CN117949821A - Asynchronous motor fault diagnosis method and related equipment - Google Patents
Asynchronous motor fault diagnosis method and related equipment Download PDFInfo
- Publication number
- CN117949821A CN117949821A CN202410048910.0A CN202410048910A CN117949821A CN 117949821 A CN117949821 A CN 117949821A CN 202410048910 A CN202410048910 A CN 202410048910A CN 117949821 A CN117949821 A CN 117949821A
- Authority
- CN
- China
- Prior art keywords
- vibration signal
- asynchronous motor
- fault diagnosis
- noise
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000003745 diagnosis Methods 0.000 title claims abstract description 60
- 238000012706 support-vector machine Methods 0.000 claims abstract description 38
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000011156 evaluation Methods 0.000 claims abstract description 18
- 230000003044 adaptive effect Effects 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 12
- 241000283153 Cetacea Species 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 28
- 230000009467 reduction Effects 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 17
- 238000009499 grossing Methods 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 4
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 3
- 238000013016 damping Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000001052 transient effect Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Abstract
The invention provides an asynchronous motor fault diagnosis method and related equipment, comprising the following steps: collecting an original vibration signal of a target asynchronous motor; denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal; decomposing the noise-reduced vibration signal through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components; calculating the kurtosis value of each eigenmode component, taking the preset kurtosis as an evaluation index, and screening the evaluation index to obtain the optimal eigenmode component; extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes; inputting a plurality of feature vector matrixes into a support vector machine model which is optimized and trained by an improved whale optimization algorithm for fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor; compared with the prior art, the accuracy of fault type diagnosis is improved.
Description
Technical Field
The invention relates to the technical field of motor fault diagnosis, in particular to an asynchronous motor fault diagnosis method and related equipment.
Background
As one of the most widely used power equipment in industry, motor faults have great influence on production efficiency and safety, and fault diagnosis is an important task in modern production. Common motor faults include rotor bar breaks, bearing faults, air gap eccentricities, and the like.
At present, the traditional motor fault diagnosis method and some new methods which are raised in recent years have been developed, for example, the application of a fault tree diagnosis method in motor fault diagnosis has achieved some remarkable results, and a fuzzy technology and a fault tree analysis method are combined to provide a motor fault diagnosis method of a fuzzy fault tree, and the motor fault diagnosis method is applied to the temperature fault diagnosis of a generator set system, and experimental research shows that the method is feasible and effective; on the basis of the traditional support vector machine, a multi-stage binary tree classifier suitable for motor fault diagnosis is constructed by integrating a fuzzy clustering technology and a support vector machine algorithm.
The traditional motor fault diagnosis method is mainly based on a signal processing method of vibration signals, such as frequency domain analysis, time domain analysis, wavelet analysis and the like; the method for diagnosing the vibration faults of the hydroelectric generating set by using the frequency spectrum method and the wavelet neural network is simple and effective. However, the vibration signal acquired under the actual operation of the motor may contain a lot of noise, and it is difficult to detect the operation state of the motor and accurately identify the type of failure of the motor only by the time-frequency domain analysis method.
Disclosure of Invention
The invention provides an asynchronous motor fault diagnosis method and related equipment, and aims to improve the accuracy of fault type diagnosis.
In order to achieve the above object, the present invention provides a fault diagnosis method for an asynchronous motor, comprising:
step 1, collecting an original vibration signal of a target asynchronous motor;
Step 2, denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal;
Step 3, decomposing the vibration signal after noise reduction through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
Step 4, calculating the kurtosis value of each intrinsic mode component, taking the preset kurtosis as an evaluation index, and screening the evaluation index to obtain the optimal intrinsic mode component;
Step 5, extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes;
step 6, optimizing and training the support vector machine model through an improved whale optimizing algorithm to obtain a trained support vector machine model;
and 7, inputting the plurality of feature vector matrixes into the trained support vector machine model for fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor.
Further, step 2 includes:
Constructing a sparse dictionary on the original vibration signal through Laplace wavelet;
Performing preliminary noise reduction on the original vibration signal by a lower bound convolution smoothing method to obtain a vibration signal after preliminary noise reduction and calculating a soft threshold of the vibration signal after preliminary noise reduction;
constructing G-log penalty through the vibration signal after preliminary noise reduction;
searching a sparse solution based on a sparse dictionary, G-log penalty, soft threshold and forward and backward splitting algorithm;
And recombining the vibration signal after preliminary noise reduction by using sparse decomposition to obtain the vibration signal after noise reduction.
Further, the expression ψ (f,ζ,τ) (t) of the laplace wavelet is:
Where f represents frequency, ζ represents damping ratio, τ represents time index, and W s represents wavelet support length.
Further, the lower bound convolution smoothing method expressionThe method comprises the following steps:
Wherein, Represents the smoothing kernel, μ represents the smoothing parameter, x represents the sparse representation coefficient, v represents the index variable of the convolution kernel.
Further, the expression ψ G-log (x) of the G-log penalty is:
where B represents a matrix that maintains the convexity of the function.
Further, the expression of the forward-backward splitting algorithm is:
Where sgn denotes the sign of the aggregate value,
Further, the improved adaptive noise set empirical mode decomposition includes:
adding Gaussian white noise to the defined original signal to obtain
x(i)＝x+β0E1[w(i)]
Wherein E 1[w(i) is Gaussian white noise, x is an original signal, x (i) is a signal sequence constructed by adding the Gaussian white noise, and beta 0 is a weighting coefficient of the Gaussian white noise;
The local mean value of x (i) is calculated by using an empirical mode decomposition algorithm, and a first residual error r 1 is obtained by calculation as follows:
wherein I is the number of data points in the original signal x, M is the local mean value of the signal generated based on the empirical mode decomposition algorithm;
calculating the IMF value of the first eigenmode function The method comprises the following steps:
The second residual r 2 is calculated as:
second eigenmode function IMF value The calculation formula is as follows:
Similarly, the kth residual r k:
calculating the IMF value of the kth eigenmode function
Stopping iteration until all the IMF values of the eigenmode functions and the corresponding residual errors are obtained, and ending the decomposition.
The invention also provides an asynchronous motor fault diagnosis device, which comprises:
the acquisition module is used for acquiring original vibration signals of the target asynchronous motor in normal operation and different fault states;
The noise reduction module is used for reducing noise of the original vibration signal through generalized logarithmic sparse regularization to obtain a noise-reduced vibration signal;
the decomposition module is used for decomposing the vibration signal after noise reduction through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
the screening module is used for calculating the kurtosis value of each intrinsic mode component, taking the preset kurtosis as an evaluation index, and screening the optimal intrinsic mode component through the evaluation index;
The extraction module is used for extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes;
The training module is used for carrying out optimization training on the support vector machine model through an improved whale optimization algorithm to obtain a trained support vector machine model;
the diagnosis module is used for inputting the feature vector matrix under different fault states into the trained support vector machine model to carry out fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor.
The present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements an asynchronous motor fault diagnosis method.
The invention also provides a terminal device which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the fault diagnosis method of the asynchronous motor when executing the computer program.
The scheme of the invention has the following beneficial effects:
The method acquires an original vibration signal of a target asynchronous motor; denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal; decomposing the noise-reduced vibration signal through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components; calculating the kurtosis value of each eigenmode component, taking the preset kurtosis as an evaluation index, and screening the evaluation index to obtain the optimal eigenmode component; extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes; inputting a plurality of feature vector matrixes into the trained support vector machine model for fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor; compared with the prior art, the sparse property can be enhanced and the noise interference can be reduced through generalized logarithmic sparsity; the problem that the accurate fault type cannot be accurately judged only through general time-frequency domain analysis under the condition that a vibration signal acquired by the asynchronous motor in an actual working environment contains a large amount of noise is solved, and the accuracy of fault type diagnosis is improved.
Other advantageous effects of the present invention will be described in detail in the detailed description section which follows.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention;
FIG. 2 is a graph of the diagnostic results of a WOA optimized support vector machine model of the prior art;
FIG. 3 is a graph of the diagnostic results of an improved WOA-optimized support vector machine model in accordance with an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages to be solved more apparent, the following detailed description will be given with reference to the accompanying drawings and specific embodiments. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, a locked connection, a removable connection, or an integral connection; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
The invention provides an asynchronous motor fault diagnosis method and related equipment aiming at the existing problems.
As shown in fig. 1, an embodiment of the present invention provides a fault diagnosis method for an asynchronous motor, including:
step 1, collecting an original vibration signal of a target asynchronous motor;
Step 2, denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal;
Step 3, decomposing the vibration signal after noise reduction through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
Step 4, calculating the kurtosis value of each intrinsic mode component, taking the preset kurtosis as an evaluation index, and screening the evaluation index to obtain the optimal intrinsic mode component;
Step 5, extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes;
step 6, optimizing and training the support vector machine model through an improved whale optimizing algorithm to obtain a trained support vector machine model;
and 7, inputting the plurality of feature vector matrixes into the trained support vector machine model for fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor.
Specifically, the embodiment of the invention collects original vibration signals of the asynchronous motor under four different working conditions such as normal state, rotor breakage, bearing fault, air gap eccentricity and the like through the vibration sensor as a training set for training the support vector machine model.
Specifically, step2 includes:
Constructing a sparse dictionary on the original vibration signal through Laplace wavelet;
Performing preliminary noise reduction on the original vibration signal by a lower bound convolution smoothing method to obtain a vibration signal after preliminary noise reduction and calculating a soft threshold of the vibration signal after preliminary noise reduction;
constructing G-log penalty through the vibration signal after preliminary noise reduction;
searching a sparse solution based on a sparse dictionary, G-log penalty, soft threshold and forward and backward splitting algorithm;
and recombining the vibration signal after preliminary noise reduction by using sparse decomposition to obtain the vibration signal after noise reduction.
In the embodiment of the invention, firstly, original vibration signals are read, a sparse representation model is established, and original vibration signals of an asynchronous motor in normal operation and different fault states are collected according to the sampling frequency of 12000Hz, wherein the vibration signals collected in practice contain noise, so that the collected vibration signals can represent superposition of repeated transient and noise, namely:
y＝y0+n
wherein y represents the measured vibration signal, y 0 represents the repeated transient caused by the local fault, and y, y 0∈RN, n represents the noise;
The repetitive transient y 0 caused by the local fault is sparsely represented by a suitable dictionary D, namely
y0＝Dx
Wherein x represents sparse coefficient representation, and the most sparse, i.e. optimal coefficient x opt, is subjected to coefficient regularization calculation, i.e.
Wherein λ represents a non-negative real number;
The sparse dictionary is constructed by using Laplace wavelet, and the expression psi (f,ζ,τ) (t) of the Laplace wavelet is as follows:
Wherein f represents frequency, ζ represents damping ratio, ζ ε [0,1 ], τ represents time index, and W s represents wavelet support length; the optimal parameters of the Laplace wavelet are determined by a related filtering method, the method compares the signals under different parameters with the inner products of the Laplace wavelet, and when the inner products reach the maximum, the optimal parameters are selected; after finding the best laplace wavelet as an atom, a sparse dictionary may ultimately be constructed by varying the interval of the sampling period.
Smoothing by lower bound convolution to further improve noise reduction and to construct G-log penalties for functions h andFrom R N to R U { + infinity is a function of }, lower bound convolution definition:
Wherein, Representing smooth kernels,/>Γ 0(RN) represents from R N to RU++ infinity, μ represents the smoothing parameter and is greater than 0, v represents the index variable of the convolution kernel;
the smoothing kernel of the lower bound convolution smoothing is selected as:
wherein log represents the natural logarithm, assuming x ε R, the expression of the smoothing function |x| The method comprises the following steps:
the smoothing function is subjected to multivariate scaling generalization, expressed as:
Wherein x, v e R n,B∈Rm×n represents a matrix which maintains convexity of the smoothing function, B i represents an ith row of matrix B, and q represents the number of rows of matrix B;
the G-log penalty is constructed by L1 norm minus the lower bound convolution of the multivariate proportional generalization function above, whose expression ψ G-log (x) is:
Wherein B represents a matrix that maintains convexity of the function;
Searching for a sparse solution through a forward-backward split FBS algorithm, and reconstructing a vibration signal according to a mean square error evaluation;
By using the G-log penalty, the function F (x) is expressed as:
wherein matrix B needs to satisfy the following two conditions:
Considering that the G-log penalty is convex, the G-log regularization problem can be solved by introducing a convex optimization algorithm by rewriting the equation for solving the thin solution into the saddle point problem, and the rewritten equation is as follows:
Wherein the method comprises the steps of
Finding a lean fluffer by an FBS algorithm, and obtaining according to the formula:
Where sgn denotes the sign of the aggregate value,
Calculating a soft threshold value of the vibration signal after preliminary noise reduction, wherein the expression is as follows:
soft(x,S)＝x·max{0,1-S/|x|}
According to the soft threshold and the FBS algorithm, the lean fluffing is obtained, and then a recombined vibration signal is obtained, namely a vibration signal after noise reduction is obtained, and the expression is as follows:
x(i+1)＝soft(z(i),ωλ)
v(i+1)＝soft(s(i),ωλ)
wherein ω represents a real number between 0 and 2.
And evaluating the accuracy of the recombined vibration signal according to the mean square error.
Specifically, step3 includes:
The vibration signal after noise reduction is decomposed into a plurality of intrinsic mode components through improved adaptive noise set empirical mode decomposition, and the specific process is as follows:
defining x as the original signal to be decomposed, let M be the operator for calculating the local mean, E k be the kth modal component generated by the empirical mode decomposition (EMPIRICAL MODE DECOMPOSITION, EMD) algorithm, the expression of the relationship between the two is:
E1x＝x-M(x)
adding gaussian white noise to the defined original signal yields:
x(i)＝x+β0E1[w(i)]
Where E 1[w(i) is Gaussian white noise, x (i) is a signal sequence constructed by adding Gaussian white noise, and β 0 is a weighting coefficient of the Gaussian white noise, which can be expressed as:
wherein std is standard deviation, epsilon 0 is the inverse of the signal-to-noise ratio between the first added Gaussian white noise and the original signal;
The local mean value of x (i) is calculated by using an empirical mode decomposition algorithm EMD, and a first residual error r 1 is obtained through calculation, namely, a residual signal obtained by removing the IMF value of the first eigenmode function after signal decomposition is as follows:
calculating the IMF value of the first eigenmode function The method comprises the following steps:
The second residual r 2 is calculated as:
second eigenmode function IMF value The calculation formula is as follows:
Similarly, the kth residual r k:
calculating the IMF value of the kth eigenmode function
Stopping iteration until all the IMF values of the eigenmode functions and the corresponding residual errors are obtained, and ending the decomposition.
Specifically, extracting the characteristic of the time-frequency domain entropy value of the optimal eigenmode component refers to extracting multi-scale permutation entropy of the optimal eigenmode component, forming a characteristic vector by the multi-scale permutation entropy, and endowing the characteristic vector matrix to a label construction characteristic vector matrix.
Specifically, the process of optimizing and training the support vector machine model through the improved whale optimizing algorithm is as follows:
Dividing the decomposed data into 200 groups, wherein each group comprises 600 data points, and extracting multi-scale arrangement entropy as a feature vector matrix;
dividing the feature vector matrix into a training set and a testing set according to the proportion of 8:2;
Taking a global optimal position obtained through an improved whale optimization algorithm as a punishment parameter in a support vector machine, and taking an optimal fitness value as a nuclear parameter in the support vector machine;
Inputting a feature vector matrix in a training set into a support vector machine model, and outputting corresponding fault state labels which are respectively 1-normal state, 2-rotor broken bar, 3-bearing fault and 4-air gap eccentricity by the support vector machine model;
the penalty factor and the kernel function parameters of the support vector machine model are respectively valued to be 0.01 and 0.1 according to experience;
Inputting the training set into a support vector machine model for training, and obtaining the support vector machine model meeting training conditions after training is completed;
Inputting the test set into a trained support vector machine model for testing to obtain a fault diagnosis junction of the target asynchronous motor, wherein the diagnosis result is shown in table 1:
TABLE 1
| Fault type | Number of test sets | Classification result/% |
| Normal state | 40 | 100 |
| Rotor broken bar | 40 | 97.5 |
| Bearing failure | 40 | 100 |
| Air gap eccentricity | 40 | 100 |
In order to make the experimental result more convincing, the embodiment of the invention inputs the same data into the WOA optimized support vector machine model for fault identification, as shown in fig. 3, and as can be seen from fig. 2 and 3, the improved WOA optimized support vector machine model in the embodiment of the invention has higher accuracy.
Extracting a characteristic vector matrix in an original vibration signal of a target asynchronous motor, inputting the characteristic vector matrix into a support vector machine model meeting training conditions, and performing fault diagnosis to obtain a fault diagnosis result of the target asynchronous motor, wherein the fault diagnosis result can be motor normal operation, rotor bar breakage, bearing fault or air gap eccentricity.
The method and the device acquire the original vibration signals of the target asynchronous motor; denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal; decomposing the noise-reduced vibration signal through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components; calculating the kurtosis value of each eigenmode component, taking the preset kurtosis as an evaluation index, and screening the evaluation index to obtain the optimal eigenmode component; extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes; inputting a plurality of feature vector matrixes into a support vector machine model which is optimized and trained by an improved whale optimization algorithm for fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor; compared with the prior art, the sparse property can be enhanced and the noise interference can be reduced through generalized logarithmic sparsity; the problem that the accurate fault type cannot be accurately judged only through general time-frequency domain analysis under the condition that a vibration signal acquired by the asynchronous motor in an actual working environment contains a large amount of noise is solved, and the accuracy of fault type diagnosis is improved.
The embodiment of the invention also provides a fault diagnosis device of the asynchronous motor, which comprises the following steps:
the acquisition module is used for acquiring an original vibration signal of the target asynchronous motor;
The noise reduction module is used for reducing noise of the original vibration signal through generalized logarithmic sparse regularization to obtain a noise-reduced vibration signal;
the decomposition module is used for decomposing the vibration signal after noise reduction through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
the screening module is used for calculating the kurtosis value of each intrinsic mode component, taking the preset kurtosis as an evaluation index, and screening the optimal intrinsic mode component through the evaluation index;
The extraction module is used for extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of eigenvector matrixes;
The training module is used for carrying out optimization training on the support vector machine model through an improved whale optimization algorithm to obtain a trained support vector machine model;
the diagnosis module is used for inputting the plurality of feature vector matrixes into the trained support vector machine model to perform fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor.
It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present invention, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiments of the present invention. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements an asynchronous motor fault diagnosis method.
The integrated modules, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the present invention may implement all or part of the flow of the above-described method embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and the computer program may implement the steps of the above-described method embodiments when executed by a processor. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to construct an apparatus/terminal equipment, recording medium, computer Memory, read-Only Memory (ROM), random access Memory (RAM, random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
The invention also provides a terminal device which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the fault diagnosis method of the asynchronous motor when executing the computer program.
The terminal equipment can be a desktop computer, a notebook computer, a palm computer, a server cluster, a cloud server and other computing equipment. The terminal device may include, but is not limited to, a processor, a memory.
The Processor referred to may be a central processing unit (CPU, central Processing Unit) or other general purpose Processor, digital signal Processor (DSP, digital Signal Processor), application SPECIFIC INTEGRATED Circuit (ASIC), field-Programmable GateArray or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may in some embodiments be an internal storage unit of the terminal device, such as a hard disk or a memory of the terminal device. The memory may in other embodiments also be an external storage device of the terminal device, such as a plug-in hard disk provided on the terminal device, a smart memory card (SMC, smart Media Card), a Secure Digital (SD) card, a flash memory card (FLASH CARD), etc. Further, the memory may also include both an internal storage unit and an external storage device of the terminal device. The memory is used for storing an operating system, application programs, boot Loader (Boot Loader), data, other programs, etc., such as program code of the computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.
It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present invention, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiments of the present invention. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.
Claims (10)
1. A method for diagnosing faults of an asynchronous motor, comprising:
step 1, collecting an original vibration signal of a target asynchronous motor;
step 2, denoising the original vibration signal through generalized logarithmic sparse regularization to obtain a denoised vibration signal;
Step 3, decomposing the noise-reduced vibration signal through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
step 4, calculating the kurtosis value of each intrinsic mode component, and screening by taking the preset kurtosis as an evaluation index to obtain the optimal intrinsic mode component;
step 5, extracting the time-frequency domain entropy value characteristics of the optimal eigenvector components to obtain a plurality of characteristic vector matrixes;
step 6, optimizing and training the support vector machine model through an improved whale optimizing algorithm to obtain a trained support vector machine model;
And 7, inputting the plurality of feature vector matrixes into a trained support vector machine model to perform fault diagnosis, and obtaining a fault diagnosis result of the target asynchronous motor.
2. The method for diagnosing faults of an asynchronous motor according to claim 1, wherein the step 2 comprises:
Constructing a sparse dictionary for the original vibration signal through Laplace wavelet;
Performing preliminary noise reduction on the original vibration signal by a lower bound convolution smoothing method to obtain a vibration signal subjected to preliminary noise reduction, and calculating a soft threshold value of the vibration signal subjected to preliminary noise reduction;
constructing G-log penalty through the vibration signal after preliminary noise reduction;
finding a sparse solution based on the sparse dictionary, the G-log penalty, the soft threshold, and a forward-backward splitting algorithm;
and recombining the primarily noise-reduced vibration signal by using sparse decomposition to obtain a noise-reduced vibration signal.
3. The asynchronous motor fault diagnosis method according to claim 2, wherein the expression ψ (f,ζ,τ) (t) of the laplace wavelet is:
Where f represents frequency, ζ represents damping ratio, τ represents time index, and W s represents wavelet support length.
4. A fault diagnosis method for an asynchronous motor according to claim 3, wherein the expression of the lower bound convolution smoothing methodThe method comprises the following steps:
Wherein, Represents the smoothing kernel, μ represents the smoothing parameter, x represents the sparse representation coefficient, v represents the index variable of the convolution kernel.
5. The method of claim 4, wherein the G-log penalty is expressed as a value of ψ G-log (x):
where B represents a matrix that maintains the convexity of the function.
6. The asynchronous motor fault diagnosis method according to claim 5, wherein the expression of the forward-backward splitting algorithm is:
Where sgn denotes the sign of the aggregate value,
7. The method of claim 6, wherein the improved adaptive noise set empirical mode decomposition comprises:
adding Gaussian white noise to the defined original signal to obtain
x(i)＝x+β0E1[w(i)]
Wherein E 1[w(i) is Gaussian white noise, x is an original signal, and x (i) is a signal sequence constructed after Gaussian white noise is added; beta 0 is the weighting coefficient of Gaussian white noise;
The local mean value of x (i) is calculated by using an empirical mode decomposition algorithm, and a first residual error r 1 is obtained by calculation as follows:
wherein I is the number of data points in the original signal x, M is the local mean value of the signal generated based on the empirical mode decomposition algorithm;
calculating the IMF value of the first eigenmode function The method comprises the following steps:
The second residual r 2 is calculated as:
second eigenmode function IMF value The calculation formula is as follows:
Similarly, the kth residual r k:
calculating the IMF value of the kth eigenmode function
Stopping iteration until all the IMF values of the eigenmode functions and the corresponding residual errors are obtained, and ending the decomposition.
8. An asynchronous motor fault diagnosis device, characterized by comprising:
the acquisition module is used for acquiring an original vibration signal of the target asynchronous motor;
the noise reduction module is used for reducing noise of the original vibration signal through generalized log sparse regularization to obtain a noise-reduced vibration signal;
the decomposition module is used for decomposing the noise-reduced vibration signal through improved adaptive noise set empirical mode decomposition to obtain a plurality of intrinsic mode components;
The screening module is used for calculating the kurtosis value of each intrinsic mode component, taking the preset kurtosis as an evaluation index, and screening the optimal intrinsic mode component through the evaluation index;
the extraction module is used for carrying out time-frequency domain entropy value feature extraction on the optimal eigenvector components to obtain a plurality of feature vector matrixes;
The training module is used for carrying out optimization training on the support vector machine model through an improved whale optimization algorithm to obtain a trained support vector machine model;
and the diagnosis module is used for inputting the plurality of feature vector matrixes into the trained support vector machine model to perform fault diagnosis, so as to obtain a fault diagnosis result of the target asynchronous motor.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the asynchronous motor fault diagnosis method according to any one of claims 1 to 7 is implemented when the computer program is executed by a processor.
10. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the asynchronous motor fault diagnosis method according to any of claims 1 to 7 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410048910.0A CN117949821A (en) | 2024-01-12 | 2024-01-12 | Asynchronous motor fault diagnosis method and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410048910.0A CN117949821A (en) | 2024-01-12 | 2024-01-12 | Asynchronous motor fault diagnosis method and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117949821A true CN117949821A (en) | 2024-04-30 |
Family
ID=90797576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410048910.0A Pending CN117949821A (en) | 2024-01-12 | 2024-01-12 | Asynchronous motor fault diagnosis method and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117949821A (en) |
-
2024
- 2024-01-12 CN CN202410048910.0A patent/CN117949821A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105275833B (en) | CEEMD (Complementary Empirical Mode Decomposition)-STFT (Short-Time Fourier Transform) time-frequency information entropy and multi-SVM (Support Vector Machine) based fault diagnosis method for centrifugal pump | |
| AU2020214409A1 (en) | Structural damage identification method based on ensemble empirical mode decomposition and convolution neural network | |
| CN110657985B (en) | Gearbox fault diagnosis method and system based on singular value spectrum manifold analysis | |
| CN112766342A (en) | Abnormity detection method for electrical equipment | |
| Liu et al. | Feature extraction of rotor fault based on EEMD and curve code | |
| CN113191240B (en) | Multi-task deep neural network method and device for bearing fault diagnosis | |
| CN110705456A (en) | Micro motor abnormity detection method based on transfer learning | |
| KR101964412B1 (en) | Method for diagnosing anomaly log of mobile commmunication data processing system and system thereof | |
| CN111680642B (en) | Terrain classification method and device | |
| CN111679654A (en) | Electric drive system vibration signal diagnosis method and device based on neural network | |
| CN111767521A (en) | Oil transfer pump rolling bearing state evaluation method based on convolutional neural network and long-short term memory network | |
| CN114037478A (en) | Advertisement abnormal flow detection method and system, electronic equipment and readable storage medium | |
| CN113390631A (en) | Fault diagnosis method for gearbox of diesel engine | |
| CN117150359A (en) | Small sample fault diagnosis method, system, device and medium based on model independent element learning | |
| CN116578843A (en) | Centrifugal pump diagnostic model training method, diagnostic method, system, device and medium | |
| CN113295413B (en) | Traction motor bearing fault diagnosis method based on indirect signals | |
| Zhang et al. | Bearing fault diagnosis with kernel sparse representation classification based on adaptive local iterative filtering-enhanced multiscale entropy features | |
| CN114118174A (en) | Rolling bearing fault diagnosis method and system | |
| CN111382792B (en) | Rolling bearing fault diagnosis method based on double-sparse dictionary sparse representation | |
| CN112801315A (en) | State diagnosis method and device for power secondary equipment and terminal | |
| CN110988673B (en) | Motor rotor fault detection method and device and terminal equipment | |
| CN117949821A (en) | Asynchronous motor fault diagnosis method and related equipment | |
| CN109239006B (en) | Substance identification method and device based on humidity compensation model and storage medium | |
| CN113128612B (en) | Processing method of abnormal value in power data and terminal equipment | |
| CN111597096B (en) | Benchmark test method, system and terminal equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |