CN113985282A - Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor - Google Patents
Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor Download PDFInfo
- Publication number
- CN113985282A CN113985282A CN202111292989.4A CN202111292989A CN113985282A CN 113985282 A CN113985282 A CN 113985282A CN 202111292989 A CN202111292989 A CN 202111292989A CN 113985282 A CN113985282 A CN 113985282A
- Authority
- CN
- China
- Prior art keywords
- permanent magnet
- magnet synchronous
- dynamic
- synchronous motor
- neural network
- 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
- 230000001360 synchronised effect Effects 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003745 diagnosis Methods 0.000 title claims abstract description 33
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 31
- 238000004088 simulation Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000004907 flux Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000008878 coupling Effects 0.000 claims 1
- 238000010168 coupling process Methods 0.000 claims 1
- 238000005859 coupling reaction Methods 0.000 claims 1
- 230000005611 electricity Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 description 13
- 238000012360 testing method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 4
- 230000003068 static effect Effects 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Electric Motors In General (AREA)
- Control Of Ac Motors In General (AREA)
Abstract
The invention relates to the technical field of motor eccentric fault diagnosis, in particular to a permanent magnet synchronous motor dynamic eccentric fault diagnosis method based on an Elman neural network. The method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network comprises the following steps: s1, establishing a characteristic database of the dynamic eccentric faults of the permanent magnet synchronous motor; s2, establishing a dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman network; s3, collecting a stray magnetic field of the permanent magnet synchronous motor to be detected, and extracting a fault characteristic value; and S4, inputting the experimental data of the step S3 into the model of the step S2, and diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor. The method can accurately diagnose whether the dynamic eccentric fault of the permanent magnet synchronous motor exists or not and the dynamic eccentricity when the dynamic eccentric fault exists; the non-invasive diagnosis method does not need to modify the motor and does not influence the normal operation of the motor; the universality is high.
Description
Technical Field
The invention relates to the technical field of motor eccentric fault diagnosis, in particular to a permanent magnet synchronous motor dynamic eccentric fault diagnosis method based on an Elman neural network.
Background
The permanent magnet synchronous motor has no brush, slip ring and excitation system, simple structure, high power density and high efficiency, and is widely applied to the fields of aerospace, national defense, numerical control machines, electric automobiles and the like. In order to improve the service performance of the motor, prolong the service life of the motor, reduce the use cost of the motor and avoid serious irreversible accidents, the method for diagnosing the motor fault by adopting an effective method has important significance.
Motor eccentricity, which means that the stator axis and the rotor axis are not coincident, is a common motor fault. Defects of a motor machining and manufacturing process, errors in an assembly process, impact and abrasion in an operation process, rotor mass unbalance and the like can cause motor eccentric faults. The motor eccentricity fault can be divided into: static eccentricity, dynamic eccentricity and mixed eccentricity. The dynamic eccentricity refers to the center C of a motor rotorRAnd the center C of the motor statorSNon-coincident, the rotor of the motor not only being around its centre CRRotating around the stator center CSRotated as shown in fig. 1. Dynamic eccentricity is generally expressed as E/l0Where e is the eccentricity,/0Is the air gap length when the motor has no eccentric fault. After the motor is dynamically eccentric, the working state of a motor bearing is rapidly deteriorated, the vibration of the motor is aggravated, the torque fluctuation is obvious, and the motor can stop running even generates permanent damage when the motor is serious. In the early stage of dynamic eccentric fault of the permanent magnet synchronous motor, if the eccentric fault is diagnosed in time, the motor can be maintained as early as possible, and the fault deterioration of the motor is avoided.
The existing motor dynamic eccentric fault detection methods can be roughly divided into the following types:
the first type: and diagnosing the eccentric fault based on the vibration signal. The method needs an acceleration sensor with high precision to measure the vibration acceleration of the surface of the shell, and the eccentricity type is judged according to the characteristic frequency appearing in a motor vibration acceleration frequency spectrogram. The patent CN105698740A is to determine whether the motor has a dynamic eccentric fault or a static eccentric fault based on the additional characteristic frequency components in the vibration acceleration spectrogram. However, the causes of the abnormal vibration of the motor are complex, other interference signals are difficult to eliminate, the diagnosis effect is not ideal, and the eccentricity degree cannot be accurately diagnosed.
The second type: diagnosis of an eccentricity fault based on a current/voltage signal in the stator winding. The method can diagnose an eccentric fault of the motor based on the characteristic frequency present in the stator current/voltage. The patent CN107091986A makes wavelet decomposition on the stator current under the eccentric fault of the motor, and diagnoses the dynamic eccentric, static eccentric and mixed eccentric faults of the motor by extracting the energy value of the characteristic frequency band in the spectrogram. Patent CN109814030A is to diagnose whether the motor has dynamic eccentricity fault by detecting whether even harmonics appear on the output voltage of the stator winding of the synchronous generator. However, the method has weak fault characteristics, is greatly influenced by load, is easily submerged by noise, is not easy to extract and is difficult to detect low-degree eccentric faults.
In the third category: an off-center fault diagnosis based on the magnetic flux signal. In patent CN103713261A, magnetic field detection rings are arranged at different axial positions on the same circumferential surface inside the stator core, and the type of the eccentric fault is determined by comparing the characteristics of the magnetic field signals. According to the patent CN108614212A, Hall sensors are arranged in the axial direction at equal intervals, the axial magnetic induction intensity of the motor is obtained, and the fault type is judged according to the fault characteristic value. However, this kind of detection method requires a built-in magnetic field detection ring or a hall sensor, and belongs to an intrusive detection method, and the process is complex, and requires a large modification to the motor, and it is difficult to structurally avoid the risk of collision between the hall sensor and the rotor for a motor with a small air gap.
The fourth type: and diagnosing the eccentric fault based on the voltage signal of the detection coil. This approach typically requires pre-embedded detection coils. Patent CN107192947A is through coiling detection coil on every stator tooth, calculates the fault characteristic value according to each coil induced voltage signal in a period of time and judges the eccentric fault, but this kind of method needs to bury a plurality of detection coils on the stator tooth in advance, belongs to the intrusive detection method, and is great to the motor change, and the technology is complicated, and the commonality is poor.
The fifth type: and diagnosing the eccentric fault based on the leakage magnetic signal. Patent CN103713261a sets a detection coil at the yoke of the stator core, measures the induced voltage and makes spectrum analysis on the detection coil voltage, and determines the eccentricity type and the minimum air gap position by the characteristic frequency and amplitude. However, in the method, the detection coil needs to be arranged at the yoke part of the stator core, the motor needs to be greatly modified, the operation is difficult, and the magnitude of the eccentric amount cannot be diagnosed.
Disclosure of Invention
The invention aims to provide a method for diagnosing dynamic eccentric faults of a permanent magnet synchronous motor based on an Elman neural network, which overcomes the defects of the prior art and adopts a non-invasive diagnosis mode. The method is simple and convenient to operate, can quickly and accurately diagnose whether the permanent magnet synchronous motor has dynamic eccentric faults or not and the dynamic eccentricity of the permanent magnet synchronous motor, and provides theoretical support for motor maintenance and repair.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a permanent magnet synchronous motor dynamic eccentric fault diagnosis method based on an Elman neural network comprises the following steps:
s1, establishing a characteristic database of the dynamic eccentric faults of the permanent magnet synchronous motor:
1.1, establishing an electromagnetic simulation model of the dynamic eccentric fault of the permanent magnet synchronous motor: establishing finite element models with different dynamic eccentricities in electromagnetic finite element software, and setting the dynamic eccentricities as 0, a, 2a, 3a, …, (n-1) a, wherein a is less than or equal to 2 percent, and (n-1) a is less than or equal to 1;
1.2 acquiring the time history of the stray magnetic field of the electromagnetic simulation model under different dynamic eccentricities: at a certain speed nrThen, respectively calculating n electromagnetic simulation models with different dynamic eccentricities, and obtaining the no-load stray magnetic field time history of each finite element model;
1.3 obtaining the amplitude-frequency diagram of the stray magnetic field under different dynamic eccentricities: fundamental wave in amplitude-frequency diagram when motor has dynamic eccentric faultF appears nearby respectivelyc±frAs shown in fig. 3, and as the dynamic eccentricity increases, the amplitude of the sideband harmonic increases;
1.4, establishing a dynamic eccentric fault feature library of the permanent magnet synchronous motor: selecting the fundamental wave amplitude B of the stray magnetic field of the ith electromagnetic simulation modeli_fcLeft band harmonic amplitude Bi_fc-frAnd the right band harmonic amplitude Bi_fc+frThe characteristic value, where i is 1,2,3 … n dynamic eccentricity fault characteristic value, can be represented by the following matrix:
the corresponding dynamic eccentricity is expressed as:
s2, establishing a dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman network:
2.1 construction of Elman neural network: an Elman neural network with a structure of 3-8-8-1 is constructed, and the structure is shown in figure 6 and mainly comprises an input layer, a hidden layer, a connecting layer and an output layer; the number of nodes of an input layer is 3, the number of nodes of an implicit layer is 8, the number of nodes of a connection layer is 8, and the number of nodes of an output layer is 1;
2.2 training of Elman neural network: a fault characteristic value B consisting of the stray magnetic field fundamental wave amplitude, the left band harmonic amplitude and the right band harmonic amplitude obtained in the step 1.4n×3And corresponding dynamic eccentricity Yn×1After normalization processing, inputting the data into a neural network model for training, wherein the training function is 'trandx', the maximum iteration number is 2000, the error tolerance is 0.00001, and multiple times of iterative training are carried out to obtain a dynamic eccentric fault diagnosis model;
s3, collecting stray magnetic fields of the permanent magnet synchronous motor to be detected, and extracting fault characteristic values:
3.1 connect the no-load stray magnetic field collection instrument of PMSM, install the magnetic density sensor, measure the stray magnetic field on casing surface: the connection mode of the permanent magnet synchronous motor to be detected and the permanent magnet synchronous motor no-load stray magnetic field acquisition instrument is shown in fig. 2, wherein a magnetic density sensor is installed on the surface of the machine shell, as shown in fig. 4, and the acquired data are guaranteed to be radial magnetic densities of stray magnetic fields;
3.2 measuring the time history of the stray magnetic field, and extracting fault characteristics: the motor to be tested is reversely dragged to a certain fixed rotating speed nrAnd measuring and recording the radial magnetic flux density of the stray magnetic field in the time from T to T + T, wherein T is a synchronous electrical period, and extracting the fundamental wave amplitude B 'from the measured experimental data after fast Fourier transform'i_fcAnd the harmonic amplitude B 'on the left'i_fc-frAnd the harmonic amplitude B 'on the right'i_fc+frForm a fault signature matrix B'1×3：
B'1×3＝[B'1_fc B'1_fc-fr B'1_fc+fr]
S4, inputting the experimental data of step S3 to the model of step S2, and diagnosing the dynamic eccentricity fault of the permanent magnet synchronous motor:
b'1×3And after normalization processing, the dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman neural network established in the step S2 is called as the input of the neural network, the output value of the neural network is calculated, and the dynamic eccentric ratio output by the diagnosis model is obtained after the calculation result is subjected to inverse normalization.
Further, in step 3.1, the no-load stray magnetic field collecting instrument of the permanent magnet synchronous motor comprises a test bench, a fixture, a servo motor, a torque and speed sensor, a tesla meter, a data collecting device, a computer and a motor control cabinet, wherein at least two fixtures are arranged on the test bench, one fixture clamps the fixed servo motor, the other fixture clamps the fixed permanent magnet synchronous motor to be detected, the servo motor is electrically connected with the motor control cabinet arranged on one side of the test bench, the servo motor is connected with the permanent magnet synchronous motor to be detected through a coupler and the torque and speed sensor, the tesla meter for collecting the stray magnetic field is arranged on a stator shell of the permanent magnet synchronous motor to be detected, the tesla meter is electrically connected with the data collecting device, and the data collecting device is electrically connected with the computer.
The invention has the beneficial effects that: compared with the prior art, the method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network has the following advantages:
(1) the dynamic eccentricity of the permanent magnet synchronous motor can be accurately diagnosed when the dynamic eccentric fault exists;
(2) the invention is a non-invasive diagnosis method of the dynamic eccentric fault of the permanent magnet synchronous motor, does not need to modify the motor, has simple and convenient operation and can not influence the normal operation of the motor;
(3) the method has high universality, is suitable for all types of permanent magnet synchronous motors, and can diagnose the dynamic eccentric faults of the motors only by establishing a dynamic eccentric fault diagnosis feature library of the type of motors according to different types of motors.
Drawings
FIG. 1 is a schematic view of dynamic eccentricity;
FIG. 2 is a schematic view of a stray magnetic field acquisition device;
FIG. 3 is a schematic of fundamental and sideband harmonics at 50% dynamic eccentricity;
FIG. 4 is a schematic view of a magnetic field position sensor mounting;
FIG. 5 is a dynamic eccentricity fault diagnosis flow chart;
FIG. 6 is a schematic diagram of an Elman neural network;
FIG. 7 is a graphical illustration of dynamic eccentricity fault diagnosis results;
in fig. 2, a motor control cabinet 1, a first clamp 2, a servo motor 3, a coupler 4, a torque and speed sensor 5, a second clamp 6, a teslameter 7, a permanent magnet synchronous motor to be measured 8, a data acquisition device 9 and a computer 10 are provided.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
Example 1
Aiming at a certain 8-pole 48-slot permanent magnet synchronous motor, the method detects the dynamic eccentric fault of the motor according to the whole process of the method, and comprises the following steps:
step S1: establishing characteristic database of dynamic eccentric faults of permanent magnet synchronous motor
1.1, establishing an electromagnetic simulation model of the dynamic eccentric fault of the permanent magnet synchronous motor:
establishing finite element models with different dynamic eccentricities in electromagnetic finite element software, and sequentially setting the dynamic eccentricities as 0, 0.02, 0.04, 0.06 and … 0.98.98 for 50 groups of data;
1.2 acquiring the time history of the stray magnetic field of the electromagnetic simulation model under different dynamic eccentricities:
under the rotating speed of 5rpm, respectively calculating 50 electromagnetic simulation models with different dynamic eccentricities, and obtaining the no-load stray magnetic field time history of each finite element model;
1.3 fast Fourier decomposition is carried out on the stray magnetic field of each simulation model to obtain the fundamental wave amplitude B of the stray magnetic field under different dynamic eccentricitiesi_fcLeft band harmonic amplitude Bi_fc-frAnd the right band harmonic amplitude Bi_fc+frWherein i is 1,2,3, … 50. Constructing a dynamic eccentricity fault eigenvalue matrix B50×3Comprises the following steps:
the corresponding dynamic eccentricity is expressed as:
step S2: elman network-based dynamic eccentric fault diagnosis model for permanent magnet synchronous motor
The structure is shown in fig. 6, the model mainly comprises an input layer, a hidden layer, a connection layer and an output layer, the input layer is provided with 3 nodes, the number of the nodes of the hidden layer and the connection layer is 8, and the number of the nodes of the output layer is 1; will imitateA fault characteristic value B consisting of truly obtained stray magnetic field fundamental wave amplitude, left band harmonic amplitude and right band harmonic amplitude50×3And the dynamic eccentricity Y after normalization processing50×1Respectively serving as input and output, substituting the input and the output into a neural network model for training, wherein the training function is 'trandx', the maximum iteration number is 2000 times, and the error tolerance is 0.00001, so as to obtain a dynamic eccentric fault diagnosis model;
step S3: collecting stray magnetic field of permanent magnet synchronous motor to be detected and extracting fault characteristic value
3.1 connect data acquisition instrument, install the magnetic density sensor, measure the stray magnetic field on casing surface:
the no-load stray magnetic field acquisition device of the permanent magnet synchronous motor is shown in fig. 2, a servo motor 3 is fixed on a test bench 11 through a first clamp 2, the servo motor 3 is controlled by a motor control cabinet 1, the left end of the servo motor is connected with a permanent magnet synchronous motor 8 to be detected through a coupler 4 and a torque and speed sensor 5, the permanent magnet synchronous motor is fixed on the test bench through a second clamp 6, a tesla meter 7 for acquiring a stray magnetic field is installed on a stator shell, the stray magnetic field is input into a computer 10 after being acquired through a data acquisition device 9, and the computer 10 is used for recording, analyzing and acquiring experimental data;
3.2 according to the flow shown in fig. 5, the following process is executed for the dynamic eccentric fault diagnosis of the permanent magnet synchronous motor to be tested:
a magnetic field sensor is arranged on a stator shell of the permanent magnet synchronous motor and used for collecting a stray magnetic field; reversely dragging the permanent magnet synchronous motor to be tested to the rotating speed of 5rpm by the servo motor in the figure 2, and collecting the time history of the stray magnetic field in one synchronous electric period; extracting fundamental wave amplitude B 'from the measured experimental data after fast Fourier transform'i_fcAnd the harmonic amplitude B 'on the left'i_fc-frAnd the harmonic amplitude B 'on the right'i_fc+frConstitute matrix B'1×3：
B'1×3＝[B'1_fc B'1_fc-fr B'1_fc+fr]
Step S4: inputting experimental data into the model established in step S2, diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor:
b'1×3And after normalization processing, the dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman neural network established in the step S2 is called as the input of the neural network, the output value of the neural network is calculated, and the dynamic eccentricity is obtained after the calculation result is subjected to inverse normalization.
Fig. 7 is a diagram illustrating a dynamic eccentricity fault diagnosis result. The total number of the training samples is 40, the number of the test samples is 10, and the dynamic eccentricity prediction value of the fault diagnosis model is relatively consistent with the actual value, the fault diagnosis precision is 98.18%, and the mean square error of the test data is 0.000144 according to the graph.
The above embodiments are only specific examples of the present invention, and the protection scope of the present invention includes but is not limited to the product forms and styles of the above embodiments, and any suitable changes or modifications made by those skilled in the art according to the claims of the present invention shall fall within the protection scope of the present invention.
Claims (6)
1. A permanent magnet synchronous motor dynamic eccentric fault diagnosis method based on an Elman neural network is characterized by comprising the following steps: the method comprises the following steps:
s1, establishing a characteristic database of the dynamic eccentric faults of the permanent magnet synchronous motor:
1.1, establishing an electromagnetic simulation model of the dynamic eccentric fault of the permanent magnet synchronous motor;
1.2, acquiring the time history of the stray magnetic field of the electromagnetic simulation model under different dynamic eccentricities;
1.3 obtaining an amplitude-frequency diagram of the stray magnetic field under different dynamic eccentricities;
1.4 establishing a dynamic eccentric fault feature library of the permanent magnet synchronous motor;
s2, establishing a dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman network;
s3, collecting a stray magnetic field of the permanent magnet synchronous motor to be detected, and extracting a fault characteristic value;
and S4, inputting the experimental data of the step S3 into the model of the step S2, and diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor.
2. The method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network as claimed in claim 1, wherein the method comprises the following steps:
1.1 of the step S1 includes the following steps: establishing finite element models with different dynamic eccentricities in electromagnetic finite element software, and setting the dynamic eccentricities as 0, a, 2a, 3a, …, (n-1) a, wherein a is less than or equal to 2 percent, and (n-1) a is less than or equal to 1;
1.2 of the step S1 includes the following steps: at a certain speed nrThen, respectively calculating n electromagnetic simulation models with different dynamic eccentricities, and obtaining the no-load stray magnetic field time history of each finite element model;
1.3 of the step S1 includes the following steps: when the motor has dynamic eccentric fault, f appears near the fundamental wave in the amplitude-frequency diagramc±frThe amplitude of the sideband harmonic wave is increased along with the increase of the dynamic eccentricity;
1.4 of the step S1 includes the following steps: selecting the fundamental wave amplitude B of the stray magnetic field of the ith electromagnetic simulation modeli_fcLeft band harmonic amplitude Bi_fc-frAnd the right band harmonic amplitude Bi_fc+frThe characteristic value, where i is 1,2,3 … n dynamic eccentricity fault characteristic value, can be represented by the following matrix:
the corresponding dynamic eccentricity is expressed as:
3. the method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network as claimed in claim 2, wherein the method comprises the following steps: the step S2 includes the following steps:
2.1 construction of Elman neural network: constructing an Elman neural network with a structure of 3-8-8-1, wherein the Elman neural network mainly comprises an input layer, a hidden layer, a connecting layer and an output layer; the number of nodes of an input layer is 3, the number of nodes of an implicit layer is 8, the number of nodes of a connection layer is 8, and the number of nodes of an output layer is 1;
2.2 training of Elman neural network: a fault characteristic value B consisting of the stray magnetic field fundamental wave amplitude, the left band harmonic amplitude and the right band harmonic amplitude obtained in the step 1.4n×3And corresponding dynamic eccentricity Yn×1After normalization processing, inputting the data into a neural network model for training, wherein the training function is 'trandx', the maximum iteration times is 2000 times, the error tolerance is 0.00001, and multiple times of iterative training are carried out to obtain a dynamic eccentric fault diagnosis model.
4. The method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network as claimed in claim 3, wherein the method comprises the following steps: the step S3 includes the following steps:
3.1 connect the no-load stray magnetic field collection instrument of PMSM, install the magnetic density sensor, measure the stray magnetic field on casing surface: the magnetic density sensor is arranged on the surface of the machine shell, and ensures that the acquired data are radial magnetic densities of stray magnetic fields;
3.2 measuring the time history of the stray magnetic field, and extracting fault characteristics: the motor to be tested is reversely dragged to a certain fixed rotating speed nrAnd measuring and recording the radial magnetic flux density of the stray magnetic field in the time from T to T + T, wherein T is a synchronous electrical period, and extracting the fundamental wave amplitude B 'from the measured experimental data after fast Fourier transform'i_fcAnd the harmonic amplitude B 'on the left'i_fc-frAnd the harmonic amplitude B 'on the right'i_fc+frForm a fault signature matrix B'1×3：
B'1×3＝[B'1_fc B'1_fc-fr B'1_fc+fr]。
5. Permanent magnet synchronous electricity based on Elman neural network according to claim 4The method for diagnosing the dynamic eccentric fault of the machine is characterized by comprising the following steps: the step S4 includes the following steps: b'1×3And after normalization processing, the dynamic eccentric fault diagnosis model of the permanent magnet synchronous motor based on the Elman neural network established in the step S2 is called as the input of the neural network, the output value of the neural network is calculated, and the dynamic eccentric ratio output by the diagnosis model is obtained after the calculation result is subjected to inverse normalization.
6. The method for diagnosing the dynamic eccentric fault of the permanent magnet synchronous motor based on the Elman neural network as claimed in claim 4, wherein the method comprises the following steps: permanent magnet synchronous motor no-load stray magnetic field acquisition instrument includes the laboratory bench, anchor clamps, servo motor, torque speed sensor, tesla meter, data acquisition device, computer and motor control cabinet, be provided with two at least anchor clamps on the laboratory bench, servo motor is fixed to one of them anchor clamps centre gripping, the permanent magnet synchronous motor that awaits measuring is fixed to another anchor clamps centre gripping, servo motor and the motor control cabinet electric connection who sets up in laboratory bench one side, servo motor passes through the shaft coupling and torque speed sensor and is connected with the permanent magnet synchronous motor that awaits measuring, the tesla meter of gathering stray magnetic field is installed on the permanent magnet synchronous motor's that awaits measuring stator shell, tesla meter and data acquisition device electric connection, data acquisition device and computer electric connection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111292989.4A CN113985282A (en) | 2021-11-03 | 2021-11-03 | Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111292989.4A CN113985282A (en) | 2021-11-03 | 2021-11-03 | Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113985282A true CN113985282A (en) | 2022-01-28 |
Family
ID=79746096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111292989.4A Pending CN113985282A (en) | 2021-11-03 | 2021-11-03 | Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113985282A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706335A (en) * | 2009-11-11 | 2010-05-12 | 华南理工大学 | Wind power forecasting method based on genetic algorithm optimization BP neural network |
| CN112881910A (en) * | 2021-01-20 | 2021-06-01 | 哈尔滨工业大学（威海） | Dynamic eccentric fault detection method for rotating permanent magnet synchronous motor |
| CN113093005A (en) * | 2021-04-06 | 2021-07-09 | 哈尔滨工业大学（威海） | Static eccentric fault diagnosis method for permanent magnet synchronous motor |
| CN113094952A (en) * | 2021-04-06 | 2021-07-09 | 哈尔滨工业大学（威海） | Stray magnetic field-based static eccentricity detection method for permanent magnet motor |
-
2021
- 2021-11-03 CN CN202111292989.4A patent/CN113985282A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706335A (en) * | 2009-11-11 | 2010-05-12 | 华南理工大学 | Wind power forecasting method based on genetic algorithm optimization BP neural network |
| CN112881910A (en) * | 2021-01-20 | 2021-06-01 | 哈尔滨工业大学（威海） | Dynamic eccentric fault detection method for rotating permanent magnet synchronous motor |
| CN113093005A (en) * | 2021-04-06 | 2021-07-09 | 哈尔滨工业大学（威海） | Static eccentric fault diagnosis method for permanent magnet synchronous motor |
| CN113094952A (en) * | 2021-04-06 | 2021-07-09 | 哈尔滨工业大学（威海） | Stray magnetic field-based static eccentricity detection method for permanent magnet motor |
Non-Patent Citations (1)
| Title |
|---|
| 陈勇;梁洪;王成栋;梁思远;钟荣强;: "基于改进小波包变换和信号融合的永磁同步电机匝间短路故障检测", 电工技术学报, no. 1 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110703091B (en) | Static eccentric fault detection method for built-in permanent magnet synchronous motor for electric automobile | |
| CN113093005B (en) | Static eccentric fault diagnosis method for permanent magnet synchronous motor | |
| CN112881910B (en) | Dynamic eccentric fault detection method for rotating permanent magnet synchronous motor | |
| CN102998591B (en) | Positioning method of generator rotor winding dynamic inter-turn short circuit fault | |
| CN103823150B (en) | Turbo generator rotor interturn short circuit fault diagnosis method of multi sensor joint | |
| CN103217644B (en) | Induction electric machine rotor defect diagonsis device, method and medium | |
| CN109541461B (en) | Permanent magnet synchronous motor eccentric fault diagnosis method based on magnetic field distribution monitoring | |
| Zhu et al. | Vold–Kalman filtering order tracking based rotor demagnetization detection in PMSM | |
| Ahmed et al. | Detection of eccentricity faults in machine usingfrequency spectrum technique | |
| CN110501640B (en) | Method for detecting static eccentricity of permanent magnet motor based on air gap magnetic field direct test | |
| Ehya et al. | Static and dynamic eccentricity fault diagnosis of large salient pole synchronous generators by means of external magnetic field | |
| CN113391235B (en) | System and method for detecting dynamic turn-to-turn short circuit fault of synchronous generator rotor | |
| KR20220079876A (en) | Fault detection in synchronous machines | |
| CN110716152A (en) | Method for monitoring turn-to-turn short circuit of generator by impedance frequency spectrum | |
| CN110470212B (en) | Dynamic eccentric fault detection method for hub motor | |
| CN115410729A (en) | Online monitoring method, system and medium for vibration of shaft seal pump of nuclear reactor | |
| CN210072012U (en) | Wind driven generator air gap eccentricity fault detection device | |
| Xu et al. | Online detecting magnet defect fault in PMSG with magnetic sensing | |
| CN113311328B (en) | Rotor static eccentricity diagnosis method based on vibration characteristics of stator-winding system | |
| Ma et al. | Open-circuit radial stray magnetic flux density based noninvasive diagnosis for mixed eccentricity parameters of interior permanent magnet synchronous motors in electric vehicles | |
| CN113985282A (en) | Elman neural network-based dynamic eccentric fault diagnosis method for permanent magnet synchronous motor | |
| Sabbatini et al. | Data acquisition and processing for tip timing and operational modal analysis of turbomachinery blades | |
| CN106403880A (en) | Method and device for detecting clearance between rotor and stator of compressor | |
| Ranga et al. | Advanced tool based condition monitoring of induction machines by using LabVIEW—A review | |
| Misra et al. | An experimental study of rotor fault detection using motor current signature analysis based on neural networks |
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 |