Industrial motor stator and rotor electrical fault diagnosis method and device and medium

Abstract

The invention relates to an industrial motor stator and rotor electrical fault diagnosis method and device and a medium, and belongs to the technical field of industrial motor fault diagnosis. The method comprises the following steps: determining the fault characteristics of the industrial motor according to the working mechanism of the industrial motor; the stator current of the industrial motor is collected, and the motor current amplitude is calculated through coordinate transformation; performing feature extraction on the fault current of the industrial motor; training and fault diagnosis are carried out by adopting a support vector machine; and outputting the diagnosis fault type of the industrial motor. By means of the technical means disclosed by the invention, electrical faults of the industrial motor stator and rotor can be found in time.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial motor fault diagnosis, and in particular, relates to an industrial motor stator and rotor electrical fault diagnosis method, device and medium. Background Art

[0002] The stable operation of industrial asynchronous motors plays a vital role in the safe production and manufacturing of enterprises. However, under frequent starting, stopping and speed regulation, the stator and rotor winding circuits of asynchronous motors are prone to generate huge energy. If the heat cannot be dissipated in time, thermal aging is prone to occur. How to detect early faults in time and take appropriate emergency measures is an important issue facing R&D personnel. Summary of the invention

[0003] The present disclosure provides an industrial motor stator and rotor electrical fault diagnosis method, device and medium to solve the above technical problems.

[0004] According to a first aspect of the present disclosure, a method for diagnosing electrical faults of stators and rotors of industrial motors is provided, the method comprising: clarifying the fault characteristics of the industrial motor according to the working mechanism of the industrial motor; collecting the stator current of the industrial motor and calculating the motor current amplitude by coordinate transformation; extracting features of the fault current of the industrial motor; using a support vector machine for training and fault diagnosis; and outputting the diagnostic fault type of the industrial motor.

[0005] In some embodiments, the fault characteristics of the industrial motor are clarified according to the working mechanism of the industrial motor, including: when a phase winding of the stator of the industrial asynchronous motor has a turn-to-turn short circuit fault, the three-phase stator winding will become asymmetric; at the same time, the three-phase stator current induces an elliptical rotating magnetic field in the air gap, and the magnetic field can be decomposed into a forward rotating stator circular rotating magnetic field and a reverse rotating stator circular rotating electric field; the forward rotating magnetic field induces a frequency of The induction current of the rotor generates a rotating magnetic field of the rotor, thereby generating a constant electromagnetic torque; the reversal of the stator rotating magnetic field will generate a negative sequence component in the stator winding; when the rotor of the industrial asynchronous motor is broken, the rotor winding is in an asymmetric operating state, The induced current flowing through the rotor winding at a high frequency will generate an elliptical rotating magnetic field in the air gap; a reverse magnetic field will be generated during asymmetric operation. The harmonic component of the frequency appears in the stator current; when the rotor winding breaks, there is a harmonic component of the frequency in the stator current. The basic fault characteristic component of the current component interacts with the air gap magnetic flux to produce pulsating torque, which in turn produces pulsating speed, causing the angle between the stator and the rotor to change, further causing the magnetic flux phase of the stator winding to change. The air gap rotating magnetic field contains not only the fundamental wave component, but also the frequencies of and The air gap harmonic magnetic field component generates two harmonic induced electromotive forces in the stator winding, which can be expressed as The harmonic induced electromotive force of the two stator windings forms a frequency of and The two harmonic induced currents.

[0006] In some embodiments, the calculation of the motor current amplitude by using coordinate transformation includes: transforming the three-phase stator current Transformed into two-dimensional coordinate representation and ; After obtaining the two-phase coordinate system, the two-phase coordinate system is converted into a single-phase signal represented by two-dimensional coordinates.

[0007] In some embodiments, the three-phase stator current is converted into Transformed into two-dimensional coordinate representation and ,include:

[0008] According to the formula: , , and get the two-dimensional coordinate representation and .

[0009] In some embodiments, after obtaining the two-phase coordinate system, converting the two-phase coordinate system into a single-phase signal represented by two-dimensional coordinates includes: according to the formula: , and get a single-phase signal.

[0010] In some embodiments, the feature extraction of the fault current of the industrial motor includes: the three-phase stator current signal of the faulty motor is expressed as ,in, represents the amplitude of the positive sequence component, represents the amplitude of the negative sequence component, represents the power frequency component, represents the initial phase of the positive sequence component, represents the initial phase of the negative sequence component; , Including DC component and twice the power frequency component. After filtering the DC component, the fault characteristic frequency is clearly displayed; the three-phase stator current signal of the faulty motor is expressed as

[0011] ,

[0012] Among them, f s represents the power frequency component, I1 represents the power frequency component f sThe amplitude of I2 represents the power frequency component (l-2s)f s The amplitude of I3 represents the power frequency component (1+2s)f s The amplitude of θ1, θ2, and θ3 represent the power frequency components f s , power frequency component (1-2s)f s and power frequency component (1+2s)f s The initial phase angle of

[0013] , and according to the formula: , and obtain the signal to be detected after removing the DC component.

[0014] In some embodiments, the use of a support vector machine for training and fault diagnosis includes: during an offline training process, collecting historical data on stator turn-to-turn short-circuit fault modes and rotor fracture fault modes under different operating conditions and loads based on simulation; preprocessing the data and dividing it into training data and test data in a ratio of 7:3; using an intelligent algorithm to train a fault classifier, and using a 5-fold cross-validation method to calculate the training and test accuracy of the algorithm; adjusting the parameters of the model to complete the acquisition of the fault classifier; performing the same data preprocessing as the offline process on the collected fault data to obtain an output vector of the fault classification model, performing calculations based on the input classification, and the output result is a fault type diagnosis result.

[0015] In some embodiments, the expression of the support vector machine is , by introducing the penalty factor C and the slack variable ξ i ,get , where ξ≥0, and the Lagrange multiplier a is introduced i (i =1, 2, ..., m) and kernel function K (x i , x j ) and then we get ,in, .

[0016] According to a second aspect of the present disclosure, there is provided an industrial motor stator and rotor electrical fault diagnosis device, comprising: a determination module, used to clarify the fault characteristics of the industrial motor according to the working mechanism of the industrial motor; a calculation module, used to collect the stator current of the industrial motor and calculate the motor current amplitude by using coordinate transformation; an extraction module, used to extract features of the fault current of the industrial motor; a diagnosis module, used to perform training and fault diagnosis by using a support vector machine; and an output module, used to output the diagnostic fault type of the industrial motor.

[0017] According to a third aspect of the present disclosure, a computer storable medium is provided, on which computer program instructions are stored, and when the instructions are executed by a processor, the above-mentioned industrial motor stator and rotor electrical fault diagnosis method is implemented.

[0018] Advantages of the present invention: The present invention is a method for diagnosing electrical faults of stators and rotors of industrial motors based on a support vector machine. The support vector machine learns the mapping relationship between fault types and fault characteristics from historical fault data, and can identify different faults under different load and speed conditions.

[0019] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0021] Figure 1 It is a flow chart of the method for diagnosing electrical faults of stators and rotors of industrial motors.

[0022] Figure 2 It is a schematic diagram of the coordinate transformation of the three-phase current of the stator of a faulty motor.

[0023] Figure 3 It is the framework of intelligent diagnosis algorithm for stator and rotor electrical faults of industrial motors.

[0024] Figure 4 It is a schematic diagram of the three-phase stator current waveform of an industrial asynchronous motor (normal operation of an industrial asynchronous motor).

[0025] Figure 5 It is a schematic diagram of the three-phase stator current waveform of an industrial asynchronous motor (stator turn-to-turn short circuit fault).

[0026] Figure 6 It is a schematic diagram of the three-phase stator current waveform of an industrial asynchronous motor (rotor fracture fault).

[0027] Figure 7 This is a schematic diagram of the detection signal of an industrial asynchronous motor during normal operation.

[0028] Figure 8 It is a schematic diagram of detection signals when a stator inter-turn short circuit fault occurs.

[0029] Fig. 9 This is a schematic diagram of the detection signal when the rotor breaks.

[0030] Fig.10 This is a diagram of the confusion matrix. DETAILED DESCRIPTION

[0031] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0032] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0033] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0035] like Figure 1 As shown, the method for diagnosing electrical faults of stators and rotors of industrial motors includes: clarifying the fault characteristics of industrial motors according to the working mechanism of industrial motors; collecting the stator current of industrial motors and calculating the motor current amplitude by coordinate transformation; extracting the characteristics of the fault current of industrial motors; using support vector machines for training and fault diagnosis; and finally, outputting the diagnostic fault type of industrial motors.

[0036] Common fault characteristics of industrial motors include stator turn-to-turn short circuit faults and rotor bar breakage faults.

[0037] When a phase winding of the stator of an industrial asynchronous motor has a turn-to-turn short circuit fault, the three-phase stator winding will become asymmetric; at the same time, the three-phase stator current induces an elliptical rotating magnetic field in the air gap, which can be decomposed into a positive stator circular rotating magnetic field and a reverse stator circular rotating electric field; the positive rotating magnetic field induces a frequency of sf in the rotor. s The induced current in the rotor generates a rotating magnetic field in the rotor, thereby generating a constant electromagnetic torque; reversing the stator rotating magnetic field will produce a negative sequence component in the stator winding, which can be used as a basis for identifying and diagnosing faults.

[0038] When the rotor of an industrial asynchronous motor breaks, the rotor winding is in an asymmetric operating state. s The induced current flowing through the rotor winding at a frequency will generate an elliptical rotating magnetic field in the air gap; asymmetric operation will produce a reverse magnetic field, which leads to a (1-2s)f s The harmonic components of the frequency appear in the stator current, so the frequency is (1-2s)f s The harmonic components of the stator current can be used as the basis for rotor fracture fault diagnosis.

[0039] When the rotor winding breaks, there is a frequency of (1-2s)f in the stator current. s The basic fault characteristic component of the current component interacts with the air gap magnetic flux to produce pulsating torque, which in turn produces pulsating speed. It will cause the angle between the stator and the rotor to change, and ultimately cause the magnetic flux phase of the stator winding to change. At this time, the air gap rotating magnetic field contains not only the fundamental wave component, but also the frequencies of (1-2s)f s and (1+2s)f s The air gap harmonic magnetic field component generates two harmonic induced electromotive forces in the stator winding, which are expressed as: The harmonic induced electromotive force of the above two stator windings forms a frequency of (1-2s)f in the closed stator winding. s and (1+2s)f s The two harmonic induced currents are generated by the reverse magnetomotive force in the rotating magnetic field of the air gap when the rotor breaks, so they can be used as a basis for diagnosing motor rotor breakage.

[0040] like Figure 2 As shown in the figure, the motor current amplitude is calculated by coordinate transformation: the coordinate transformation used is applied to the stator turn-to-turn short circuit fault model and the rotor fracture fault model. The principle is to transform the three-phase stator current (i A ,i B ,i C ) is transformed into i in the two-dimensional coordinate representation d and i q After obtaining the two-phase coordinate system, it is necessary to convert it into a single-phase signal represented by a two-dimensional coordinate system. Therefore, I s =i d +ji q The modulus square is calculated and finally converted into a detection signal with time as the horizontal axis and the modulus square as the vertical axis;

[0041] The three-phase coordinate system is converted into a two-phase coordinate system through coordinate transformation. The conversion formula is: , the transformation matrix from the two-phase stationary coordinate system to the two-phase rotating coordinate system is , and then taking the square of the transformation modulus, we get .

[0042] Characterizing the fault current of industrial motors:

[0043] According to the above fault mechanism analysis, the negative sequence component of the three-phase stator current is used as the fault characteristic signal. The three-phase stator current signal of the faulty motor can be expressed as: , where I1 is the amplitude of the positive sequence component, I2 is the amplitude of the negative sequence component, ω is the power frequency component, θ1 is the initial phase of the positive sequence component, and θ2 is the initial phase of the negative sequence component.

[0044] By converting, we get , it can be seen that I s It only contains DC component and twice the power frequency component. After filtering the DC component, the fault characteristic frequency can be clearly displayed.

[0045] In the detection of motor rotor fracture fault, the fault characteristic frequency component (1±2s)f s Used as a fault characteristic signal, the collected three-phase stator current signal of the faulty motor can be expressed as: , where f s is the power frequency component; I1, I2 and I3 are the power frequency components f s 、(l-2s)f s and (1+2s)f s The amplitude of θ1, θ2, and θ3 are the power frequency components f s 、(1-2s)f s Component sum (1+2s)f s The initial phase angle of the component.

[0046] By transforming, we can get In order to eliminate the influence of DC components on the detection results and observe the side frequency components in the detection results more intuitively, it is necessary to s for further processing;

[0047] According to I s The properties of the expression are averaged to obtain the mean value (I s ), the calculation result only includes I s The DC component in the signal can then be obtained by using the equation to remove the DC component. The equation is: .

[0048] The support vector machine is used for training and fault diagnosis, which is divided into two parts: offline training and online diagnosis.

[0049] In the offline training process, firstly, historical data of stator turn-to-turn short-circuit fault mode and rotor fracture fault mode under different operating conditions and loads are collected based on simulation; after preprocessing the data, it is divided into training data and test data in a ratio of 7:3; then, an intelligent algorithm is used to train the fault classifier, and a 5-fold cross-validation method is used to calculate the training and test accuracy of the algorithm; finally, the fault classifier is acquired by adjusting and optimizing the parameters of the model; in online diagnosis, the collected fault data is first preprocessed in the same way as the offline process to obtain the output vector of the fault classification model, which is calculated according to the input classification, and the output result is the fault type diagnosis result;

[0050] The fault data of industrial asynchronous motors is difficult to obtain in actual work. The amount of fault data is relatively small and belongs to small sample information. Therefore, support vector machine is selected as the classifier for fault diagnosis. The basic expression of support vector machine is: , by introducing the penalty factor C and the slack variable ξ i , can get , where ξ≥0, and the Lagrange multiplier is introduced and kernel function After that, you can get ,in, .

[0051] In order to better train the classifier and make it suitable for fault waveform recognition under different working conditions and different load conditions. The fault asynchronous motor model is written using the S function in Matlab / Simulink, and the motor control strategy is vector control. The positive three-phase current data of the motor under normal operating conditions, the stator inter-turn short circuit fault and rotor fracture fault mode under variable speed and variable load conditions are collected, and a historical database is established. The parameters of the industrial asynchronous motor are shown in the following table:

[0052] Table 1. Motor parameters

[0053]

[0054] The load torque of the industrial asynchronous motor drive system is set to 900 N•m, and the given speed is 1000r / min. Taking this as an example, the simulation obtains the three-phase stator current under normal motor, stator turn short circuit fault mode and rotor fracture fault mode, as shown in Figure 4 , Figure 5 ,and Figure 6 .

[0055] This method converts the three-phase stator current into a one-dimensional detection signal through coordinate transformation and vector modulus square function, thereby completing the extraction of fault features. The detection signals under the motor normal, stator turn short circuit fault mode and rotor fracture fault mode are as follows: Figure 7 , Figure 8 , Fig. 9 shown.

[0056] In order to verify the performance of the offline training classifier, 420 sets of new test data were obtained by setting different fault times, different fault modes and different operating conditions of the system. After coordinate transformation and fault feature extraction, the data were input into the classifier for fault type identification and location. The confusion matrix output by the classifier is shown in Fig.10 As shown, the test accuracy is 99.3%.

[0057] The present disclosure also provides an industrial motor stator and rotor electrical fault diagnosis device, including: a determination module, used to clarify the fault characteristics of the industrial motor according to the working mechanism of the industrial motor; a calculation module, used to collect the stator current of the industrial motor and calculate the motor current amplitude using coordinate transformation; an extraction module, used to extract features of the fault current of the industrial motor; a diagnosis module, used to use a support vector machine for training and fault diagnosis; and an output module, used to output the diagnostic fault type of the industrial motor.

[0058] The present disclosure also provides a computer storable medium having computer program instructions stored thereon, which, when executed by a processor, implement the above-mentioned industrial motor stator and rotor electrical fault diagnosis method.

[0059] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for diagnosing electrical faults of industrial motor stators and rotors, characterized in that: The method comprises: According to the working mechanism of industrial motors, clarify the fault characteristics of industrial motors; Collect the stator current of industrial motors and calculate the motor current amplitude using coordinate transformation; Extract features of fault current of industrial motors; Support vector machines are used for training and fault diagnosis; Outputs diagnostic fault types of industrial motors.

2. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 1, characterized in that: According to the working mechanism of industrial motors, the fault characteristics of industrial motors are clarified, including: When a phase winding of the stator of an industrial asynchronous motor has a turn-to-turn short circuit fault, the three-phase stator winding will become asymmetric; at the same time, the three-phase stator current induces an elliptical rotating magnetic field in the air gap, which can be decomposed into a circular rotating magnetic field of the forward stator and a circular rotating electric field of the reverse stator; the forward rotating magnetic field induces an elliptical rotating magnetic field with a frequency of The induced current of the rotor generates a rotating magnetic field of the rotor, thereby generating a constant electromagnetic torque; the reversal of the stator rotating magnetic field will generate a negative sequence component in the stator winding; When the rotor of an industrial asynchronous motor breaks, the rotor winding is in an asymmetric operating state. The induced current flowing through the rotor winding at a high frequency will generate an elliptical rotating magnetic field in the air gap; a reverse magnetic field will be generated during asymmetric operation. Harmonic components of the frequency appear in the stator current; When the rotor winding breaks, there is a frequency in the stator current. The basic fault characteristic component of the current component interacts with the air gap magnetic flux to produce pulsating torque, which in turn produces pulsating speed, causing the angle between the stator and the rotor to change, further causing the magnetic flux phase of the stator winding to change. The air gap rotating magnetic field contains not only the fundamental wave component, but also the frequencies of and The air gap harmonic magnetic field component generates two harmonic induced electromotive forces in the stator winding, which can be expressed as The harmonic induced electromotive force of the two stator windings forms a frequency of and The two harmonic induced currents.

3. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 2, characterized in that: The method of calculating the motor current amplitude by using coordinate transformation includes: Based on Park transformation, the three-phase stator current Transformed into two-dimensional coordinate representation and ; After the two-phase coordinate system is obtained, the two-phase coordinate system is converted into a single-phase signal represented by a two-dimensional coordinate system.

4. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 3, characterized in that: The three-phase stator current is converted into Transformed into two-dimensional coordinate representation and ,include: According to the formula: , , and get the two-dimensional coordinate representation and .

5. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 4, characterized in that: After the two-phase coordinate system is obtained, the two-phase coordinate system is converted into a single-phase signal represented by a two-dimensional coordinate system, including: According to the formula: , and get a single-phase signal.

6. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 5, characterized in that: The feature extraction of the fault current of the industrial motor includes: The three-phase stator current signal of the faulty motor is expressed as ,in, represents the amplitude of the positive sequence component, represents the amplitude of the negative sequence component, represents the power frequency component, represents the initial phase of the positive sequence component, represents the initial phase of the negative sequence component; , Including DC component and twice the power frequency component. After filtering the DC component, the fault characteristic frequency is clearly displayed; The three-phase stator current signal of the faulty motor is expressed as ， Among them, f s represents the power frequency component, I1 represents the power frequency component f s The amplitude of I2 represents the power frequency component (l-2s)f s The amplitude of I3 represents the power frequency component (1+2s)f s The amplitude of θ1, θ2, and θ3 represent the power frequency components f s , power frequency component (1-2s)f s and power frequency component (1+2s)f s The initial phase angle of , and according to the formula: , and obtain the signal to be detected after removing the DC component.

7. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 1, characterized in that: The support vector machine is used for training and fault diagnosis, including: During the offline training process, historical data of stator turn-to-turn short-circuit fault mode and rotor fracture fault mode under different operating conditions and loads are collected based on simulation; After preprocessing, the data is divided into training data and test data in a ratio of 7:3; Use intelligent algorithms to train fault classifiers, and use a 5-fold cross-validation method to calculate the training and testing accuracy of the algorithm; Adjust the parameters of the model to obtain the fault classifier; The collected fault data is preprocessed in the same way as the offline process to obtain the output vector of the fault classification model. The calculation is performed based on the input classification, and the output result is the fault type diagnosis result.

8. The method for diagnosing electrical faults of stators and rotors of industrial motors according to claim 7, characterized in that: The expression of the support vector machine is , by introducing the penalty factor C and the slack variable ξ i ,get , where ξ≥0, and the Lagrange multiplier a is introduced i (i =1, 2,..., m) and kernel function K (x i , x j ) and then we get ,in, .

9. An industrial motor stator and rotor electrical fault diagnosis device, characterized in that: include: A determination module is used to identify the fault characteristics of the industrial motor according to the working mechanism of the industrial motor; A calculation module is used to collect the stator current of the industrial motor and calculate the motor current amplitude using coordinate transformation; An extraction module, used for extracting features of fault current of industrial motors; A diagnostic module for training and fault diagnosis using a support vector machine; Output module, used to output the diagnostic fault type of industrial motors.

10. A computer storable medium, characterized in that: Computer program instructions are stored thereon, and when the instructions are executed by a processor, the method for diagnosing electrical faults of stators and rotors of industrial motors as claimed in any one of claims 1 to 8 is implemented.

Images