CN112364762B - Mechanical transmission fault detection method based on step error frequency spectrum characteristics - Google Patents
Mechanical transmission fault detection method based on step error frequency spectrum characteristics Download PDFInfo
- Publication number
- CN112364762B CN112364762B CN202011247189.6A CN202011247189A CN112364762B CN 112364762 B CN112364762 B CN 112364762B CN 202011247189 A CN202011247189 A CN 202011247189A CN 112364762 B CN112364762 B CN 112364762B
- Authority
- CN
- China
- Prior art keywords
- error
- time sequence
- data
- steps
- mechanical transmission
- 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
- 238000001228 spectrum Methods 0.000 title claims abstract description 26
- 230000009347 mechanical transmission Effects 0.000 title claims abstract description 16
- 238000001514 detection method Methods 0.000 title claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000013499 data model Methods 0.000 claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 238000012544 monitoring process Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims abstract description 5
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 4
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000013145 classification model Methods 0.000 claims description 9
- 238000013480 data collection Methods 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 230000009466 transformation Effects 0.000 abstract 1
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 208000032369 Primary transmission Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a mechanical transmission fault detection method based on a step error frequency spectrum characteristic, which comprises a data acquisition step, a step error characteristic extraction step, a machine learning model training step and a new data model testing step; firstly, collecting time sequence signals of the operation of a part through a sensor at monitoring points such as a gear, a bearing and the like; then, carrying out transformation processing on the signals by a frequency spectrum analysis method, and extracting the step error frequency spectrum characteristics; and finally, training a machine learning model and testing a new data model through the extracted features. According to the invention, the fault signal can be effectively captured through the extracted step error characteristics, so that the performance and efficiency of the machine learning model are greatly improved, the deployment is easy to realize, and the applicability is strong.
Description
Technical Field
The invention relates to a mechanical transmission fault detection method based on a step error frequency spectrum characteristic, and belongs to the field of transmission mechanical fault detection.
Background
Wind power generation has received a great deal of attention from a worldwide perspective due to its advantages of cleanliness and reproducibility. However, fans often operate in harsh environments and are extremely labor intensive. On one hand, the damage from the fan blade, the tower to the power generation box, even the damage of the bearing and the gear can bring great safety threat and economic loss to the production process; on the other hand, the manual judgment of whether a fan is out of order by the fault diagnosis expert has high time and labor costs. These are all difficulties in fan fault detection and repair, and therefore, a way of detecting and analyzing fan faults using machine learning techniques has emerged.
Machine learning techniques require a large amount of data as training support, especially deep learning techniques. In addition, the noise tolerance of the model for data and marker information is low, and small changes in the data can lead to large changes in model predictions. However, on one hand, training data in the fan fault detection field is relatively less and has a lot of noise (such as improper sensor installation position); on the other hand, fan fault detection is a time sequence signal, the input dimension is high, the capacity requirement on the model is high, and the model is easy to be over-fitted. The prior machine learning technology needs to use the characteristics of expert design and assist in shallow model when processing mechanical transmission time sequence signals, and the method encounters performance bottleneck, so that an intelligent solution is needed.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides the mechanical transmission fault detection method based on the step error frequency spectrum characteristics, which can effectively capture fault signals through the extracted step error characteristics, greatly improves the performance and efficiency of a machine learning model, is easy to realize deployment and has strong applicability.
The technical scheme adopted by the invention is as follows:
a mechanical transmission fault detection method based on step error frequency spectrum characteristics comprises the following steps:
step one, data collection: collecting time sequence characteristics of each monitoring point through a sensor;
step two, extracting step error frequency spectrum characteristics: extracting step error spectrum characteristics through spectrum analysis;
training a machine learning model;
and step four, testing a new data model.
Preferably, the data collection in the first step includes the following steps:
step 100: determining a fault monitoring point of the transmission machinery;
step 101: deploying a signal collection sensor;
step 102: collecting a transmission mechanical rotating speed time sequence signal;
step 103: the collected data are formed into a plurality of groups of data in the forms of 'time sequence signal has faults' and 'time sequence signal has no faults'.
Further preferably, the step two step error spectrum feature extraction includes the steps of:
step 200: converting the original transmission mechanical rotating speed time sequence signal T into frequency domain data F0, and extracting a spectrogram F1 in a fixed frequency domain range;
step 201: traversing different window sizes W1, W2, …, WN in proportion;
step 202: for a given window size Wi, traversing each window in the spectrogram in turn to take the maximum value, and removing the phenomenon of 'burrs' beside;
step 203: taking the obtained index value of the peak;
step 204: calculating first-order, second-order and third-order differences of the index values, and sequencing all the difference values to obtain a peak index value difference curve C;
step 205: performing linear fitting on the peak index value difference curve C, and taking the Error value as a step Error characteristic, namely a Stage Error and a se characteristic;
step 206: repeating steps 202-205 for all window sizes W1, W2, …, WN, extracting step error features se1, se2, …, seN for each window;
step 207: and calculating the mean value, variance, maximum value and minimum value of all the step error features, and putting all the step error features together to serve as the step error features finally extracted.
Further preferably, the machine learning model training in the third step includes the steps of:
step 300: the step error characteristics extracted from each section of time sequence data are organized into a vector Vi;
step 301: the component corresponding to the time sequence data has no fault, and Yi=0 is set; otherwise, the component corresponding to the time sequence data has faults, and Yi=1 is set;
step 302: the classification model M is trained on the training data "Vi, yi".
Further preferably, the new data model test in the fourth step includes the steps of:
step 400: collecting mechanical transmission time sequence characteristic data of a part to be predicted;
step 401: extracting step error spectrum characteristics;
step 402: predicting by using the trained classification model M;
step 403: the predicted value is less than or equal to 0.5, and no fault is output;
step 404: and if the predicted value is more than 0.5, outputting the fault.
The invention has the beneficial effects that:
the method can effectively remove the burrs, adopts a plurality of sliding windows to ensure the robustness of the model, can effectively capture fault signals through extracted step error characteristics, greatly improves the performance and efficiency of the machine learning model, is easy to realize deployment and has strong applicability.
Detailed Description
The present invention will be specifically described with reference to examples.
Example 1: the embodiment is a mechanical transmission fault detection method based on step error frequency spectrum characteristics, which comprises the following steps:
step one, data collection: collecting time sequence characteristics of each monitoring point through a sensor;
step two, extracting step error frequency spectrum characteristics: extracting step error spectrum characteristics through spectrum analysis;
training a machine learning model;
and step four, testing a new data model.
Wherein, the data collection in the first step comprises the following steps:
step 100: determining a fault monitoring point of the transmission machinery;
step 101: deploying a signal collection sensor;
step 102: collecting a transmission mechanical rotating speed time sequence signal;
step 103: the collected data are formed into a plurality of groups of data in the forms of 'time sequence signal has faults' and 'time sequence signal has no faults'.
Secondly, the step error spectrum characteristic extraction in the step two comprises the following steps:
step 200: converting the original transmission mechanical rotating speed time sequence signal T into frequency domain data F0, and extracting a spectrogram F1 in a fixed frequency domain range;
step 201: traversing different window sizes W1, W2, …, WN in proportion;
step 202: for a given window size Wi, traversing each window in the spectrogram in turn to take the maximum value, and removing the phenomenon of 'burrs' beside;
step 203: taking the obtained index value of the peak;
step 204: calculating first-order, second-order and third-order differences of the index values, and sequencing all the difference values to obtain a peak index value difference curve C;
step 205: performing linear fitting on the peak index value difference curve C, and taking the Error value as a step Error characteristic, namely a Stage Error and a se characteristic;
step 206: repeating steps 202-205 for all window sizes W1, W2, …, WN, extracting step error features se1, se2, …, seN for each window;
step 207: and calculating the mean value, variance, maximum value and minimum value of all the step error features, and putting all the step error features together to serve as the step error features finally extracted.
Then, the machine learning model training in the third step includes the steps of:
step 300: the step error characteristics extracted from each section of time sequence data are organized into a vector Vi;
step 301: the component corresponding to the time sequence data has no fault, and Yi=0 is set; otherwise, the component corresponding to the time sequence data has faults, and Yi=1 is set;
step 302: the classification model M is trained on the training data "Vi, yi".
Then, the new data model test in the fourth step includes the following steps:
step 400: collecting mechanical transmission time sequence characteristic data of a part to be predicted;
step 401: extracting step error spectrum characteristics;
step 402: predicting by using the trained classification model M;
step 403: the predicted value is less than or equal to 0.5, and no fault is output;
step 404: and if the predicted value is more than 0.5, outputting the fault.
In practical application, the data collection steps are as follows: the bearing is selected as a main measuring point, a rotation speed sensor is deployed, the low-frequency vibration is measured in three directions of horizontal, vertical and axial (step 100, step 101), rotation speed time sequence signals corresponding to fans with faults and without faults are collected (step 102), each time sequence signal is collected for 4s-30s, data are organized into a time sequence signal, whether the time sequence signal has faults or not is stored, namely, the time sequence signal has faults and the time sequence signal has no faults are stored (step 103).
The step error spectrum characteristic extraction steps are as follows: converting the primary transmission mechanical rotation speed time sequence signal T into frequency domain data F0, and extracting a spectrogram F1 in a fixed frequency domain range (step 200); traversing different window sizes W1, W2, … and WN according to proportion (step 201), taking the maximum value for each window in a spectrogram for a given window size Wi, removing the phenomenon of 'burrs' beside the window in sequence (step 202) to obtain a peak index value (step 203), calculating first-order, second-order and third-order differential values of the index value, sequencing all differential values to obtain a peak index value differential curve C (step 204), performing linear fitting on the peak index value differential curve C, and taking the Error value as a step Error characteristic, namely Stage Error and se characteristic (step 205); steps 202-205 are repeated for all window sizes W1, W2, …, WN, step error features se1, se2, …, seN under each window are extracted (step 206), and then the mean, variance, maximum, minimum values of all the above step error features are calculated and all the step error features are put together as the final extracted step error feature (step 207).
The training steps of the machine learning model are as follows: step error characteristics extracted from each time series data are organized into a vector Vi (step 300), if a component corresponding to the time series data has no fault, yi=0 is set, otherwise Yi=1 (step 301), a classification model M is trained according to training data Vi and Yi (step 302), and in the embodiment, the classification model M adopts a support vector machine model; in practical application, the classification model M may also be a random forest model.
The new data model test steps are as follows: mechanical transmission time sequence data of a part to be predicted is collected (step 400), step error spectrum characteristics are extracted (step 401), the trained model M is utilized for prediction (step 402), no fault is output (step 403) if the predicted value is less than or equal to 0.5, and otherwise, the fault is output (step 404).
The foregoing is merely illustrative of the preferred embodiments of this invention, and it will be appreciated by those skilled in the art that variations and modifications may be made without departing from the principles of this invention, and such variations and modifications are to be regarded as being within the scope of this invention.
Claims (4)
1. A mechanical transmission fault detection method based on step error frequency spectrum features is characterized by comprising the following steps: the method comprises the following steps:
step one, data collection: collecting time sequence characteristics of each monitoring point through a sensor;
step two, extracting step error frequency spectrum characteristics: extracting step error spectrum characteristics through spectrum analysis;
training a machine learning model;
step four, testing a new data model;
the step-error spectrum feature extraction in the step two comprises the following steps:
step 200: converting the original transmission mechanical rotating speed time sequence signal T into frequency domain data F0, and extracting a spectrogram F1 in a fixed frequency domain range;
step 201: traversing different window sizes W1, W2, …, WN in proportion;
step 202: for a given window size Wi, traversing each window in the spectrogram in turn to take the maximum value, and removing the phenomenon of 'burrs' beside;
step 203: taking the obtained index value of the peak;
step 204: calculating first-order, second-order and third-order differences of the index values, and sequencing all the difference values to obtain a peak index value difference curve C;
step 205: performing linear fitting on the peak index value difference curve C, and taking the Error value as a step Error characteristic, namely a Stage Error and a se characteristic;
step 206: repeating steps 202-205 for all window sizes W1, W2, …, WN, extracting step error features se1, se2, …, seN for each window;
step 207: and calculating the mean value, variance, maximum value and minimum value of all the step error features, and putting all the step error features together to serve as the step error features finally extracted.
2. The method for detecting mechanical transmission failure based on step error spectrum features according to claim 1, wherein the step one data collection comprises the steps of:
step 100: determining a fault monitoring point of the transmission machinery;
step 101: deploying a signal collection sensor;
step 102: collecting a transmission mechanical rotating speed time sequence signal;
step 103: the collected data are formed into a plurality of groups of data in the forms of 'time sequence signal has faults' and 'time sequence signal has no faults'.
3. The method for detecting mechanical transmission failure based on step error spectrum features according to claim 1, wherein the training of the machine learning model in the step three comprises the following steps:
step 300: the step error characteristics extracted from each section of time sequence data are organized into a vector Vi;
step 301: the component corresponding to the time sequence data has no fault, and Yi=0 is set; otherwise, the component corresponding to the time sequence data has faults, and Yi=1 is set;
step 302: the classification model M is trained on the training data "Vi, yi".
4. A mechanical transmission failure detection method based on step error spectrum features according to claim 3, wherein the new data model test in the fourth step comprises the steps of:
step 400: collecting mechanical transmission time sequence characteristic data of a part to be predicted;
step 401: extracting step error spectrum characteristics;
step 402: predicting by using the trained classification model M;
step 403: the predicted value is less than or equal to 0.5, and no fault is output;
step 404: and if the predicted value is more than 0.5, outputting the fault.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011247189.6A CN112364762B (en) | 2020-11-10 | 2020-11-10 | Mechanical transmission fault detection method based on step error frequency spectrum characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011247189.6A CN112364762B (en) | 2020-11-10 | 2020-11-10 | Mechanical transmission fault detection method based on step error frequency spectrum characteristics |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112364762A CN112364762A (en) | 2021-02-12 |
CN112364762B true CN112364762B (en) | 2024-02-27 |
Family
ID=74508517
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011247189.6A Active CN112364762B (en) | 2020-11-10 | 2020-11-10 | Mechanical transmission fault detection method based on step error frequency spectrum characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112364762B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102095430A (en) * | 2010-11-18 | 2011-06-15 | 合肥工业大学 | Sensor dynamic error frequency-domain correction technology based on step response |
CN103983343A (en) * | 2014-05-29 | 2014-08-13 | 武汉大学 | Satellite platform chattering detection method and system based on multispectral image |
GB201415567D0 (en) * | 2014-09-03 | 2014-10-15 | Ucl Business Plc | Method and apparatus for the detection of faults in data computations |
CN105628649A (en) * | 2015-12-24 | 2016-06-01 | 山东省科学院激光研究所 | Method for extracting absorption characteristic peak in gas field monitoring |
CN108489529A (en) * | 2018-02-09 | 2018-09-04 | 中国科学院国家空间科学中心 | A kind of method for detecting weak signals based on high-order statistic |
CN108613802A (en) * | 2018-05-10 | 2018-10-02 | 重庆大学 | A kind of mechanical failure diagnostic method based on depth mixed network structure |
CN109102005A (en) * | 2018-07-23 | 2018-12-28 | 杭州电子科技大学 | Small sample deep learning method based on shallow Model knowledge migration |
CN109918417A (en) * | 2019-02-28 | 2019-06-21 | 西安交通大学 | Time series data adaptive segmentation, dimensionality reduction and characterizing method and application based on wavelet transformation |
CN110196173A (en) * | 2019-05-05 | 2019-09-03 | 成都大学 | A kind of strong noise environment machine operation troubles online test method based on self study |
CN110598676A (en) * | 2019-09-25 | 2019-12-20 | 南京邮电大学 | Deep learning gesture electromyographic signal identification method based on confidence score model |
CN111537056A (en) * | 2020-07-08 | 2020-08-14 | 浙江浙能天然气运行有限公司 | Pipeline along-line third-party construction dynamic early warning method based on SVM and time-frequency domain characteristics |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10782675B2 (en) * | 2018-12-13 | 2020-09-22 | Analog Devices, Inc | Sensor mounting assembly for machine condition monitoring |
-
2020
- 2020-11-10 CN CN202011247189.6A patent/CN112364762B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102095430A (en) * | 2010-11-18 | 2011-06-15 | 合肥工业大学 | Sensor dynamic error frequency-domain correction technology based on step response |
CN103983343A (en) * | 2014-05-29 | 2014-08-13 | 武汉大学 | Satellite platform chattering detection method and system based on multispectral image |
GB201415567D0 (en) * | 2014-09-03 | 2014-10-15 | Ucl Business Plc | Method and apparatus for the detection of faults in data computations |
CN105628649A (en) * | 2015-12-24 | 2016-06-01 | 山东省科学院激光研究所 | Method for extracting absorption characteristic peak in gas field monitoring |
CN108489529A (en) * | 2018-02-09 | 2018-09-04 | 中国科学院国家空间科学中心 | A kind of method for detecting weak signals based on high-order statistic |
CN108613802A (en) * | 2018-05-10 | 2018-10-02 | 重庆大学 | A kind of mechanical failure diagnostic method based on depth mixed network structure |
CN109102005A (en) * | 2018-07-23 | 2018-12-28 | 杭州电子科技大学 | Small sample deep learning method based on shallow Model knowledge migration |
CN109918417A (en) * | 2019-02-28 | 2019-06-21 | 西安交通大学 | Time series data adaptive segmentation, dimensionality reduction and characterizing method and application based on wavelet transformation |
CN110196173A (en) * | 2019-05-05 | 2019-09-03 | 成都大学 | A kind of strong noise environment machine operation troubles online test method based on self study |
CN110598676A (en) * | 2019-09-25 | 2019-12-20 | 南京邮电大学 | Deep learning gesture electromyographic signal identification method based on confidence score model |
CN111537056A (en) * | 2020-07-08 | 2020-08-14 | 浙江浙能天然气运行有限公司 | Pipeline along-line third-party construction dynamic early warning method based on SVM and time-frequency domain characteristics |
Non-Patent Citations (6)
Title |
---|
A spectral self-focusing fault diagnosis method for automotive transmissions under gear-shifting conditions;Xiwei Li 等;《Mechanical Systems and Signal Processing》;1-18 * |
地震和***事件源波形信号的卷积神经网络分类研究;陈润航 等;《地球物理学进展》;第33卷(第4期);1331-1338 * |
基于NS3-gym框架的智能干扰规避***设计与实现;陈海涛 等;《计算机工程与应用》;1-9 * |
基于基座多振动信号融合的转子故障诊断;张田 等;《机械工程师》;14-15 * |
基于超声导波的航空航天结构损伤诊断成像技术研究进展;郑跃滨 等;《航空控制技术》;24-43 * |
纸质心电图数字化信息提取与分析;孙小涵;《中国优秀硕士学位论文全文数据库 (医药卫生科技辑)》;E060-16 * |
Also Published As
Publication number | Publication date |
---|---|
CN112364762A (en) | 2021-02-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109655266B (en) | Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis | |
CN102798529A (en) | Method and system for diagnosing bearing faults of large-size wind turbine bearing | |
CN106650122B (en) | A kind of equipment variable parameter operation methods of risk assessment | |
CN101603854A (en) | The rotating machinery non-stationery vibration signal instantaneous frequency estimation algorithm in start and stop period | |
CN112347944B (en) | Mechanical transmission fault detection method based on zero value proportion frequency spectrum characteristics | |
Zhang et al. | Time–frequency analysis via complementary ensemble adaptive local iterative filtering and enhanced maximum correlation kurtosis deconvolution for wind turbine fault diagnosis | |
CN111209934A (en) | Fan fault prediction and alarm method and system | |
CN113391207A (en) | Motor fault detection method, medium and system | |
CN112177865B (en) | Method for solving marking noise and insufficient marks in fan fault detection | |
CN116358871B (en) | Rolling bearing weak signal composite fault diagnosis method based on graph rolling network | |
CN103926075B (en) | Method for improving fault analysis precision of wind turbine generating set gear box | |
CN114742111B (en) | Fault diagnosis method and system based on parameter adaptive characteristic modal decomposition | |
Berntsen et al. | Enhanced demodulation band selection based on Operational Modal Analysis (OMA) for bearing diagnostics | |
Xu et al. | Caputo-Fabrizio fractional order derivative stochastic resonance enhanced by ADOF and its application in fault diagnosis of wind turbine drivetrain | |
CN111046790A (en) | Pump bearing fault diagnosis method | |
CN112364762B (en) | Mechanical transmission fault detection method based on step error frequency spectrum characteristics | |
CN113776818A (en) | Offshore wind turbine generator system gear system fault diagnosis method based on Park transformation | |
CN117093938A (en) | Fan bearing fault detection method and system based on deep learning | |
CN116821762A (en) | Mechanical fault diagnosis method based on multi-scale graph attention fusion network | |
Koukoura et al. | On the use of AI based vibration condition monitoring of wind turbine gearboxes | |
EP3953580B1 (en) | A method for computer-implemented monitoring of a component of a wind turbine | |
CN113447267B (en) | Vibration signal analysis-based complete gearbox state evaluation method and system | |
Zheng et al. | Wavelet packet decomposition and neural network based fault diagnosis for elevator excessive vibration | |
CN113947017B (en) | Method for predicting residual service life of rolling bearing | |
Xue et al. | Structural fault diagnosis of rotating machinery based on distinctive frequency components and support vector machines |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20221209 Address after: 210023 163 Xianlin Road, Qixia District, Nanjing, Jiangsu Applicant after: NANJING University Applicant after: Nanjing Zhigu Artificial Intelligence Research Institute Co.,Ltd. Address before: 210000 building C4, Hongfeng Science Park, Nanjing Economic and Technological Development Zone, Nanjing City, Jiangsu Province Applicant before: Nanjing Zhigu Artificial Intelligence Research Institute Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |