CN110567721A - rolling bearing fault diagnosis method and system - Google Patents
rolling bearing fault diagnosis method and system Download PDFInfo
- Publication number
- CN110567721A CN110567721A CN201911033248.7A CN201911033248A CN110567721A CN 110567721 A CN110567721 A CN 110567721A CN 201911033248 A CN201911033248 A CN 201911033248A CN 110567721 A CN110567721 A CN 110567721A
- Authority
- CN
- China
- Prior art keywords
- fault
- rolling bearing
- sample set
- clustering
- energy
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
the invention discloses a rolling bearing fault diagnosis method and a system, wherein the method comprises the following steps: acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set; extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set; carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm; and extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
Description
Technical Field
the invention relates to the technical field of fault diagnosis of industrial equipment, in particular to a rolling bearing fault diagnosis method and system based on a fuzzy C-means clustering algorithm of an improved particle swarm optimization algorithm.
Background
since the 21 st century, with the rapid development of national economy, the industry of China is rapidly developed, and the method becomes a world large industrial country and greatly promotes the development of various industries of China. Among them, the rapid development of industry is not independent of the use of industrial equipment in large quantities, and the rolling bearing is one of the key components of industrial equipment, and the operating condition of the rolling bearing directly affects the reliability of the whole system, so that it becomes more important to perform condition monitoring and fault diagnosis.
the PSO algorithm is an intelligent optimization algorithm which is inspired by bird swarm flying and foraging behaviors by Kennedy and the like, has a simple structure and strong operability, is convenient to implement, and obtains the attention and research of a plurality of scholars, but the PSO algorithm is premature, and easily falls into local extrema when processing high-dimensional complex functions, and has low search precision and the like.
Clustering analysis is an important data mining and pattern recognition method, aims to find a cluster structure contained in a data set, clusters the data set according to data attributes, and is widely applied to the field of fault diagnosis. The K-means algorithm is a classical pattern recognition algorithm, is simple and high in convergence rate, and is widely applied to the academic and industrial fields. However, the clustering result of the algorithm is easily affected by the initial clustering center, and is easily involved in the situation of local optimal solution. The fault of the industrial equipment is usually a gradual process, so the extracted fault characteristics are usually fuzzy, and the fault diagnosis directly through the fault characteristics has certain difficulty. The fuzzy clustering analysis method provides an effective solution for solving the problems. The FCM algorithm is optimization and expansion of a K-means algorithm, and the basic idea is to increase fuzzy membership degree on the basis of the K-means algorithm, so that the algorithm can avoid the condition of trapping a local optimal solution.
the inventor finds that the existing industrial equipment fault diagnosis method has the following problems in the research and development process:
(1) the traditional PSO algorithm is easy to fall into local extremum when processing high-dimensional complex functions, and has the problems of low convergence speed, low search precision and the like;
(2) The fault of the industrial equipment is generally a gradual process, so that the extracted fault characteristics are generally fuzzy, fault diagnosis directly through the fault characteristics is difficult, and the traditional FCM algorithm does not consider the influence of the overall variation on the distance, so that the fault diagnosis accuracy is low.
Disclosure of Invention
in order to overcome the defects of the prior art, the invention provides a rolling bearing fault diagnosis method and a rolling bearing fault diagnosis system based on an FCM algorithm of an improved PSO algorithm, wherein the speed and the position of particles are updated by the improved PSO algorithm to obtain a clustering center, then the improved FCM algorithm is used for carrying out clustering analysis on data samples, and finally the rolling bearing is subjected to fault diagnosis; the PSO algorithm and the FCM algorithm are fused to give full play to respective advantages, and the fault diagnosis accuracy of the rolling bearing is improved.
The technical scheme of the fault diagnosis method for the rolling bearing provided by the invention on the one hand is as follows:
a rolling bearing fault diagnosis method, comprising the steps of:
acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
And extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
The technical scheme of the fault diagnosis system for the rolling bearing provided by the invention is as follows:
A rolling bearing fault diagnostic system, the system comprising:
the data acquisition module is used for acquiring vibration data of the rolling bearing and dividing the vibration data into a test sample set and a training sample set;
the characteristic extraction module is used for extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on the vibration data of the rolling bearings in the training sample set;
the fault clustering module is used for carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by utilizing an FCM algorithm based on an improved PSO algorithm;
and the fault diagnosis module is used for extracting the fault characteristics of the rolling bearing with the concentrated test samples and judging the fault type of the rolling bearing with the concentrated test samples according to the fault clustering result.
another aspect of the present invention provides a computer-readable storage medium, wherein:
a computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
Acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
Carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
and extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
Another aspect of the present invention provides a processing apparatus, including:
A processing apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps when executing the program:
acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
Carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
And extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
Through the technical scheme, the invention has the beneficial effects that:
the method comprises the steps of firstly updating the speed and the position of particles by using an improved PSO algorithm to obtain a clustering center, then performing clustering analysis on data samples by using an improved FCM algorithm, and finally performing fault diagnosis on a rolling bearing; the PSO algorithm and the FCM algorithm are fused to give full play to respective advantages, and the fault diagnosis accuracy of the rolling bearing is improved.
drawings
the accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the application and not to limit the invention.
FIG. 1 is a flowchart of a rolling bearing fault diagnosis method according to an embodiment;
FIG. 2 is a flow chart of fault clustering based on the FCM algorithm of the improved PSO according to the first embodiment;
FIG. 3 is a diagram illustrating comparison of the calculation results of two algorithms according to the first embodiment.
Detailed Description
The invention is further described with reference to the following figures and examples.
it is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
it is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
the noun explains:
(1) and a fuzzy C-means clustering algorithm (FCM) is used for obtaining the membership degree of each sample point to all class centers by optimizing an objective function so as to determine the class of the sample points and achieve the purpose of automatically classifying the sample data.
(2) a Particle Swarm Optimization (PSO), also called a PSO or a bird swarm foraging algorithm, utilizes sharing of information by individuals in a swarm to enable movement of the whole swarm to generate an evolution process from disorder to order in a problem solving space on the basis of observation of animal swarm movement behaviors, and accordingly an optimal solution is obtained.
(3) a K-means clustering algorithm (K-means) gives a data point set and the required clustering number K, K is specified by a user, and the K-means algorithm repeatedly divides data into K clusters according to a certain distance function.
example one
Fig. 1 is a flowchart of a rolling bearing fault diagnosis method based on the FCM algorithm of the improved PSO algorithm according to the present embodiment. As shown in fig. 1, the rolling bearing fault diagnosis method includes the steps of:
s101, acquiring a vibration data set of the rolling bearing, and dividing the vibration data set into a test sample set and a training sample set.
Specifically, a vibration data set of the rolling bearing is acquired, which includes vibration data of the rolling bearing in states of normal, inner ring failure, outer ring failure, rolling body failure, and the like, and is some waveform signals.
And dividing one part of the vibration data in the obtained vibration data set of the rolling bearing into a training sample set, and taking the other part of the vibration data as a test sample set.
and S102, extracting the energy characteristics of the rolling bearings concentrated by the training samples.
Feature extraction is very important for fault classification and state monitoring of all vibration signals. The dynamic response caused by some faults of the rolling bearing becomes a non-smooth process, so that the vibration signal also becomes a non-smooth signal. The fault feature of the rolling bearing extracted in the present embodiment is an energy feature.
In the embodiment, a wavelet packet decomposition method is selected to extract the energy characteristics of the rolling bearing. The specific implementation mode is as follows:
the vibration data x (t) of the rolling bearing is subjected to i-layer wavelet packet decomposition, and the wavelet packet coefficient of a node (i, j) can be expressed aswhere N is the number of sample points of the original signal, the signal energy at the node of the wavelet packet (i, j)Expressed as:
Normalizing the wavelet packet energy to obtain a normalized energy characteristic, which is expressed as:
Carrying out i-layer wavelet packet decomposition on vibration data x (t) of the rolling bearing to obtain 2jnormalized energy characteristics, forming an energy characteristic set xi＝(xi 1,xi 2,…,xi 2j)。
S103, carrying out fault clustering on the energy characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on the improved PSO algorithm.
(1) The improved FCM algorithm:
The fuzzy C-means clustering algorithm determines the degree to which each data point belongs to a certain cluster by using the degree of membership, and the fuzzy C-means clustering algorithm uses fuzzy division so that the value of each given data point is between [0,1], and the sum of the degrees of membership of a data set is equal to 1. Expressed as:
its objective function is defined as follows:
In the formula: j is an objective function; n is the number of samples; c is the number of clusters; mu.sikexpressed as fuzzy degree of membershipThe formula is as follows:
Where m is a weighted index (i.e., fuzzy index m)>1)；is the ith data sample in the kth class; v. ofkIs the cluster center of the kth cluster, and the expression is:
the distance from each data to a clustering center is calculated by using a euclidean distance formula in the conventional fuzzy C-means clustering algorithm, but the euclidean distance does not consider the influence of overall variation on the distance, so that the distance from each data sample to the clustering center is represented by adopting a fuzzy paste schedule in the embodiment, the smaller the fuzzy paste schedule is, the closer the sample data is to a certain class is represented, and the fuzzy paste schedule expression is as follows:
wherein the content of the first and second substances,is the ith data sample in the kth class; v. ofkis the cluster center of the kth cluster; mu.sikIs the fuzzy degree of membership.
(2) Improving PSO algorithm:
A. Fitness function:
According to the actual demand of clustering, taking a target function of the improved fuzzy C-means clustering algorithm as a particle fitness function, wherein the expression is as follows;
Wherein the content of the first and second substances,is the ith data sample in the kth class; v. ofkis the cluster center of the kth cluster; mu.sikis fuzzy membership; n is the number of samples; c is the number of clusters; m is a weighted index.
the clustering effect is closely related to J, and as the target function is based on the fuzzy closeness, the smaller the fuzzy closeness, the closer the sample data is to a certain class, which indicates that the smaller the fitness value of the particle is, the better the corresponding clustering effect is.
B. inertial weight:
the coefficient of the particle flight velocity is represented by an inertial weight, maintaining a balance between the global and local search capabilities of the particle population. The traditional method for updating the inertia weight ensures that the variable quantity of the inertia weight is fixed in each iteration, thereby reducing the convergence speed of the algorithm. The embodiment adopts an adaptive adjustment method to update the inertia weight. The inertial weight of the ith particle in the t iteration is:
where t denotes the current number of iterations, wmax、wminRespectively representing a maximum inertial weight and a minimum inertial weight,a fitness value representing a global optimal solution,Representing fitness values of the individual best solutions.
The inertia weight is adjusted through the formula, on one hand, the inertia weight is reduced along with the gradual increase of the iteration times, and the particle swarm algorithm is favorably prevented from easily falling into a local extreme point when processing a high-dimensional complex function; on the other hand, the strategy can adaptively update the inertia weight along with the fitness values of the local global optimal solution and the individual optimal solution, and is favorable for accelerating the convergence speed of the particle swarm algorithm.
C. Learning factor:
in order to improve the convergence accuracy of the algorithm, the learning factor c is dynamically adjusted along with the gradual increase of the iteration times1,c2As follows:
Wherein, tmaxdenotes the maximum number of iterations, cmax，cminrespectively representing a maximum learning factor and a minimum learning factor; c. C1C decreases with increasing number of iterations, c2The value of (2) is increased along with the increase of the iteration times, so that the population is prevented from falling into a local extreme point, and the convergence speed and precision of the algorithm are effectively improved.
D. particle velocity and position update:
for the ith particle, its velocity component and position component in the t +1 th iteration are:
By introducing an inertia weight formula, a particle velocity update formula is improved, and the method comprises the following steps:
wherein v isijRepresents the velocity of the ith particle, and the rand interval [0,1]]the random number of (2) is greater than,Representing the individual optimal solution found by the ith particle,Representing a global optimal solution of the whole population; since the particle position is constantly changing with time, the particle position update formula is improved by introducing a time-of-flight factor, which is:
xij(t+1)＝xij(t)+s(t)×vij(t+1)(12)
Wherein x isijdenotes the position of the ith particle, s (t) denotes the flight factor, which is an important coefficient for adjusting the particle search step, s0Representing the time-of-flight constant. The flight factor is reduced along with the increase of the iteration number, and correspondingly, the particle searching step size is also reduced, so that the convergence speed of the algorithm is increased.
(3) and carrying out fault clustering on the energy characteristics of the rolling bearing based on the FCM algorithm of the improved PSO algorithm.
As shown in fig. 2, the specific implementation steps of fault clustering the energy characteristics of the rolling bearing based on the improved PSO algorithm and the FCM algorithm are as follows:
a. Initializing parameters;
b. Calculating the fitness value of each energy characteristic according to a formula (8), and updating the individual optimal solution and the global optimal solution of the energy characteristics;
c. Updating the speed and position of each energy feature according to equations (9) (10) (11) (12) and (13);
d. obtaining a clustering center according to the position of the energy characteristics, clustering and dividing each energy characteristic by adopting an improved FCM algorithm, and recalculating the clustering center according to a new dividing result;
Judging whether the clustering center changes and whether the maximum iteration times is reached, and if so, finishing clustering; if not, returning to the step b.
Specifically, in the step b, the specific implementation steps of clustering and partitioning each energy feature by using the improved FCM algorithm are as follows:
Calculating the fuzzy membership degree of each energy characteristic belonging to each cluster;
calculating the fuzzy pasting progress of each energy characteristic to each clustering center;
each energy feature is classified into the class that is closest to it.
And S104, carrying out fault diagnosis on the rolling bearing according to the fault clustering result.
specifically, the specific implementation process of fault diagnosis on the rolling bearing is as follows:
Carrying out fault clustering on the energy characteristics of the rolling bearings in the training set based on the FCM algorithm of the improved PSO algorithm to obtain a clustering center set of different bearing faults, wherein the clustering center set consists of the energy characteristics;
Extracting energy characteristics of vibration data of each rolling bearing in a test sample set;
and comparing the energy characteristics of each rolling bearing in the test sample set with the cluster center sets according to the cluster center sets, wherein the closer to which cluster center set the rolling bearing is, the fault type of the rolling bearing belongs to.
The following exemplifies a specific embodiment, and verifies the rolling bearing fault diagnosis method proposed in the above embodiment.
(1) Experimental preparation and feature extraction
The rolling bearing vibration data set of the university of kassingapond in the united states is used in the embodiment, the rotating speed is 1772r/min, the fault diameter is 0.1778mm, the rotating speed is 12kHz, four states including normal state, inner ring fault, outer ring fault in the 12:00 direction and rolling body fault are included, 50 samples are selected in each state, and the number of data points of each sample is 1024-point bearing vibration data. The energy characteristics of the rolling bearing are extracted by adopting a wavelet packet decomposition characteristic extraction method, and the partial normalized experimental data are shown in table 1. Selecting 30 fault states as training samples and 120 feature vectors for each fault state; for each fault state, 20 are selected as test samples, and the maximum number of iterations is 200, wherein the total number of the feature vectors is 80.
TABLE 1 partial bearing energy characteristic values
(2) analysis of Experimental results
and (4) selecting an objective function of the improved FCM algorithm as a fitness function value of the two models in an experiment. When the fitness function value is converged, the model is represented to find the best clustering center, and the smaller the fitness function value is, the better the clustering effect of the model is proved to be. As can be seen from fig. 3, when the FCM algorithm based on the improved PSO finds the best cluster center, the algorithm iterates to 38 th; when the FCM algorithm based on the traditional PSO finds the best cluster center, the algorithm iterates to 56 th. Thus, it can be concluded that the improved algorithm converges faster than the conventional algorithm.
TABLE 2 comparison of Performance of three algorithms
As can be seen from the table above, compared with the K-means algorithm based on the traditional PSO, the other two PSO-based FCM algorithms have obvious improvements in iteration times, running time and test accuracy, and the combination of the PSO and the FCM algorithms is helpful for solving the problem of fault ambiguity. Compared with the traditional PSO-based FCM algorithm, the improved PSO-based FCM algorithm has higher convergence speed and higher test accuracy, and solves the problems that the traditional PSO algorithm is easy to fall into local extrema, has low convergence speed, low search accuracy and the like when processing high-dimensional complex functions, and the traditional FCM algorithm does not consider the influence of overall variation on the distance.
Example two
The embodiment provides a rolling bearing fault diagnosis system, this system includes:
The data acquisition module is used for acquiring vibration data of the rolling bearing and dividing the vibration data into a test sample set and a training sample set;
The characteristic extraction module is used for extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on the vibration data of the rolling bearings in the training sample set;
The fault clustering module is used for carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by utilizing an FCM algorithm based on an improved PSO algorithm;
and the fault diagnosis module is used for extracting the fault characteristics of the rolling bearing with the concentrated test samples and judging the fault type of the rolling bearing with the concentrated test samples according to the fault clustering result.
EXAMPLE III
the present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
And extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
example four
The present embodiment provides a processing apparatus, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the following steps:
Acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
And extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A fault diagnosis method for a rolling bearing is characterized by comprising the following steps:
acquiring vibration data of a rolling bearing, and dividing the vibration data into a test sample set and a training sample set;
extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on vibration data of the rolling bearings in the training sample set;
Carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by using an FCM algorithm based on an improved PSO algorithm;
and extracting the fault characteristics of the rolling bearings in the test sample set, and judging the fault type of the rolling bearings in the test sample set according to the fault clustering result.
2. The rolling bearing fault diagnosis method according to claim 1, wherein the method of extracting the fault characteristics of the rolling bearing is:
carrying out multi-layer wavelet packet decomposition on the vibration data of the rolling bearing, and calculating the signal energy at the node of each layer of wavelet packet;
And normalizing the energy of all the wavelet packets to obtain normalized energy characteristics to form an energy characteristic set.
3. the rolling bearing fault diagnosis method according to claim 1, wherein the fitness function of the FCM algorithm based on the improved PSO algorithm is:
wherein the content of the first and second substances,is the ith data sample in the kth class; v. ofkIs the cluster center of the kth cluster; mu.sikIs fuzzy membership; n is the number of samples; c is the number of clusters; m is a weighted index.
4. The rolling bearing fault diagnosis method according to claim 1, wherein the inertia weight of the FCM algorithm based on the improved PSO algorithm is:
Where t denotes the current number of iterations, wmax、wminrespectively representing a maximum inertial weight and a minimum inertial weight,A fitness value representing a global optimal solution,A fitness value representing an individual optimal solution;
The learning factors are:
Wherein, tmaxDenotes the maximum number of iterations, cmax，cminrespectively representing a maximum learning factor and a minimum learning factor; c. C1、c2Is a learning factor;
the particle velocity update formula is:
wherein v isijRepresents the velocity of the ith particle, and the rand interval [0,1]]the random number of (2) is greater than,Representing the individual optimal solution found by the ith particle,Representing a global optimal solution of the whole population;
the particle position update formula is:
xij(t+1)＝xij(t)+s(t)×vij(t+1)
Wherein x isijDenotes the position of the ith particle, s (t) denotes the flight factor, which is an important coefficient for adjusting the particle search step, s0representing the time-of-flight constant.
5. the rolling bearing fault diagnosis method according to claim 1, wherein the step of fault clustering the fault characteristics of the rolling bearing using the FCM algorithm based on the improved PSO algorithm comprises:
initializing parameters of an FCM algorithm based on an improved PSO algorithm;
calculating the fitness value of each energy characteristic, and updating the individual optimal solution and the global optimal solution of the energy characteristics;
Updating the speed and position of each energy feature;
obtaining a clustering center according to the position of the energy characteristics, clustering and dividing each energy characteristic by adopting an improved FCM algorithm, and recalculating the clustering center according to a new dividing result;
and judging whether the clustering center changes and whether the maximum iteration times is reached, and if so, finishing clustering.
6. the rolling bearing fault diagnosis method according to claim 1, wherein the step of clustering each energy feature using the modified FCM algorithm comprises:
Respectively calculating the fuzzy membership degree of each energy characteristic belonging to each cluster and the fuzzy attaching degree of each energy characteristic to each cluster center;
Each energy feature is classified into the class that is closest to it.
7. the rolling bearing fault diagnosis method according to claim 1, wherein the method of determining the type of fault of the rolling bearing is:
carrying out fault clustering on the energy characteristics of the rolling bearings in the training set based on an improved PSO algorithm and an FCM algorithm to obtain a clustering center set of different bearing faults, wherein the clustering center set consists of the energy characteristics;
Extracting energy characteristics of vibration data of each rolling bearing in a test sample set;
And comparing the energy characteristics of each rolling bearing in the test sample set with the clustering center set according to the clustering center set, and judging the class of the clustering center with the closest energy characteristics of the rolling bearing as the fault type of the rolling bearing.
8. a rolling bearing fault diagnosis system is characterized by comprising:
the data acquisition module is used for acquiring vibration data of the rolling bearing and dividing the vibration data into a test sample set and a training sample set;
the characteristic extraction module is used for extracting fault characteristics of the rolling bearings in the training sample set by adopting a wavelet packet decomposition method based on the vibration data of the rolling bearings in the training sample set;
The fault clustering module is used for carrying out fault clustering on the fault characteristics of the rolling bearings in the training sample set by utilizing an FCM algorithm based on an improved PSO algorithm;
And the fault diagnosis module is used for extracting the fault characteristics of the rolling bearing with the concentrated test samples and judging the fault type of the rolling bearing with the concentrated test samples according to the fault clustering result.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method for diagnosing a failure of a rolling bearing according to any one of claims 1 to 7.
10. a processing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps in the rolling bearing fault diagnosis method according to any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911033248.7A CN110567721B (en) | 2019-10-28 | 2019-10-28 | Rolling bearing fault diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911033248.7A CN110567721B (en) | 2019-10-28 | 2019-10-28 | Rolling bearing fault diagnosis method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110567721A true CN110567721A (en) | 2019-12-13 |
| CN110567721B CN110567721B (en) | 2021-08-17 |
Family
ID=68786050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911033248.7A Active CN110567721B (en) | 2019-10-28 | 2019-10-28 | Rolling bearing fault diagnosis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110567721B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111811818A (en) * | 2020-06-02 | 2020-10-23 | 桂林电子科技大学 | Rolling bearing fault diagnosis method based on AP clustering algorithm of specified clustering number |
| CN113609901A (en) * | 2021-06-25 | 2021-11-05 | 国网山东省电力公司泗水县供电公司 | Power transmission and transformation equipment fault monitoring method and system |
| CN114964777A (en) * | 2022-05-11 | 2022-08-30 | 盐城工学院 | Rolling bearing fault detection method |
| CN115700363A (en) * | 2022-11-07 | 2023-02-07 | 南京工业大学 | Fault diagnosis method and system for rolling bearing of coal mining machine, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944418A (en) * | 2012-12-11 | 2013-02-27 | 东南大学 | Wind turbine generator group blade fault diagnosis method |
| KR101254177B1 (en) * | 2011-10-07 | 2013-04-19 | 위아코퍼레이션 주식회사 | A system for real-time recognizing a face using radial basis function neural network algorithms |
| CN103886146A (en) * | 2014-03-12 | 2014-06-25 | 河海大学 | Control parameter optimization method for improving phase commutating failure restraining capacity of direct current system |
| CN204346711U (en) * | 2015-01-21 | 2015-05-20 | 山西潞安环保能源开发股份有限公司 | A kind of portable bearing fault diagnosing apparatus based on vibration detection |
| CN109470477A (en) * | 2018-09-27 | 2019-03-15 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on improved PSO algorithm optimization FSVM |
| CN109934810A (en) * | 2019-03-08 | 2019-06-25 | 太原理工大学 | A kind of defect classification method based on improvement population wavelet neural network |
| CN109947047A (en) * | 2019-03-28 | 2019-06-28 | 西安科技大学 | A kind of electro spindle imbalance fault diagnosis method |
-
2019
- 2019-10-28 CN CN201911033248.7A patent/CN110567721B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101254177B1 (en) * | 2011-10-07 | 2013-04-19 | 위아코퍼레이션 주식회사 | A system for real-time recognizing a face using radial basis function neural network algorithms |
| CN102944418A (en) * | 2012-12-11 | 2013-02-27 | 东南大学 | Wind turbine generator group blade fault diagnosis method |
| CN103886146A (en) * | 2014-03-12 | 2014-06-25 | 河海大学 | Control parameter optimization method for improving phase commutating failure restraining capacity of direct current system |
| CN204346711U (en) * | 2015-01-21 | 2015-05-20 | 山西潞安环保能源开发股份有限公司 | A kind of portable bearing fault diagnosing apparatus based on vibration detection |
| CN109470477A (en) * | 2018-09-27 | 2019-03-15 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on improved PSO algorithm optimization FSVM |
| CN109934810A (en) * | 2019-03-08 | 2019-06-25 | 太原理工大学 | A kind of defect classification method based on improvement population wavelet neural network |
| CN109947047A (en) * | 2019-03-28 | 2019-06-28 | 西安科技大学 | A kind of electro spindle imbalance fault diagnosis method |
Non-Patent Citations (8)
| Title |
|---|
| XIAO-JIN WAN等: "Fault diagnosis of rolling bearing based on optimized soft competitive learning Fuzzy ART and similarity evaluation technique", 《ADVANCED ENGINEERING INFORMATICS》 * |
| 周飞红等: "自适应惯性权重的分组并行粒子群优化算法", 《计算机工程与应用》 * |
| 孙希: "基于小波包和FCM多分类器组的轴承故障诊断", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 张学林等: "带飞行时间因子的改进粒子群优化算法", 《实验技术与管理》 * |
| 王凯丽等: "基于IPSO算法的光伏阵列多峰值MPPT研究", 《电器工程学报》 * |
| 蔡新等: "《工程结构优化设计》", 31 October 2003, 中国水利水电出版社 * |
| 赵琦等: "基于模糊聚类及神经网络的连铸漏钢预报", 《中国冶金》 * |
| 陈燕等: "《数据挖掘与聚类分析》", 30 November 2012, 连海事大学出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111811818A (en) * | 2020-06-02 | 2020-10-23 | 桂林电子科技大学 | Rolling bearing fault diagnosis method based on AP clustering algorithm of specified clustering number |
| CN111811818B (en) * | 2020-06-02 | 2022-02-01 | 桂林电子科技大学 | Rolling bearing fault diagnosis method based on AP clustering algorithm of specified clustering number |
| CN113609901A (en) * | 2021-06-25 | 2021-11-05 | 国网山东省电力公司泗水县供电公司 | Power transmission and transformation equipment fault monitoring method and system |
| CN114964777A (en) * | 2022-05-11 | 2022-08-30 | 盐城工学院 | Rolling bearing fault detection method |
| CN115700363A (en) * | 2022-11-07 | 2023-02-07 | 南京工业大学 | Fault diagnosis method and system for rolling bearing of coal mining machine, electronic equipment and storage medium |
| CN115700363B (en) * | 2022-11-07 | 2023-08-08 | 南京工业大学 | Method and system for diagnosing faults of rolling bearing of coal mining machine, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110567721B (en) | 2021-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110567721B (en) | Rolling bearing fault diagnosis method and system | |
| CN111161879B (en) | Disease prediction system based on big data | |
| CN109460793A (en) | A kind of method of node-classification, the method and device of model training | |
| Zhou et al. | Abrupt motion tracking via adaptive stochastic approximation Monte Carlo sampling | |
| CN110333076B (en) | Bearing fault diagnosis method based on CNN-Stacking | |
| CN115096590A (en) | Rolling bearing fault diagnosis method based on IWOA-ELM | |
| CN110987436B (en) | Bearing fault diagnosis method based on excitation mechanism | |
| CN112465160A (en) | VR-based vehicle maintenance auxiliary system | |
| CN105913078A (en) | Multi-mode soft measurement method for improving adaptive affine propagation clustering | |
| CN113255873A (en) | Clustering longicorn herd optimization method, system, computer equipment and storage medium | |
| CN113722980A (en) | Ocean wave height prediction method, system, computer equipment, storage medium and terminal | |
| CN106326188B (en) | Task dividing system and its method based on backward learning radius particle group optimizing | |
| CN110929761A (en) | Balance method for collecting samples in situation awareness framework of intelligent system security system | |
| CN113989655A (en) | Radar or sonar image target detection and classification method based on automatic deep learning | |
| CN113190931A (en) | Sub-health state identification method for improving optimized DBN-ELM of wolf | |
| CN113627075A (en) | Projectile aerodynamic coefficient identification method based on adaptive particle swarm optimization extreme learning | |
| CN115442887B (en) | Indoor positioning method based on RSSI of cellular network | |
| CN113807005A (en) | Bearing residual life prediction method based on improved FPA-DBN | |
| CN115778372A (en) | Knee joint angle estimation method based on surface electromyogram signals | |
| CN112446435B (en) | City data classification method and system | |
| CN111476321B (en) | Air flyer identification method based on feature weighting Bayes optimization algorithm | |
| CN113989543A (en) | COVID-19 medical image detection and classification method and device | |
| CN112183884A (en) | Grain storage quality prediction method and device | |
| CN113341379A (en) | Radar signal sorting method based on adaptive threshold and iterative control | |
| Ren et al. | Research on multimodal algorithms for multi-routes planning based on niche technique |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |