CN109947047B - Electric spindle unbalance fault diagnosis method - Google Patents
Electric spindle unbalance fault diagnosis method Download PDFInfo
- Publication number
- CN109947047B CN109947047B CN201910245484.9A CN201910245484A CN109947047B CN 109947047 B CN109947047 B CN 109947047B CN 201910245484 A CN201910245484 A CN 201910245484A CN 109947047 B CN109947047 B CN 109947047B
- Authority
- CN
- China
- Prior art keywords
- electric spindle
- fault
- unbalance
- data
- matrix
- 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.)
- Active
Links
Images
Landscapes
- Image Processing (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses an unbalance fault diagnosis method for an electric spindle, which comprises the steps of collecting vibration data of the electric spindle in different unbalance states, carrying out noise reduction pretreatment on the vibration data, converting one-dimensional vibration signals of the electric spindle into two-dimensional snowflake images according to a symmetric polar coordinate method, and simply and intuitively judging the severity of the unbalance state of the electric spindle according to the change rule of the two-dimensional snowflake images; and extracting image characteristic parameters of a snowflake image generated by a sample with a known fault by using a gray level co-occurrence matrix to obtain a characteristic parameter matrix, using the characteristic parameter matrix as the input of FCM clustering to obtain a clustering center, extracting the image characteristic parameters of the sample data to be detected, calculating the degree of adherence between the sample data to be detected and the clustering center by using the Haiming adherence progress, and further obtaining the fault category of the sample to be detected. The method realizes the function of unbalance fault diagnosis of the electric spindle, reduces the difficulty of fault diagnosis and improves the intelligent degree and accuracy of diagnosis.
Description
Technical Field
The invention relates to the field of electric spindle fault diagnosis, in particular to an electric spindle unbalance fault diagnosis method, and particularly relates to a diagnosis method for realizing electric spindle unbalance faults by combining a symmetric polar coordinate image method and a fuzzy C-means clustering algorithm.
Background
The numerical control machine tool is an industrial master machine in the equipment manufacturing industry, and the design, the manufacturing capability and the level of key technology are one of main indexes for measuring the national industrialization level. High speed, precision and intellectualization become the development trend of the high-end numerical control machine tool at present, and the electric spindle is one of the core components and key technologies for realizing high speed and high precision technology of the high-grade numerical control machine tool. As a result of the perfect combination of a conventional mechanical spindle with an electric motor, the electric spindle is a core component of high-end machine tools. However, in the process of manufacturing and assembling the spindle, the mass center of the spindle is inevitably inconsistent with the rotation center of the spindle due to uneven material, processing and assembling errors and the like, so that mass imbalance is generated. The electric spindle has high rotating speed, and even small unbalance can cause great centrifugal force to excite the system to vibrate violently, so that the timely diagnosis of the unbalance vibration of the electric spindle has great significance for the development of high-speed and high-precision machine tools.
At present, the conventional fault diagnosis method for the rotary machine mainly depends on vibration signals, and takes signal time/frequency domain amplitude, frequency or energy as fault characteristic parameters, so as to obtain a fault diagnosis result. However, in the fault identification of the method, parameters under various conditions need to be manually compared and analyzed, the judgment of the final result has high dependence on the level of people, and in the use field of mechanical equipment, most of the results are workers in the same line, so that the fault identification is difficult to be quickly and accurately carried out; with the intelligent manufacturing, the efficiency of the conventional fault diagnosis method is too low, and a relatively accurate conclusion is difficult to be quickly obtained.
Disclosure of Invention
The invention aims to provide an unbalance fault diagnosis method for an electric spindle, aiming at the defects of the existing fault diagnosis method, the invention combines a symmetric polar coordinate image method and a fuzzy C-means (FCM) clustering algorithm to image a one-dimensional vibration signal in a two-dimensional mode, is simple and visual, can more intelligently identify the fault degree, and improves the accuracy of fault diagnosis.
In order to achieve the purpose, the invention adopts the following technical scheme:
an electric spindle unbalance fault diagnosis method comprises the following steps:
1) vibration data of the electric spindle in different running states are measured through a vibration acceleration sensor arranged on the electric spindle, the vibration data are collected through a multi-channel collection card, and then the data are transmitted to an upper computer;
2) exporting the vibration data of the motorized spindle obtained in the step 1), and carrying out noise reduction pretreatment on the vibration data;
3) setting the amplitude value at the moment i as x in the discrete sampling data sequence of the vibration time domain signal obtained after the noise reduction processing in the step 2)iThe amplitude at time i + L is xi+LThe method is substituted into a symmetric polar coordinate calculation formula to carry out the following operation, can be converted into points in a polar coordinate space P (gamma (i), theta (i) and phi (i)), and finally can be converted into a hexagonal snowflake mirror symmetric image under the polar coordinate, and the severity of the unbalance fault can be directly judged through the generated snowflake image.
Wherein x ismaxFor maximum value of sampled data, xminIs the minimum value of the sampled data; l is a time interval, and the value of L is taken as 3; g is an angle amplification factor, and the value of g is 30 degrees; theta is an initial line rotation angle, namely a mirror symmetry plane rotation angle, and is taken as 60 degrees;
4) extracting image features of the snowflake image generated in the step 3) by adopting a gray level co-occurrence matrix method, selecting 6 feature statistics in 4 directions to perform image texture recognition, and taking the mean value of the 4 direction co-occurrence matrices as a final gray level co-occurrence matrix, namely a feature vector matrix;
5) standardizing a feature vector matrix generated by the snowflake image generated by the known fault sample according to the formula (4) to obtain an initial membership matrix, and using the initial membership matrix as FCM clustering input to obtain a membership matrix U and a clustering center C of the known fault sample, wherein the obtained clustering center C can be used as a standard mode for judging unbalance faults of the electric spindle;
wherein x isijFor any ith row and jth column data, x, of the eigenvector matrixjJ column data of the feature vector matrix X; the denominator is the difference between the maximum value and the minimum value in each element of the jth column in the original data matrix. Data compression at [0,1 ] by normalization]Within a closed interval.
6) And extracting image characteristic parameters from the to-be-detected fault sample data through a gray level co-occurrence matrix, standardizing by using the formula (4), and calculating according to the Haiming post progress to judge the to-be-detected fault category.
Further, different operating states of the electric spindle refer to different unbalance states.
Further, different operating states of the electric spindle can be realized by screwing screws with different masses into the round holes on the counterweight plate.
Further, the noise reduction preprocessing specifically includes: and (3) after Empirical Mode Decomposition (EMD) is carried out on the collected vibration data of the electric spindle, the component of the contained main fault characteristic information is reserved, and other components are discarded.
Further, the method for intuitively judging the severity of the unbalance fault by directly generating the snowflake image specifically comprises the following steps: as the amount of imbalance increases, the petals of the image become progressively thinner, and the gap between the first pair of petals formed when the initial line is 0 ° becomes progressively larger.
Further, θ is an initial line rotation angle, that is, a rotation angle of a mirror symmetry plane, where θ is 60 °, and the mirror symmetry planes are 0 °, 60 °, 120 °, 180 °, 240 °, and 300 °, respectively, and the six mirror planes are overlapped to form the snowflake-shaped image.
Further, the 4 directions include four generation directions of the gray level co-occurrence matrix, i.e., 0 °, 45 °, 90 °, and 135 °; the 6 feature statistics include contrast, correlation, energy, inverse distance, maximum probability, and entropy.
Further, the cluster center C can be used as a standard mode for judging the unbalance of the electric spindle, and each row of the cluster center C is a cluster center of sample data in different unbalance states.
Further, the Haiming posting progress is the posting degree of each row of the standard mode C obtained in the step 4) and the fault sample to be detected is calculated according to the following formula (5), and the sample to be detected is judged as the fault type with the maximum posting degree value.
Where A and B are two fuzzy subsets, X ═ X1,x2,…,xnIs a finite set, n is the number of subsets in the finite set X, and k is the finite set XThe kth column subset of (a), N (a, B) represents the degree of tiling of the two fuzzy sets.
Compared with the prior art, the invention has the following beneficial technical effects:
the method can convert one-dimensional vibration signals of the electric spindle in different unbalance states into two-dimensional image signals, judge the severity of the unbalance state of the electric spindle according to the change of the image, specifically convert time domain signals of the electric spindle vibration into snowflake images by adopting a symmetrical polar coordinate image method, judge the severity of a fault through the change of the petal size and the petal gap of the snowflake images, can judge the fault type more simply and intuitively, and reduce the diagnosis complexity; according to the method, the characteristic parameter matrix after the known fault sample is standardized can be used as the input of the FCM clustering to obtain the clustering center of the known sample, the fault type of the sample to be tested is judged by calculating the hamming degree of the fault sample to be tested and the obtained clustering center of the known fault sample, the severity of the unbalance fault of the electric spindle can be judged more intelligently and efficiently, and the intelligent degree and accuracy of fault diagnosis are improved.
Drawings
FIG. 1 is a block diagram of a process for implementing unbalance fault diagnosis of an electric spindle according to the present invention;
fig. 2 is a two-dimensional snowflake pattern of the electric spindle in three different unbalance states, wherein (a) indicates a normal state, (b) indicates an unbalance amount of 29.37g · cm, and (c) indicates an unbalance amount of 42.24g · cm.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
referring to fig. 1, a method for diagnosing unbalance faults of an electric spindle includes the steps of after time domain vibration signals of the electric spindle in different unbalance states are measured, converting the time domain vibration signals into hexagonal snowflake mirror symmetry images through noise reduction processing by using a symmetric polar coordinate image method, and judging the severity of the unbalance faults of the electric spindle through changes of the snowflake images; and extracting a characteristic parameter matrix of the snowflake image through the gray level co-occurrence matrix and using the characteristic parameter matrix as the input of FCM clustering to obtain a clustering center, and judging the fault category of the fault sample to be detected according to the calculation of the Haiming post progress.
The diagnosis method comprises the following specific steps:
1) vibration data of the electric spindle in different unbalanced states are measured through a vibration acceleration sensor arranged on the electric spindle, the vibration data are collected through a multi-channel collection card, and then the data are transmitted to an upper computer;
2) deriving the vibration data of the electric spindle obtained in the step 1), and carrying out noise reduction pretreatment on the vibration data. The noise reduction pretreatment specifically comprises the following steps: and (3) after Empirical Mode Decomposition (EMD) is carried out on the collected vibration data of the electric spindle, the component of the contained main fault characteristic information is reserved, and other components are discarded.
3) Setting the amplitude value at the moment i as x in the discrete sampling data sequence of the vibration time domain signal obtained after the noise reduction processing in the step 2)iThe amplitude at time i + L is xi+LThe method is substituted into a symmetric polar coordinate calculation formula to carry out the following operation, can be converted into points in a polar coordinate space P (gamma (i), theta (i) and phi (i)), and finally can be converted into a hexagonal snowflake mirror symmetric image under the polar coordinate, and the severity of the unbalance fault can be directly judged through the generated snowflake image. The image change rule is as follows: as the amount of imbalance increases, the petals of the image become progressively thinner, and the gap between the first pair of petals formed when the initial line is 0 ° becomes progressively larger.
Wherein x ismaxFor maximum value of sampled data, xminIs the minimum value of the sampled data; l is a time interval, and the value of L is taken as 3; g is an angle amplification factor, and the value of g is 30 degrees; theta is the initial line rotation angle, i.e. the mirror symmetry plane rotation angle, taking theta equal to 60 deg., which isThe mirror symmetry planes are respectively 0 degrees, 60 degrees, 120 degrees, 180 degrees, 240 degrees and 300 degrees, and the six mirror symmetry planes are overlapped to form the snowflake-shaped image.
4) Extracting image characteristics of the snowflake image generated in the step 3) by adopting a gray level co-occurrence matrix method, selecting 6 characteristic statistics in 4 directions to perform image texture recognition, and taking the mean value of the 4 direction co-occurrence matrices as a final gray level co-occurrence matrix, namely a characteristic vector matrix. The 4 directions include four generation directions of the gray level co-occurrence matrix, i.e., 0 °, 45 °, 90 °, and 135 °; the 6 feature statistics include contrast, correlation, energy, inverse distance, maximum probability, and entropy.
5) Standardizing the eigenvector matrix generated by the snowflake image generated by the known fault sample in the step 4) according to the following formula (4) to obtain an initial membership matrix, and taking the initial membership matrix as FCM clustering input to obtain a membership matrix U of the known fault sample and a clustering center C, wherein the obtained clustering center C can be used as a standard mode for judging the unbalance fault of the electric spindle, and each row of the clustering center C is respectively the clustering center of sample data in different unbalance states.
Wherein x isijFor any ith row and jth column data, x, of the eigenvector matrixjJ column data of the feature vector matrix X; the denominator is the difference between the maximum value and the minimum value in each element of the jth column in the original data matrix. Data compression at [0,1 ] by normalization]Within a closed interval.
6) Extracting image characteristic parameters from the to-be-detected fault sample data through a gray level co-occurrence matrix, standardizing by using the formula (4), and then calculating the closeness of the to-be-detected fault sample and each row of the standard mode C obtained in the step 4) according to the Haiming closeness degree and the following formula (5), wherein the to-be-detected sample is judged to be the fault type with the largest closeness value.
Wherein, A and B are two fuzzy subsets, and X ═ X1,x2,…,xnAnd N is the number of subsets in the finite set X, k is the kth column subset in the finite set X, and N (A, B) represents the pasting progress of the two fuzzy sets.
The present invention is described in further detail below with reference to specific examples:
the vibration data of the electric spindle shows certain regularity along with the increase of the unbalance amount. According to the method, a one-dimensional vibration signal is converted into a two-dimensional snowflake image signal, and the degree of a fault is obtained by comparing the change rule of a snowflake image; extracting image characteristic parameters of known fault samples by utilizing the gray level co-occurrence matrix, taking the characteristic parameter matrix as the input of FCM clustering, obtaining a clustering center, extracting the image characteristic parameters of the fault samples to be detected, and then judging the fault types according to the calculation of the Haiming post progress. The complexity of the unbalance diagnosis of the electric spindle is reduced, and the intelligent degree and accuracy of the unbalance diagnosis of the electric spindle are improved.
The method comprises the following steps:
a grinding spindle test platform based on 170MD12Y16 is designed, when the output rotating speed of a motor is 1800r/min, screws with the mass of 3.88g and that of 5.58g are respectively screwed into holes in the circumference of a counterweight disc with the radius of 7.57cm, and acceleration sensors are utilized to test vibration signals of an electric spindle under normal and two different unbalance quantities, and the specific results are as follows:
1. and (6) data acquisition. Measuring vibration data of a plurality of groups of electric spindles in three different unbalance states of normal state, unbalance amount of 29.37 g-cm and unbalance amount of 42.24 g-cm by using a vibration acceleration sensor arranged on the electric spindles, collecting the vibration data by using a multi-channel acquisition card, and transmitting the data to an upper computer;
2. and (5) noise reduction pretreatment. And (3) deriving the obtained vibration data of the electric spindle, performing Empirical Mode Decomposition (EMD), and taking the IMF component corresponding to the maximum first 5 phase relation numbers as a characteristic component, wherein the rest corresponding IMF components are taken as interference noise.
3. And (4) generating a two-dimensional snowflake image. Discrete sampling data sequence of vibration time domain signal obtained after noise reduction processingIn, let the amplitude at time i be xiThe amplitude at time i + L is xi+LThe method is substituted into a symmetric polar coordinate calculation formula to carry out the following operation, can be converted into points in a polar coordinate space P (gamma (i), theta (i) and phi (i)), and finally can be converted into a hexagonal snowflake mirror symmetric image under the polar coordinate, and the severity of the unbalance fault can be directly judged through the generated snowflake image. The snowflake images generated in three different unbalance states of normal, unbalance amount of 29.37g cm and unbalance amount of 42.24g cm are shown in figure 2, and the image change rule is as follows: as the amount of imbalance increases, the petals of the image become progressively thinner, and the gap between the first pair of petals formed when the initial line is 0 ° becomes progressively larger.
Wherein x ismaxFor maximum value of sampled data, xminIs the minimum value of the sampled data; l is a time interval, and the value of L is taken as 3; g is an angle amplification factor, and the value of g is 30 degrees; and theta is an initial line rotation angle, namely a rotation angle of a mirror symmetry plane, wherein theta is 60 degrees, the mirror symmetry planes are respectively 0 degrees, 60 degrees, 120 degrees, 180 degrees, 240 degrees and 300 degrees, and the six mirror planes are overlapped to form the snowflake-shaped image.
4. And (5) extracting image features. And (3) extracting image characteristics of the generated snowflake image by adopting a gray level co-occurrence matrix method, selecting 6 characteristic statistics in 4 directions to perform image texture recognition, taking the mean value of the co-occurrence matrices in 4 directions as a final gray level co-occurrence matrix, namely a characteristic vector matrix, and obtaining 9 groups of image characteristic parameters of known fault samples as shown in the table 1.
TABLE 19 set of image characteristic parameters for known failure samples
And 5, calculating the FCM clustering center. Normalizing the feature vector matrix generated in the step 4 of the snowflake image generated by the known fault sample according to the following formula (4) to obtain an initial membership matrix, and taking the initial membership matrix as FCM clustering input to obtain a membership matrix U of the known fault sample, wherein the clustering center C is as follows:
the obtained cluster center C can be used as a standard mode for judging the unbalance fault of the electric spindle, and the 1 st, 2 nd and 3 rd rows of the cluster center C are respectively the cluster centers of sample data of normal, unbalance amount of 29.37g cm and unbalance amount of 42.24g cm.
Wherein x isijFor any ith row and jth column data, x, of the eigenvector matrixjJ column data of the characteristic matrix X; the denominator is the difference between the maximum value and the minimum value in each element of the jth column in the original data matrix. Data compression at [0,1 ] by normalization]Within a closed interval.
6. And (4) fault identification. Extracting image characteristic parameters from the to-be-detected fault sample data through a gray level co-occurrence matrix, standardizing by using the formula (4), and then calculating the closeness of the to-be-detected fault sample and each row of the standard mode C obtained in the step 4) according to the Haiming closeness degree and the following formula (5), wherein the to-be-detected sample is judged to be the fault type with the largest closeness value. The identification results of the 9 sets of fault samples to be tested are shown in table 2.
Where A and B are two fuzzy subsets, X ═ X1,x2,…,xnAnd N is the number of subsets in the finite set X, k is the kth column subset in the finite set X, and N (A, B) represents the pasting progress of the two fuzzy sets.
Table 29 identification results of to-be-detected fault samples
And (3) test results: based on actual equipment, after multiple tests, the designed electric spindle has the rotating speed of 1800r/min, 9 groups of sample data to be tested are measured in three different unbalance states of normal, unbalance 29.37 g-cm and unbalance 42.24 g-cm, the results are shown in a table 2, and the diagnosis result can be seen to be matched with the preset equipment fault condition, so that the diagnosis method can accurately diagnose the corresponding equipment fault.
It should be noted that modifications can be made by other researchers in the field without departing from the principles of the present invention. The main technical solutions described in this embodiment are implemented based on MATLAB software, and all components that are not specified can be implemented by using existing technologies and other programming software. Such modifications are also to be considered as within the scope of the present invention.
Claims (9)
1. The method for diagnosing the unbalance fault of the electric spindle is characterized by comprising the following steps of:
1) vibration data of the electric spindle in different running states are measured through a vibration acceleration sensor arranged on the electric spindle, and the vibration data are collected through a multi-channel collection card and then transmitted to an upper computer;
2) exporting the vibration data of the motorized spindle obtained in the step 1), and carrying out noise reduction pretreatment on the vibration data;
3) setting the amplitude value at the moment i as x in the discrete sampling data sequence of the vibration time domain signal obtained after the noise reduction processing in the step 2)iThe amplitude at time i + L is xi+LSubstitution symmetryThe polar coordinate calculation formula is converted into a point in a polar coordinate space P (gamma (i), theta (i) and phi (i)), and finally converted into a hexagonal snowflake mirror symmetry image under a polar coordinate, and the severity of the unbalance fault can be directly and intuitively judged through the generated snowflake image;
wherein x ismaxFor maximum value of sampled data, xminIs the minimum value of the sampled data; l is a time interval; g is an angle amplification factor; theta is an initial line rotation angle, namely a mirror symmetry plane rotation angle;
4) extracting image features of the snowflake image generated in the step 3) by adopting a gray level co-occurrence matrix method, selecting 6 feature statistics in 4 directions to perform image texture recognition, and taking the mean value of the 4 direction co-occurrence matrices as a final gray level co-occurrence matrix, namely a feature vector matrix;
5) standardizing a feature vector matrix generated by the snowflake image generated by the known fault sample according to the formula (4) to obtain an initial membership matrix, using the initial membership matrix as FCM clustering input to obtain a membership matrix U and a clustering center C of the known fault sample, and using the obtained clustering center C as a standard mode for judging unbalance faults of the electric spindle;
wherein x isijFor any ith row and jth column data of the eigenvector matrix,xjj column data of the feature vector matrix X; the denominator is the difference between the maximum value and the minimum value in each element of the jth row in the original data matrix;
6) and extracting image characteristic parameters from the to-be-detected fault sample data through a gray level co-occurrence matrix, standardizing by using the formula (4), and calculating according to the Haiming post progress to judge the to-be-detected fault category.
2. The method as claimed in claim 1, wherein the different operating states of the electric spindle refer to different unbalance states.
3. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein different operation states of the electric spindle are realized by screwing screws with different masses into round holes on the counterweight plate.
4. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein the noise reduction preprocessing specifically comprises the following steps: and after the acquired vibration data of the electric spindle is subjected to empirical mode decomposition, components containing main fault characteristic information are reserved, and other components are discarded.
5. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein the severity of the unbalance fault is directly determined through the generated snowflake image in the step 3), and specifically: as the amount of imbalance increases, the petals of the image become progressively thinner, and the gap between the first pair of petals formed when the initial line is 0 ° becomes progressively larger.
6. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein θ is 60 °, mirror symmetry planes of the six mirror planes are respectively 0 °, 60 °, 120 °, 180 °, 240 ° and 300 °, and the six mirror planes are overlapped to form a snowflake-shaped image.
7. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein the 4 directions in the step 4) comprise four generation directions of the gray level co-occurrence matrix, namely 0 °, 45 °, 90 ° and 135 °; the 6 feature statistics include contrast, correlation, energy, inverse distance, maximum probability, and entropy.
8. The method as claimed in claim 1, wherein a cluster center C is used as a standard mode for determining the unbalance of the electric spindle, and each row of the cluster center C is a cluster center of sample data in different unbalance states.
9. The method for diagnosing the unbalance fault of the electric spindle according to claim 1, wherein the hamming progress is obtained by calculating the closeness of the fault sample to be detected and each row of the standard mode C obtained in the step 4) according to the following formula (5), and the sample to be detected is determined as a fault category with the maximum closeness value;
where A and B are two fuzzy subsets, X ═ X1,x2,…,xnAnd N is the number of subsets in the finite set X, k is the kth column subset in the finite set X, and N (A, B) represents the pasting progress of the two fuzzy sets.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910245484.9A CN109947047B (en) | 2019-03-28 | 2019-03-28 | Electric spindle unbalance fault diagnosis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910245484.9A CN109947047B (en) | 2019-03-28 | 2019-03-28 | Electric spindle unbalance fault diagnosis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109947047A CN109947047A (en) | 2019-06-28 |
| CN109947047B true CN109947047B (en) | 2021-03-23 |
Family
ID=67012430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910245484.9A Active CN109947047B (en) | 2019-03-28 | 2019-03-28 | Electric spindle unbalance fault diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947047B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110728300A (en) * | 2019-09-09 | 2020-01-24 | 交控科技股份有限公司 | Method and system for identifying fault type based on turnout action current curve |
| CN110595780B (en) * | 2019-09-20 | 2021-12-14 | 西安科技大学 | Bearing fault identification method based on vibration gray level image and convolution neural network |
| CN110567721B (en) * | 2019-10-28 | 2021-08-17 | 齐鲁工业大学 | Rolling bearing fault diagnosis method and system |
| CN110880178B (en) * | 2019-11-27 | 2023-05-09 | 济南大学 | Rolling bearing fault diagnosis method based on improved Manhattan distance |
| CN116991115B (en) * | 2023-09-27 | 2023-12-29 | 中科航迈数控软件(深圳)有限公司 | Method, device, equipment and medium for monitoring state of main shaft of numerical control machine tool |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003315213A (en) * | 2002-04-25 | 2003-11-06 | Toshiba Corp | Apparatus and method for diagnosing vibration |
| CN101545824A (en) * | 2008-03-25 | 2009-09-30 | 唐德尧 | Fault diagnosis technology for mechanical tower |
| CN108776457A (en) * | 2018-07-20 | 2018-11-09 | 安徽江机重型数控机床股份有限公司 | A kind of numerically-controlled machine tool cutter head fault diagnosis system |
| CN109253882A (en) * | 2018-10-08 | 2019-01-22 | 桂林理工大学 | A kind of rotor crack fault diagnostic method based on variation mode decomposition and gray level co-occurrence matrixes |
| KR20190092076A (en) * | 2018-01-30 | 2019-08-07 | 현대자동차주식회사 | Appratus and method for detecting steering roll connector assembly status using steering angle sensor |
-
2019
- 2019-03-28 CN CN201910245484.9A patent/CN109947047B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003315213A (en) * | 2002-04-25 | 2003-11-06 | Toshiba Corp | Apparatus and method for diagnosing vibration |
| CN101545824A (en) * | 2008-03-25 | 2009-09-30 | 唐德尧 | Fault diagnosis technology for mechanical tower |
| KR20190092076A (en) * | 2018-01-30 | 2019-08-07 | 현대자동차주식회사 | Appratus and method for detecting steering roll connector assembly status using steering angle sensor |
| CN108776457A (en) * | 2018-07-20 | 2018-11-09 | 安徽江机重型数控机床股份有限公司 | A kind of numerically-controlled machine tool cutter head fault diagnosis system |
| CN109253882A (en) * | 2018-10-08 | 2019-01-22 | 桂林理工大学 | A kind of rotor crack fault diagnostic method based on variation mode decomposition and gray level co-occurrence matrixes |
Non-Patent Citations (3)
| Title |
|---|
| Fault detection of motorized spindle based on clustering by density peaks;Qiang Li 等;《2018 12th International Conference on Reliability, Maintainability, and Safety》;20181231;全文 * |
| 一种对称极坐标图像模糊Ｃ均值聚类的电主轴失衡故障诊断方法;樊红卫 等;《西安交通大学学报》;20191231;第53卷(第12期);全文 * |
| 基于振动谱图像识别的智能故障诊断;林勇 等;《振动、测试与诊断》;20100430;第30卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109947047A (en) | 2019-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109947047B (en) | Electric spindle unbalance fault diagnosis method | |
| CN110110768B (en) | Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers | |
| CN111103139A (en) | Rolling bearing fault diagnosis method based on GRCMSE and manifold learning | |
| CN111761409A (en) | Multi-sensor numerical control machine tool cutter wear monitoring method based on deep learning | |
| CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| CN109000921B (en) | Method for diagnosing main shaft fault of wind turbine generator | |
| CN109375010A (en) | Winding in Power Transformer method for diagnosing faults based on vibration distribution characteristic model | |
| CN106198020A (en) | Wind turbines bearing failure diagnosis method based on subspace and fuzzy C-means clustering | |
| CN112098088B (en) | Rolling bearing fault diagnosis method based on KICA-fractal theory | |
| CN109605128B (en) | Milling chatter online detection method based on power spectrum entropy difference | |
| CN110346130B (en) | Boring flutter detection method based on empirical mode decomposition and time-frequency multi-feature | |
| CN112528955B (en) | High-frequency element machining size precision prediction method and system | |
| CN116787225A (en) | Rotary cutter abrasion monitoring method based on sound vibration signal characteristic fusion graph | |
| Du et al. | Intelligent turning tool monitoring with neural network adaptive learning | |
| CN114354184A (en) | Deep learning-based method and device for establishing health early warning model of main shaft of large-scale rotating equipment | |
| CN108393744A (en) | A kind of more sensor monitoring methods of cutting tool state | |
| CN114462480A (en) | Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set | |
| Lu et al. | An early chatter detection method based on multivariate variational mode decomposition and chatter correlation factor | |
| CN114417924A (en) | Rolling bearing fault diagnosis method based on undirected graph adjacency matrix of mixed features | |
| CN113807431A (en) | Intelligent spindle state evaluation method and system based on multi-source information fusion | |
| Xu et al. | CapsFormer: A Novel Bearing Intelligent Fault Diagnosis Framework With Negligible Speed Change Under Small-Sample Conditions | |
| CN108427375B (en) | Method for monitoring cutter state based on band-pass filtering processing multi-sensor | |
| CN110749443B (en) | Rolling bearing fault diagnosis method and system based on high-order origin moment | |
| CN114310483B (en) | Numerical control machining dimension error prediction method |
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 |