publication number
- CN108614940A CN108614940A CN201810411798.7A CN201810411798A CN108614940A CN 108614940 A CN108614940 A CN 108614940A CN 201810411798 A CN201810411798 A CN 201810411798A CN 108614940 A CN108614940 A CN 108614940A
- Authority
- CN
- China
- Prior art keywords
- permanent magnet
- magnet synchronous
- motor
- synchronous motor
- assessed
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Control Of Ac Motors In General (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
The invention discloses a kind of permanent magnet synchronous motor performance degradation assessment method and system, disclosed method includes step S100：Acquisition permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are in motor stator current signal and vibration signal under different degenerate states；Step S200：It is in motor performance degenerative character vector under different conditions according to motor stator current signal in step S100 and vibration signal extraction motor；Step S300：Motor performance degenerative character vector, which is generated, with step S200 trains category support vector machines model；Step S400：Acquire the motor stator current signal and vibration signal extraction motor performance degenerative character vector of permanent magnet synchronous motor to be assessed, and judge whether permanent magnet synchronous motor to be assessed is in normal condition using supporting vector machine model under normal condition and predetermined threshold value, calculate permanent magnet synchronous motor performance degradation index to be assessed if in abnormal state.The degree of motor performance degeneration can be detected or measured during motor performance is degenerated.
Description
Technical field
The present invention relates to a kind of permanent magnet synchronous motor technical fields more particularly to a kind of permanent magnet synchronous motor performance degradation to comment
Estimate method and system.
Background technology
Permanent magnet synchronous motor is a kind of electronic product of complexity, in the normal operation of the motor the degeneration of its performance be often because
Caused by its component performance degradation, for example, bearing wear degenerate, permanent magnet demagnetization, insulation performance aging, rotor eccentricity by
It is cumulative big etc., when its performance degradation will break down and fail to a certain extent, so permanent magnet synchronous motor is in longtime running
The Potential feasibility to break down in the process can accordingly increase.Once the critical component of motor breaks down, so that it may can cause whole
The collapse of platform equipment shuts down and influences entire production process, causes huge economic loss.If it is possible to be moved back in motor performance
The degree of motor performance degeneration is detected or measured during change, then production and motor dimension can be organized targetedly
It repaiies, the generation for preventing motor abnormality from failing.
Invention content
The object of the present invention is to provide a kind of permanent magnet synchronous motor performance degradation assessment method and systems, can be in motor
The degree that motor performance degeneration is detected or measured during performance degradation can targetedly organize production and motor dimension
It repaiies, the generation for preventing motor abnormality from failing.
In order to solve the above technical problems, the present invention provides a kind of permanent magnet synchronous motor performance degradation assessment method, the side
Method includes the following steps：
Step S100：Acquisition permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are under different degenerate states
Motor stator current signal and vibration signal；
Step S200：It is under different degenerate states according to permanent magnet synchronous motor normal condition and permanent magnet synchronous motor
Motor stator current signal and vibration signal extraction motor are in motor performance degenerative character vector under different conditions；
Step S300：The motor performance degenerative character vector being under different conditions with motor is respectively trained its corresponding point
Class support vector machines model；
Step S400：Acquire the motor stator current signal and vibration signal extraction motor of permanent magnet synchronous motor to be assessed
Energy degenerative character vector, and judged using permanent magnet synchronous motor supporting vector machine model under normal condition and predetermined threshold value
Whether permanent magnet synchronous motor to be assessed is in normal condition, and permanent magnet synchronous motor to be assessed is calculated if in abnormal state
Can degenerate index.
Preferably, step S200 is specially：
Step S210：Permanent magnet synchronous motor is in normal condition and permanent magnet synchronous motor is under different degenerate states
Motor stator current signal and vibration signal carry out empirical mode decomposition, choose several IMF components of each signal；
Step S220：Calculate the energy of all IMF components；
Step S230：All IMF component energies are normalized respectively；
Step S240：Motor performance under different conditions is built respectively with the normalization IMF component energies under different conditions to move back
Change feature vector.
Preferably, step S300 is specially：It is moved back with motor performance under Lagrange multiplier, gaussian kernel function and different conditions
Change feature vector and category support vector machines model under different conditions is respectively trained, and obtain different conditions drag suprasphere, obtains
To the centre of sphere and radius of the suprasphere of each model.
Preferably, it is specially in the step 400：
Step S410：It calculates under the permanent magnet synchronous motor model suprasphere centre of sphere under different degenerate states and normal condition
Generalized distance between the permanent magnet synchronous motor model suprasphere centre of sphere and relative distance；
Step S420：It acquires the motor stator current signal of permanent magnet synchronous motor to be assessed and vibration signal and extracts motor
Performance degradation feature vector；
Step S430：It calculates and classifies under the motor performance degenerative character vector and different conditions of permanent magnet synchronous motor to be assessed
Generalized distance between the suprasphere centre of sphere of supporting vector machine model and relative distance；
Step S440：Judge permanent magnet synchronous motor to be assessed motor performance degenerative character vector and normal condition under classify
Whether relative distance is less than predetermined threshold value between the suprasphere centre of sphere of supporting vector machine model, if the permanent-magnet synchronous to be assessed less than if
Motor is in normal condition, on the contrary then enter step S450；
Step S450：The motor performance degenerative character vector of permanent magnet synchronous motor more to be assessed and lower point of abnormal condition
Relative distance between the suprasphere centre of sphere of class support vector machines model, and then calculate permanent magnet synchronous motor performance degradation to be assessed and refer to
Mark.
Preferably, permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are in different degenerations in the step S100
State is specially：The permanent magnet synchronous motor of 5 models of the same race is taken, wherein the 1st is completely new normal permanent magnet synchronous motor, the 2nd
Platform is the permanent magnet synchronous motor of bearing performance high degradation, the 3rd be permanent magnet performance high degradation permanent magnet synchronous motor, the
4 permanent magnet synchronous motors degenerated for rotor eccentricity, the 5th permanent magnet synchronous motor for insulation performance high degradation.
Preferably, the step S100 is specially：Normal motor is run 50 minutes, every 1 minute to stator current and shaking
Dynamic signal is sampled, each sample-duration 5 seconds；Remaining 4 run 10 points in different degenerate state permanent magnet synchronous motors
Clock sampled stator current signal and vibration signal every 12 seconds, each sample-duration 5 seconds.
Preferably, the vibration signal refers to the displacement of vibration, speed, acceleration signal.
Preferably, several IMF components that each signal is chosen in the step S210 are specially to choose each signal
Preceding 6 IMF components.
The present invention also provides a kind of permanent magnet synchronous motor performance degradation assessment systems, including signal acquisition module, feature
Extraction module, model building module and performance degradation assessment module, wherein：
Signal acquisition module is in different degenerations for acquiring permanent magnet synchronous motor normal condition and permanent magnet synchronous motor
Motor stator current signal and vibration signal under state are sent to characteristic extracting module；
Characteristic extracting module, motor stator current signal and vibration signal for being sent according to signal acquisition module extract
Motor performance degenerative character vector, is sent to model building module under different conditions；
Model building module, motor performance degenerative character is vectorial under the different conditions for being sent according to characteristic extracting module
Category support vector machines model under different conditions is respectively trained；
Performance degradation assessment module, motor stator current signal and vibration for acquiring permanent magnet synchronous motor to be assessed are believed
Number extraction motor performance degenerative character vector, using the support under normal condition of the permanent magnet synchronous motor of model building module
Vector machine model and predetermined threshold value judge whether permanent magnet synchronous motor to be assessed is in normal condition, if in abnormal state
Calculate permanent magnet synchronous motor performance degradation index to be assessed.
This method and system can detect or measure the degree of motor performance degeneration during motor performance is degenerated,
So as to targetedly organize production and electrical machmary maintenance, the generation for preventing motor abnormality from failing.
Description of the drawings
Fig. 1 is a kind of flow chart for permanent magnet synchronous motor performance degradation assessment method that the first embodiment provides；
Fig. 2 is a kind of flow chart for permanent magnet synchronous motor performance degradation assessment method that second of embodiment provides；
Fig. 3 is a kind of structure diagram of permanent magnet synchronous motor performance degradation assessment system provided by the invention；
Fig. 4 is the signal acquisition module schematic diagram of permanent magnet synchronous motor performance degradation assessment system provided by the invention.
Specific implementation mode
In order that those skilled in the art will better understand the technical solution of the present invention, below in conjunction with the accompanying drawings to the present invention
It is described in further detail.
Referring to Fig. 1, Fig. 1 is a kind of stream for permanent magnet synchronous motor performance degradation assessment method that the first embodiment provides
Cheng Tu.
A kind of permanent magnet synchronous motor performance degradation assessment method, the described method comprises the following steps：
Step S100：Acquisition permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are under different degenerate states
Motor stator current signal and vibration signal；
Step S200：It is under different degenerate states according to permanent magnet synchronous motor normal condition and permanent magnet synchronous motor
Motor stator current signal and vibration signal extraction motor are in motor performance degenerative character vector under different conditions；
Step S300：The motor performance degenerative character vector being under different conditions with motor is respectively trained its corresponding point
Class support vector machines model；
Step S400：Acquire the motor stator current signal and vibration signal extraction motor of permanent magnet synchronous motor to be assessed
Energy degenerative character vector, and judged using permanent magnet synchronous motor supporting vector machine model under normal condition and predetermined threshold value
Whether permanent magnet synchronous motor to be assessed is in normal condition, and permanent magnet synchronous motor to be assessed is calculated if in abnormal state
Can degenerate index.
It acquisition permanent magnet synchronous motor normal condition and motor stator current signal under different degenerate states and shakes
Dynamic signal.Motor stator current signal is acquired by current sensor, and vibration signal is acquired by vibrating sensor.According to acquisition
Motor stator current signal and vibration signal extraction motor be under different conditions motor performance degenerative character vectorial, and electricity consumption
Its corresponding category support vector machines model is respectively trained in the motor performance degenerative character vector that machine is under different conditions.Acquisition
Motor stator current signal and vibration signal the extraction motor performance degenerative character vector of permanent magnet synchronous motor to be assessed, and utilize
Judge permanent magnet synchronous electric to be assessed using permanent magnet synchronous motor supporting vector machine model under normal condition and predetermined threshold value
Whether machine is in normal condition, and permanent magnet synchronous motor performance degradation index to be assessed is calculated if in abnormal state.Pass through
Permanent magnet synchronous motor performance degradation index judges the degree that motor performance is degenerated.
The method achieve the degree that motor performance degeneration is detected or measured during motor performance is degenerated, can be with
Targetedly tissue production and electrical machmary maintenance, the generation for preventing motor abnormality from failing.
Referring to Fig. 2, Fig. 2 is a kind of stream for permanent magnet synchronous motor performance degradation assessment method that second of embodiment provides
Cheng Tu.
A kind of permanent magnet synchronous motor performance degradation assessment method, the described method comprises the following steps：
Step S100：Permanent magnet synchronous motor and 4 of the acquisition 1 in normal condition are under different degenerate states
The motor stator current signal and vibration signal of permanent magnet synchronous motor, the vibration signal refer to displacement, speed, the acceleration of vibration
Spend signal.
The permanent magnet synchronous motor of 5 models of the same race is taken, wherein the 1st is completely new normal permanent magnet synchronous motor M1, the 2nd
It is seriously worn for bearing, the permanent magnet synchronous motor M of bearing performance high degradation2, the 3rd demagnetizes for motor permanent magnet, permanent magnet
The permanent magnet synchronous motor M of performance high degradation3, the 4th occurs the permanent magnet synchronous electric eccentric, rotor eccentricity is degenerated for rotor
Machine M4, the 5th slightly short-circuit for interturn in stator windings, the permanent magnet synchronous motor M of insulation performance high degradation5。
Normal motor is run 50 minutes, every 1 minute to stator current I1j(t) and vibration signal Vd1j(t)、Vs1j(t)、
Va1j(t) it is sampled, each sample-duration 5 seconds, wherein j=1,2 ... 50.I1j(t) indicate what the 1st motor jth time sampled
Stator current signal, Vd1j(t)、Vs1j(t)、Va1j(t) position of the motor oscillating of the 1st motor jth time sampling is indicated respectively
Shifting, speed, acceleration signal, using them as the data under motor normal condition.
Remaining 4 run 10 minutes in different degenerate state permanent magnet synchronous motors, believe stator current every 12 seconds
Number Iij(t) and vibration signal Vdij(t)、Vsij(t)、Vaij(t) it is sampled, each sample-duration 5 seconds.Wherein, i=2,3 ... 5,
J=1,2 ... 50.Iij(t) stator current signal of i-th motor jth time sampling, V are indicateddij(t)、Vsij(t)、Vaij(t)
The displacement of the motor oscillating of i-th motor jth time sampling of expression, speed, acceleration signal respectively, using them as motor
Data when different components are seriously degenerated.
Step S210：Permanent magnet synchronous motor is in normal condition and permanent magnet synchronous motor is under different degenerate states
Motor stator current signal and vibration signal carry out empirical mode decomposition, choose several IMF components of each signal.It is preferred that
Preceding 6 IMF components of each signal are chosen on ground.
To the motor stator current signal I of step S100 acquisitionsij(t) and vibration signal Vdij(t)、Vsij(t)、Vaij(t) into
Row empirical mode decomposition chooses preceding 6 IMF components (the Intrinsic Mode Function, i.e. natural mode of vibration of each signal
Function) as analysis object, totally 24 IMF components, i.e. preceding 6 IMF components of motor stator current signal, vibration displacement signal
Preceding 6 IMF components, preceding 6 IMF components of vibration velocity signal, preceding 6 IMF components of vibration acceleration signal.Motor is fixed
Electron current signal IMF components are respectively imfij1, imfij2…imfij6, motor oscillating signal IMF components are respectively imfij7,
imfij8...imfij24, wherein i=1,2 ... 5, j=1,2 ... 50.
Original signal and IMF components relationship such as following formula：
Formula (1), in (2), (3), (4), imfij1, imfij2…imfij6It is sampled when the jth time sampling for indicating i-th motor
The IMF components that the motor stator current signal come is obtained through empirical mode decomposition, Rmij6Indicate the jth time sampling of i-th of motor
When sampling come motor stator current signal through 6 obtained residual signals of empirical mode decomposition.imfij7,
imfij8...imfij24Indicate the displacement of the motor oscillating of the jth time acquisition of i-th of motor, speed, acceleration signal is through experience
The IMF components that mode decomposition obtains, Rdij6, Rsij6, Raij6The motor oscillating acquired when the jth time sampling for indicating i-th of motor
Displacement, speed, acceleration signal is through 6 obtained residual signals of empirical mode decomposition.
Step S220：Calculate the ENERGY E of all IMF componentsij1, Eij2…Eij24：
Wherein, EijkIndicate the energy of k-th of IMF component of motor signal of i-th of motor jth time sampling, imfijkd
Indicate the signal amplitude of k-th of motor signal, d-th point of the IMF components of i-th of motor jth time sampling, n is IMF signal datas
The number of point.Since sensor samples 5 seconds, sample frequency 4KHz every time, so the 5 seconds electric currents and vibration signal that acquire every time
There are 20000 data points, so n=20000, k=1,2 ... 24.
Step S230：When signal energy is larger, EijkUsual numerical value is larger, is unfavorable for post analysis calculating, and can shadow
The effect for building model in next step is rung, therefore all IMF component energies are normalized respectively.
Wherein, eijkIndicate the normalized value for k-th of IMF component energy of motor signal that i-th motor jth time sampling comes.
Step S240：Motor performance under different conditions is built respectively with the normalization IMF component energies under different conditions to move back
Change feature vector Tij=[eij1, eij2…eij24]。
Wherein, TijIndicate the feature vector that the motor signal from i-th motor jth time sampling is extracted.
Step S300：With motor performance degenerative character vector under Lagrange multiplier, gaussian kernel function and different conditions point
Not Xun Lian category support vector machines model under different conditions, and obtain different conditions drag suprasphere, obtain each model
The O of suprasphereiWith radius ri。
The feature vector T obtained with step S240ijTraining monodrome category support vector machines model, by T1j, j=1,2 ... 50
Normal as motor, performance trains motor without SVM models SVM when degenerating without sample when degenerating1.By T2j, j=1,2 ...
50 wear as motor bearings, and sample when bearing degradation is come SVM models SVM when motor bearings being trained seriously to degenerate2.It will
T3j, j=1,2 ... 50 sample when seriously demagnetizing as motor permanent magnet is come SVM moulds when motor permanent magnet being trained seriously to degenerate
Type SVM3.By T4j, j=1,2 ... 50 as rotor bias when sample train SVM moulds when rotor height bias
Type SVM4.By T5j, j=1,2 ... 50 are used as motor insulation ag(e)ings, and sample when interturn in stator windings short circuit is serious to train motor insulation
SVM models SVM when aging5。
Monodrome category support vector machines model is different from common supporting vector machine model, its basic thought is generation one
A suprasphere that can include all target sample feature vectors, can construct monodrome category support vector machines model by following formula：
O in formula (7) and (8)i, riFor i-th of motor SVMiThe suprasphere centre of sphere and radius of model, Ci, Li, ξimRespectively
I-th of motor SVMiThe punishment parameter of model, object function, relaxation factor, TimIt degenerates for i-th of motor, m-th of motor performance
Feature vector.
Introduce Lagrange multiplier αimWith gaussian kernel function K (x, y), following quadratic programming formula can be obtained：
In formula (9) and (10), αimFor i-th of motor SVMiM-th of Lagrange multiplier of model, K (x, y) indicate Gaussian kernel
Function, TinFor i-th of motor, n-th of motor performance degenerative character vector.
If sample point is inside suprasphere, αim=0；If sample point is outside suprasphere, αim=Ci；If sample
This point is just on the boundary of suprasphere, then 0≤αim≤Ci.In actual conditions, most of target samples all inside suprasphere,
Only a small amount of sample is known as supporting vector i.e. on suprasphere boundary or outside suprasphere, on suprasphere boundary and external sample
Meet 0≤αim≤CiThe sample of condition is supporting vector, after acquiring supporting vector, so as to find out the ball of SVM supraspheres
Heart OiWith radius ri, calculation formula is as follows：
Wherein, TisIndicate i-th of SVMiThe supporting vector of training pattern, αinFor i-th of motor SVMiN-th of glug of model
Bright day multiplier.
Step S410：It calculates under the permanent magnet synchronous motor model suprasphere centre of sphere under different degenerate states and normal condition
Generalized distance do between the permanent magnet synchronous motor model suprasphere centre of sphereiWith relative distance Doi。
Wherein, doiIndicate i-th of SVM model suprasphere centre ofs sphere OiWith normal condition SVM model suprasphere centre ofs sphere O1Between
Generalized distance, DoiIndicate i-th of SVM model suprasphere centre ofs sphere OiWith normal condition SVM model suprasphere centre ofs sphere O1Between
Relative distance, i=2,3,4,5.
Step S420：Acquire the motor stator current signal i of permanent magnet synchronous motor to be assessedtest(t) and vibration signal
Vdtest(t)、Vstest(t)、Vatest(t) and motor performance degenerative character vector is extracted.
Acquire the motor stator current signal i of permanent magnet synchronous motor to be assessedtest(t) and vibration signal Vdtest(t)、Vstest
(t)、Vatest(t) and according to step S210 to step S240 its feature vector T is extracted.Sample frequency 4KHz, sample-duration 5 seconds.
Step S430：Calculate the motor performance degenerative character vector T and lower point of different conditions of permanent magnet synchronous motor to be assessed
Generalized distance dt between the suprasphere centre of sphere of class support vector machines modeliWith relative distance Dti。
Relative distance DtiIt can reflect departure degree of the tested sample relative to target sample, Dt1Indicate tested motor and electricity
The departure degree of machine normal condition, distance is remoter, shows more serious, the Dt of motor performance degenerationi, i=2,3 ... 5 indicate tested electricity
The departure degree of machine and the serious degenerate state of motor various parts illustrates that certain performance degradation of motor is more serious apart from smaller.
Step S440：Judge permanent magnet synchronous motor to be assessed motor performance degenerative character vector and normal condition under classify
Whether relative distance is less than predetermined threshold value between the suprasphere centre of sphere of supporting vector machine model, if the permanent-magnet synchronous to be assessed less than if
Motor is in normal condition, on the contrary then enter step S450；
Judge Dt1Whether predetermined threshold value is less than, permanent magnet synchronous motor to be assessed is in normal condition if less than if, instead
Then enter step S450；The threshold value can rule of thumb be set.
Step S450：The motor performance degenerative character vector of permanent magnet synchronous motor more to be assessed and lower point of abnormal condition
Relative distance between the suprasphere centre of sphere of class support vector machines model, and then calculate permanent magnet synchronous motor performance degradation to be assessed and refer to
Mark.
Compare Dt2, Dt3, Dt4, Dt5Size, find out minimum Dti, i=2,3,4,5, if DtiMinimum shows motor
It is more biased towards i-th kind of failure to degenerate, calculates motor performance amount of degradation DI according to the following formula：
DI is bigger, illustrates that motor degeneration is more serious, given threshold 0.9 works as DI>When 0.9, motor performance is degenerated seriously, and
It is likely to occur i-th kind of failure.Wherein, i=2 indicates that bearing is gradually worn out to bearing fault, and i=3 indicates that permanent magnet demagnetizes to forever
Magnet failure, i=4 indicate that rotor is gradually eccentric to rotor fault, and i=5 indicates insulation ag(e)ing to interturn in stator windings failure.
Referring to Fig. 3 to Fig. 4, Fig. 3 is a kind of structure of permanent magnet synchronous motor performance degradation assessment system provided by the invention
Block diagram, Fig. 4 are the signal acquisition module schematic diagrames of permanent magnet synchronous motor performance degradation assessment system provided by the invention.
The present invention also provides a kind of permanent magnet synchronous motor performance degradation assessment systems, including signal acquisition module 1, feature
Extraction module 2, model building module 3 and performance degradation assessment module 4, wherein：
Signal acquisition module 1 is moved back for acquiring permanent magnet synchronous motor normal condition and permanent magnet synchronous motor in difference
Motor stator current signal and vibration signal under change state are sent to characteristic extracting module 2；
Characteristic extracting module 2, motor stator current signal and vibration signal for being sent according to signal acquisition module 1 carry
Motor performance degenerative character vector under different conditions is taken, model building module 3 is sent to；
Model building module 3, under the different conditions for being sent according to characteristic extracting module 2 motor performance degenerative character to
Category support vector machines model under different conditions is respectively trained in amount；
Performance degradation assessment module 4, the motor stator current signal for acquiring permanent magnet synchronous motor to be assessed and vibration
Signal extraction motor performance degenerative character vector, the permanent magnet synchronous motor using model building module 3 are under normal condition
Supporting vector machine model and predetermined threshold value judge whether permanent magnet synchronous motor to be assessed is in normal condition, if being in abnormal shape
State then calculates permanent magnet synchronous motor performance degradation index to be assessed.
It acquisition permanent magnet synchronous motor normal condition and motor stator current signal under different degenerate states and shakes
Dynamic signal.Motor stator current signal is acquired by current sensor, and vibration signal is acquired by vibrating sensor.According to acquisition
Motor stator current signal and vibration signal extraction motor be under different conditions motor performance degenerative character vectorial, use motor
Its corresponding category support vector machines model is respectively trained in motor performance degenerative character vector under different conditions.Acquisition waits for
The motor stator current signal and vibration signal extraction motor performance degenerative character vector of permanent magnet synchronous motor are assessed, and using forever
Whether magnetic-synchro motor supporting vector machine model under normal condition and predetermined threshold value judge permanent magnet synchronous motor to be assessed
In normal condition, permanent magnet synchronous motor performance degradation index to be assessed is calculated if in abnormal state.It is same by permanent magnetism
Step motor performance degeneration index judges the degree that motor performance is degenerated.The threshold value can rule of thumb be set.
The system can detect or measure the degree of motor performance degeneration during motor performance is degenerated, so as to
Targetedly to organize production and electrical machmary maintenance, the generation for preventing motor abnormality from failing.
Fig. 4 is the signal acquisition module schematic diagram of permanent magnet synchronous motor performance degradation assessment system provided by the invention.Its
220V AC conversions are that 24V direct currents are supplied to drive system, driver to use certain model AC servo by middle regulated power supply
Driver is suitble to driving 200W small-sized AC permanent magnet synchronous motors below, experiment to use certain model permanent magnet synchronous motor.Servo
Controller uses matched micro-machine controller.Stator current signal is acquired by current sensor, and motor shakes
Dynamic signal is acquired by vibrating sensor, and vibrating sensor is mounted on motor case, and sample frequency is 4KHz.
A kind of permanent magnet synchronous motor performance degradation assessment method and system provided by the present invention has been carried out in detail above
It introduces.Principle and implementation of the present invention are described for specific case used herein, the explanation of above example
It is merely used to help understand the core idea of the present invention.It should be pointed out that for those skilled in the art,
Without departing from the principles of the invention, can be with several improvements and modifications are made to the present invention, these improvement and modification are also fallen
Enter in the protection domain of the claims in the present invention.
Claims (9)
1. a kind of permanent magnet synchronous motor performance degradation assessment method, which is characterized in that the described method comprises the following steps：
Step S100：Acquisition permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are in the motor under different degenerate states
Stator current signal and vibration signal；
Step S200：The motor being according to permanent magnet synchronous motor normal condition and permanent magnet synchronous motor under different degenerate states
Stator current signal and vibration signal extraction motor are in motor performance degenerative character vector under different conditions；
Step S300：Its corresponding classification branch is respectively trained in the motor performance degenerative character vector being under different conditions with motor
Hold vector machine model；
Step S400：The motor stator current signal and vibration signal of permanent magnet synchronous motor to be assessed are acquired, and extracts its motor
Performance degradation feature vector, and sentenced using permanent magnet synchronous motor supporting vector machine model under normal condition and predetermined threshold value
Whether the permanent magnet synchronous motor to be assessed that breaks is in normal condition, and permanent magnet synchronous motor to be assessed is calculated if in abnormal state
Performance degradation index.
2. permanent magnet synchronous motor performance degradation assessment method according to claim 1, which is characterized in that step S200 is specific
For：
Step S210：Permanent magnet synchronous motor is in normal condition and permanent magnet synchronous motor is in electricity under different degenerate states
Machine stator current signal and vibration signal carry out empirical mode decomposition, choose several IMF components of each signal；
Step S220：Calculate the energy of all IMF components；
Step S230：All IMF component energies are normalized respectively；
Step S240：It is special that motor performance degeneration under different conditions is built respectively with the normalization IMF component energies under different conditions
Sign vector.
3. permanent magnet synchronous motor performance degradation assessment method according to claim 2, which is characterized in that step S300 is specific
For：Different conditions are respectively trained with motor performance degenerative character vector under Lagrange multiplier, gaussian kernel function and different conditions
Lower category support vector machines model, and obtain different conditions drag suprasphere, obtain the suprasphere of each model the centre of sphere and
Radius.
4. permanent magnet synchronous motor performance degradation assessment method according to claim 3, which is characterized in that in the step
It is specially in 400：
Step S410：It calculates in the permanent magnet synchronous motor model suprasphere centre of sphere under different degenerate states and permanent magnetism under normal condition
Generalized distance between the Synchronous Machine Models suprasphere centre of sphere and relative distance；
Step S420：It acquires the motor stator current signal of permanent magnet synchronous motor to be assessed and vibration signal and extracts motor performance
Degenerative character vector；
Step S430：Calculate the motor performance degenerative character vector of permanent magnet synchronous motor to be assessed and support of classifying under different conditions
Generalized distance between the suprasphere centre of sphere of vector machine model and relative distance；
Step S440：Judge the motor performance degenerative character vector of permanent magnet synchronous motor to be assessed and support of classifying under normal condition
Whether relative distance is less than predetermined threshold value between the suprasphere centre of sphere of vector machine model, if the permanent magnet synchronous motor to be assessed less than if
It is on the contrary then enter step S450 in normal condition；
Step S450：The motor performance degenerative character vector of permanent magnet synchronous motor more to be assessed and branch of classifying under abnormal condition
Relative distance between the suprasphere centre of sphere of vector machine model is held, and then calculates permanent magnet synchronous motor performance degradation index to be assessed.
5. permanent magnet synchronous motor performance degradation assessment method according to claim 4, which is characterized in that the step S100
Middle permanent magnet synchronous motor normal condition and permanent magnet synchronous motor are in different degenerate states：Take 5 models of the same race
Permanent magnet synchronous motor, wherein the 1st is completely new normal permanent magnet synchronous motor, the 2nd permanent magnetism for bearing performance high degradation
Synchronous motor, the 3rd permanent magnet synchronous motor for permanent magnet performance high degradation, the 4th permanent magnetism for rotor eccentricity degeneration are same
Walk motor, the 5th permanent magnet synchronous motor for insulation performance high degradation.
6. permanent magnet synchronous motor performance degradation assessment method according to claim 5, which is characterized in that the step S100
Specially：Normal motor is run 50 minutes, is sampled to stator current and vibration signal every 1 minute, each sample-duration 5
Second；Remaining 4 run 10 minutes in different degenerate state permanent magnet synchronous motors, every 12 seconds to stator current signal and shaking
Dynamic signal is sampled, each sample-duration 5 seconds.
7. permanent magnet synchronous motor performance degradation assessment method according to claim 6, which is characterized in that the vibration signal
Refer to the displacement of vibration, speed, acceleration signal.
8. permanent magnet synchronous motor performance degradation assessment method according to claim 7, which is characterized in that the step S210
Middle several IMF components for choosing each signal are specially the preceding 6 IMF components for choosing each signal.
9. a kind of permanent magnet synchronous motor performance degradation assessment system, which is characterized in that including signal acquisition module, feature extraction mould
Block, model building module and performance degradation assessment module, wherein：
Signal acquisition module is in different degenerate states for acquiring permanent magnet synchronous motor normal condition and permanent magnet synchronous motor
Under motor stator current signal and vibration signal be sent to characteristic extracting module；
Characteristic extracting module, motor stator current signal and vibration signal for being sent according to signal acquisition module extract different
Motor performance degenerative character vector, is sent to model building module under state；
Model building module, motor performance degenerative character vector difference under the different conditions for being sent according to characteristic extracting module
Category support vector machines model under training different conditions；
Performance degradation assessment module, motor stator current signal and vibration signal for acquiring permanent magnet synchronous motor to be assessed carry
Take motor performance degenerative character vectorial, using the permanent magnet synchronous motor of model building module supporting vector under normal condition
Machine model and predetermined threshold value judge whether permanent magnet synchronous motor to be assessed is in normal condition, are calculated if in abnormal state
Permanent magnet synchronous motor performance degradation index to be assessed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411798.7A CN108614940A (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor performance degradation assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411798.7A CN108614940A (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor performance degradation assessment method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108614940A true CN108614940A (en) | 2018-10-02 |
Family
ID=63661799
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810411798.7A Pending CN108614940A (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor performance degradation assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108614940A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807399A (en) * | 2019-10-29 | 2020-02-18 | 北京师范大学 | Single-category support vector machine-based collapse and slide hidden danger point detection method |
| CN110954827A (en) * | 2019-12-17 | 2020-04-03 | 北京昊鹏智能技术有限公司 | Fault diagnosis method and device, electronic equipment and system |
| CN113777487A (en) * | 2021-09-13 | 2021-12-10 | 哈尔滨工业大学 | Demagnetization environment simulation device and method for predicting aging demagnetization risk of permanent magnet motor |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145750A (en) * | 2007-09-18 | 2008-03-19 | 湖南大学 | Multi-model integrated intelligent control method of large generator group |
| CN101251579A (en) * | 2008-03-05 | 2008-08-27 | 湖南大学 | Analog circuit failure diagnosis method based on supporting vector machine |
| CN102854015A (en) * | 2012-10-15 | 2013-01-02 | 哈尔滨理工大学 | Diagnosis method for fault position and performance degradation degree of rolling bearing |
| CN104729853A (en) * | 2015-04-10 | 2015-06-24 | 华东交通大学 | Rolling bearing performance degradation evaluation device and method |
| CN104751420A (en) * | 2015-03-06 | 2015-07-01 | 湖南大学 | Sparse representation and multi-objective optimization based blind restoration method |
| CN106053066A (en) * | 2016-05-23 | 2016-10-26 | 华东交通大学 | Antifriction bearing performance degradation assessment method based on empirical mode decomposition and logistic regression |
| CN107091737A (en) * | 2017-06-06 | 2017-08-25 | 太原理工大学 | A kind of study of typical faults of rotor systems combining diagnostic method based on current signal |
| CN107219457A (en) * | 2017-06-15 | 2017-09-29 | 河北工业大学 | Frame-type circuit breaker fault diagnosis and degree assessment method based on operation annex electric current |
-
2018
- 2018-05-02 CN CN201810411798.7A patent/CN108614940A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145750A (en) * | 2007-09-18 | 2008-03-19 | 湖南大学 | Multi-model integrated intelligent control method of large generator group |
| CN101251579A (en) * | 2008-03-05 | 2008-08-27 | 湖南大学 | Analog circuit failure diagnosis method based on supporting vector machine |
| CN102854015A (en) * | 2012-10-15 | 2013-01-02 | 哈尔滨理工大学 | Diagnosis method for fault position and performance degradation degree of rolling bearing |
| CN104751420A (en) * | 2015-03-06 | 2015-07-01 | 湖南大学 | Sparse representation and multi-objective optimization based blind restoration method |
| CN104729853A (en) * | 2015-04-10 | 2015-06-24 | 华东交通大学 | Rolling bearing performance degradation evaluation device and method |
| CN106053066A (en) * | 2016-05-23 | 2016-10-26 | 华东交通大学 | Antifriction bearing performance degradation assessment method based on empirical mode decomposition and logistic regression |
| CN107091737A (en) * | 2017-06-06 | 2017-08-25 | 太原理工大学 | A kind of study of typical faults of rotor systems combining diagnostic method based on current signal |
| CN107219457A (en) * | 2017-06-15 | 2017-09-29 | 河北工业大学 | Frame-type circuit breaker fault diagnosis and degree assessment method based on operation annex electric current |
Non-Patent Citations (2)
| Title |
|---|
| 刘国飞: ""基于多信号融合的电主轴故障预警方法及系统研究"", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
| 郭伟杰: ""基于电机电流的转子系统典型故障识别方法"", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807399A (en) * | 2019-10-29 | 2020-02-18 | 北京师范大学 | Single-category support vector machine-based collapse and slide hidden danger point detection method |
| CN110954827A (en) * | 2019-12-17 | 2020-04-03 | 北京昊鹏智能技术有限公司 | Fault diagnosis method and device, electronic equipment and system |
| CN113777487A (en) * | 2021-09-13 | 2021-12-10 | 哈尔滨工业大学 | Demagnetization environment simulation device and method for predicting aging demagnetization risk of permanent magnet motor |
| CN113777487B (en) * | 2021-09-13 | 2024-04-19 | 哈尔滨工业大学 | Demagnetizing environment simulation device and method for predicting aging demagnetizing risk of permanent magnet motor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108614940A (en) | A kind of permanent magnet synchronous motor performance degradation assessment method and system | |
| CN105426905B (en) | Robot barrier object recognition methods based on histogram of gradients and support vector machines | |
| CN110132598A (en) | Slewing rolling bearing fault noise diagnostics algorithm | |
| CN104200396B (en) | A kind of wind turbine component fault early warning method | |
| Hosseinpoor et al. | Virtual sensors for fault diagnosis: A case of induction motor broken rotor bar | |
| CN109344905A (en) | A kind of transmission facility automatic fault recognition methods based on integrated study | |
| CN105021334B (en) | Fan energy consumption monitoring recognition methods based on multiple weighing value neutral net | |
| CN109324604A (en) | A kind of intelligent train resultant fault analysis method based on source signal | |
| Song et al. | A new demagnetization fault recognition and classification method for DPMSLM | |
| CN110046641B (en) | Partial recognition and lesion characterization using deep learning | |
| CN109100648A (en) | Ocean current generator impeller based on CNN-ARMA-Softmax winds failure fusion diagnosis method | |
| CN108919116A (en) | Ocean current generator imbalance stator current method for diagnosing faults based on MCCKAF-FFT-Softmax | |
| Khlaief et al. | Feature engineering for ball bearing combined-fault detection and diagnostic | |
| CN111814404A (en) | Locomotive bogie bolt loosening fault detection method | |
| Hasegawa et al. | Adaptive training of vibration-based anomaly detector for wind turbine condition monitoring | |
| CN112816200B (en) | Rotary machine axis track identification method based on deep learning | |
| Mustafa et al. | Faults classification scheme for three phase induction motor | |
| Fang et al. | Intermittent fault recognition of analog circuits in the presence of outliers via density peak clustering with adaptive weighted distance | |
| CN115906350A (en) | Micro-grid fault positioning and diagnosing method | |
| Wen et al. | Fault diagnosis of rolling bearings of different working conditions based on multi-feature spatial domain adaptation | |
| Sinha et al. | Machine learning-based explainable stator fault diagnosis in induction motor using vibration signal | |
| Senanayaka et al. | Autoencoders and data fusion based hybrid health indicator for detecting bearing and stator winding faults in electric motors | |
| Ma et al. | GIS mechanical fault diagnosis and test analysis based on S transform-SVM-AlexNet model | |
| CN115563529A (en) | Fault diagnosis method for superconducting magnet of ultra-high-speed low-vacuum pipeline magnetic levitation traffic system | |
| CN105628380A (en) | Adjustable neutrosophic model bearing fault classifier and fault classification method thereof |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181002 |
|
| RJ01 | Rejection of invention patent application after publication |