CN114036998A - Method and system for fault detection of industrial hardware based on machine learning - Google Patents
Method and system for fault detection of industrial hardware based on machine learning Download PDFInfo
- Publication number
- CN114036998A CN114036998A CN202111120733.5A CN202111120733A CN114036998A CN 114036998 A CN114036998 A CN 114036998A CN 202111120733 A CN202111120733 A CN 202111120733A CN 114036998 A CN114036998 A CN 114036998A
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- industrial hardware
- class
- clustering
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000010801 machine learning Methods 0.000 title claims abstract description 22
- 238000013145 classification model Methods 0.000 claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000007477 logistic regression Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 6
- 238000007637 random forest analysis Methods 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 6
- 238000012706 support-vector machine Methods 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method and a system for carrying out fault detection on industrial hardware based on machine learning, which relate to the technical field of fault detection and are realized by the following steps: collecting historical data and latest data generated by the operation of industrial hardware; setting a frequency threshold value based on the frequency of the industrial hardware faults, wherein historical data smaller than the frequency threshold value is used as first-class fault data, and historical data larger than the frequency threshold value is used as second-class fault data; constructing a first-class fault model, training the first-class fault model by using first-class fault data, and completing parameter adjustment; clustering the second-class fault data by using a clustering method, and classifying the faults according to the working state of the industrial hardware and a clustering result; constructing a classification model, taking the second-class fault data and the working state of the industrial hardware as input, taking the fault classification result as output, and training the classification model; and inputting the latest data into the trained fault models and classification models in sequence to finish fault type judgment. The invention can detect and judge the fault type.
Description
Technical Field
The invention relates to the technical field of fault detection, in particular to a method and a system for detecting faults of industrial hardware based on machine learning.
Background
At present, the technology of artificial intelligence internet of things is gradually emerging, and more industrial hardware can find corresponding sensors to monitor the working state of the sensors.
At present, the mainstream monitoring means is to monitor each index of hardware by using a sensor, define a safety range for each index, and send out a safety alarm if data uploaded by the sensor exceeds the set safety range. However, the method only aims at a single variable during detection, and does not link all variables; secondly, even if the same hardware is used, the safety range of the hardware is different due to different places where the hardware is arranged. Therefore, it is difficult to set the safety range in a targeted manner.
Disclosure of Invention
Aiming at the requirements and the defects of the prior art development, the invention provides a method and a system for detecting the fault of industrial hardware based on machine learning.
Firstly, the invention provides a method for detecting faults of industrial hardware based on machine learning, and the technical scheme adopted for solving the technical problems is as follows:
a method for fault detection of industrial hardware based on machine learning includes the following steps:
s1, aiming at the same type of industrial hardware, collecting historical data generated by the operation of different industrial hardware;
step S2, setting a frequency threshold value based on the frequency of the industrial hardware faults, taking the industrial hardware historical data smaller than the frequency threshold value as first-class fault data, and taking the industrial hardware historical data larger than the frequency threshold value as second-class fault data;
s3, constructing a class I fault model based on a DBSCAN algorithm and an isolated forest algorithm, training the class I fault model by using class I fault data, and completing parameter adjustment to obtain a trained class I fault model;
s4, clustering the two types of fault data by using a clustering method, carrying out fault classification by professional technicians according to the working state of the industrial hardware and the clustering result, constructing a classification model, taking the two types of fault data and the working state of the industrial hardware as input, and taking the fault classification result as output, and finishing the training of the classification model;
step S5, collecting the latest data generated by the operation of the industrial hardware, using the trained fault model to judge whether the latest data belongs to the fault data, if so, sending an alarm, if not, inputting the latest data into the trained classification model, if the trained classification model outputs a specific classification fault, judging the latest data as the fault data of the second type, and sending the alarm.
Optionally, the frequency threshold includes a minimum value and a maximum value;
and taking the industrial hardware historical data smaller than the minimum value as one type of fault data, and taking the industrial hardware historical data larger than the maximum value as two types of fault data.
Optionally, when step S4 is executed, the two types of fault data are clustered by using three clustering methods, namely a gaussian mixture model, spectral clustering and K-means, and then, by combining results of the three clustering methods, a professional performs fault classification according to the working state of the industrial hardware.
Further optionally, the working states of the industrial hardware include a normal working state and a fault working state;
and aiming at the clustering result of the second-class fault data, the working state of the industrial hardware is used as a label of the second-class fault data, and the labeled second-class fault data is used as the input of a classification model.
Optionally, a classification model is constructed by using any one of methods of logistic regression, random forest, support vector machine, linear discriminant analysis, quadratic discriminant analysis, naive Bayes and K nearest neighbor;
and based on the two types of fault data, verifying the reliability and stability of the constructed classification model by using a cross verification method.
Secondly, the invention provides a system for detecting faults of industrial hardware based on machine learning, and the technical scheme adopted for solving the technical problems is as follows:
a system for fault detection of industrial hardware based on machine learning is provided, which comprises:
the acquisition module is used for acquiring historical data and latest data generated by the operation of different industrial hardware under the same type;
the setting and dividing module is used for setting a frequency threshold according to the frequency of the industrial hardware faults, taking industrial hardware historical data smaller than the frequency threshold as first-class fault data, and taking industrial hardware historical data larger than the frequency threshold as second-class fault data;
the first training module is used for constructing a class I fault model based on a DBSCAN algorithm and an isolated forest algorithm, training the class I fault model by using class I fault data, and completing parameter adjustment to obtain a trained class I fault model;
the clustering module is used for clustering the two types of fault data by using a clustering method;
the marking module is used for assisting a professional technician to classify faults according to the working state of the industrial hardware and the clustering result of the clustering module;
a second training module is constructed and used for constructing a classification model, taking the second-class fault data and the working state of the industrial hardware as input, taking the fault classification result as output and finishing the training of the classification model;
the trained fault model judges whether the latest data generated by the operation of the industrial hardware belongs to the first class fault data, if so, an alarm is sent, if not, the latest data is input into the trained classification model, and if the trained classification model outputs a specific classification fault, the latest data is judged to be the second class fault data and an alarm is sent.
Optionally, the frequency threshold set by the setting and dividing module includes a minimum value and a maximum value;
and taking the industrial hardware historical data smaller than the minimum value as one type of fault data, and taking the industrial hardware historical data larger than the maximum value as two types of fault data.
Optionally, the related clustering module uses three clustering methods of a gaussian mixture model, spectral clustering and K-means to cluster the two types of fault data respectively, and a professional carries out fault classification according to the working state of the industrial hardware by combining the results of the three clustering methods;
the working state of the industrial hardware comprises a normal working state and a fault working state, and for the clustering result of the two types of fault data, the professional technicians use the working state of the industrial hardware as the labels of the two types of fault data by means of the marking module, and the labeled two types of fault data are used as the input of the classification model.
Further optionally, the related training module II is used for constructing a classification model by using any one of logistic regression, random forest, support vector machine, linear discriminant analysis, secondary discriminant analysis, naive Bayes and K nearest neighbor.
Compared with the prior art, the method and the system for detecting the fault of the industrial hardware based on the machine learning have the beneficial effects that:
(1) firstly, training a class I fault model and a class I classification model by dividing historical data generated by the operation of industrial hardware, then, inputting latest data generated by the industrial hardware into the class I fault model and the class I classification model which are trained, detecting whether a fault occurs in the latest data, judging the type of the fault, and finally, giving corresponding alarm according to a detection result;
(2) the invention can realize fault detection, judge the fault type, and monitor whether certain industrial hardware in the production chain normally operates very conveniently, thereby having stronger applicability.
Drawings
FIG. 1 is a flow chart diagram of a first embodiment of the present invention;
fig. 2 is a connection block diagram of the second embodiment of the present invention.
The reference information in the drawings indicates:
1. an acquisition module 2, a setting and dividing module 3, a first training module,
4. a clustering module 5, a marking module 6, a second construction training module,
7. a class of fault models, 7', a class of fault models that are trained,
8. and 8', training the finished classification model.
Detailed Description
In order to make the technical scheme, the technical problems to be solved and the technical effects of the present invention more clearly apparent, the following technical scheme of the present invention is clearly and completely described with reference to the specific embodiments.
The first embodiment is as follows:
with reference to fig. 1, the present embodiment provides a method for performing fault detection on industrial hardware based on machine learning, and the implementation content includes:
and step S1, aiming at the same type of industrial hardware, acquiring historical data generated by the operation of different industrial hardware by using a sensor.
And step S2, setting a frequency threshold value based on the frequency of the industrial hardware faults, wherein the frequency threshold value comprises a minimum value and a maximum value, the historical data of the industrial hardware smaller than the minimum value is used as first-class fault data, and the historical data of the industrial hardware larger than the maximum value is used as second-class fault data.
Step S3, constructing a class I fault model 7 based on the DBSCAN algorithm and the isolated forest algorithm, training the class I fault model 7 by using class I fault data, and completing parameter adjustment to obtain a class I fault model 7' after training.
S4, clustering two types of fault data by using a Gaussian mixture model, spectral clustering and a K-means clustering method, and classifying faults according to the working state of the industrial hardware by professional technicians by combining the results of the three clustering methods, wherein the working state of the industrial hardware comprises a normal working state and a fault working state, the working state of the industrial hardware is used as a label of the two types of fault data according to the clustering result of the two types of fault data, and the labeled two types of fault data are used as the input of a classification model;
and (3) constructing a classification model 8 by using any one of methods of logistic regression, random forest, support vector machine, linear discriminant analysis, quadratic discriminant analysis, naive Bayes and K nearest neighbor, and finishing the training of the classification model 8 by using the two types of fault data and the working state of industrial hardware as input and using the fault classification result as output.
Based on the two types of fault data, the reliability and stability of the constructed classification model 8 are verified by using a cross validation method.
Step S5, collecting the latest data generated by the operation of the industrial hardware, using the one-class fault model 7 ' after training to judge whether the latest data belongs to the one-class fault data, if so, sending an alarm, if not, inputting the latest data into the classification model 8 ' after training, if the classification model 8 ' after training outputs a specific classification fault, judging the latest data as the two-class fault data, and sending an alarm.
Example two:
with reference to fig. 2, the present embodiment provides a system for performing fault detection on industrial hardware based on machine learning, and the structure of the system includes:
the acquisition module 1 is used for acquiring historical data and latest data generated by the operation of different industrial hardware under the same type;
the setting and dividing module 2 is used for setting a frequency threshold according to the frequency of the industrial hardware faults, wherein the set frequency threshold comprises a minimum value and a maximum value, the historical data of the industrial hardware smaller than the minimum value is used as first-class fault data, and the historical data of the industrial hardware larger than the maximum value is used as second-class fault data;
the building training module I3 is used for building a class I fault model 7 based on a DBSCAN algorithm and an isolated forest algorithm, training the class I fault model 7 by using class I fault data, completing parameter adjustment and obtaining a trained class I fault model 7';
the clustering module 4 is used for clustering the two types of fault data by using a Gaussian mixture model, a spectral clustering method and a K-means clustering method respectively;
the marking module 5 is used for assisting a professional to classify the fault according to the working state of the industrial hardware and the three clustering results of the clustering module 4, wherein the working state of the industrial hardware comprises a normal working state and a fault working state, and for the clustering results of the two types of fault data, the professional uses the working state of the industrial hardware as a label of the two types of fault data by means of the marking module 5;
a second training module 6 is constructed, the second training module is used for constructing a classification model 8 by using any one of methods of logistic regression, random forest, support vector machine, linear discriminant analysis, secondary discriminant analysis, naive Bayes and K nearest neighbor, the labeled second-class fault data is used as the input of the classification model 8, and the fault classification result is used as the output to finish the training of the classification model 8;
the trained first-class fault model 7 ' judges whether the latest data generated by the operation of the industrial hardware belongs to first-class fault data, if the latest data belongs to the first-class fault data, an alarm is sent out, if the latest data does not belong to the first-class fault data, the latest data is input into the trained classification model 8 ', and if the trained classification model 8 ' outputs specific classification faults, the latest data is judged to be second-class fault data, and an alarm is sent out.
In summary, the method and the system for detecting the fault of the industrial hardware based on the machine learning can detect whether the industrial hardware has the fault during operation, judge the specific type of the fault, and give corresponding alarm according to the detection result, so that the method and the system have strong applicability.
Based on the above embodiments of the present invention, those skilled in the art should make any improvements and modifications to the present invention without departing from the principle of the present invention, and therefore, the present invention should fall into the protection scope of the present invention.
Claims (9)
1. A method for fault detection of industrial hardware based on machine learning is characterized in that the implementation content comprises the following steps:
s1, aiming at the same type of industrial hardware, collecting historical data generated by the operation of different industrial hardware;
step S2, setting a frequency threshold value based on the frequency of the industrial hardware faults, taking the industrial hardware historical data smaller than the frequency threshold value as first-class fault data, and taking the industrial hardware historical data larger than the frequency threshold value as second-class fault data;
s3, constructing a class I fault model based on a DBSCAN algorithm and an isolated forest algorithm, training the class I fault model by using class I fault data, and completing parameter adjustment to obtain a trained class I fault model;
s4, clustering the two types of fault data by using a clustering method, carrying out fault classification by professional technicians according to the working state of the industrial hardware and the clustering result, constructing a classification model, taking the two types of fault data and the working state of the industrial hardware as input, and taking the fault classification result as output, and finishing the training of the classification model;
step S5, collecting the latest data generated by the operation of the industrial hardware, using the trained fault model to judge whether the latest data belongs to the fault data, if so, sending an alarm, if not, inputting the latest data into the trained classification model, if the trained classification model outputs a specific classification fault, judging the latest data as the fault data of the second type, and sending the alarm.
2. The method of machine learning-based fault detection for industrial hardware of claim 1, wherein said frequency threshold comprises a minimum and a maximum;
and taking the industrial hardware historical data smaller than the minimum value as one type of fault data, and taking the industrial hardware historical data larger than the maximum value as two types of fault data.
3. The method for fault detection of industrial hardware based on machine learning of claim 1, wherein in step S4, the two types of fault data are clustered by using three clustering methods of gaussian mixture model, spectral clustering and K-means, and the professional carries out fault classification according to the working state of the industrial hardware by combining the results of the three clustering methods.
4. The method for fault detection of industrial hardware based on machine learning according to claim 3, wherein the working state of the industrial hardware comprises a normal working state and a fault working state;
and aiming at the clustering result of the second-class fault data, the working state of the industrial hardware is used as a label of the second-class fault data, and the labeled second-class fault data is used as the input of a classification model.
5. The method for fault detection of industrial hardware based on machine learning according to claim 3, wherein a classification model is constructed by using any one of logistic regression, random forest, support vector machine, linear discriminant analysis, quadratic discriminant analysis, naive Bayes and K nearest neighbor method;
and based on the two types of fault data, verifying the reliability and stability of the constructed classification model by using a cross verification method.
6. A system for fault detection of industrial hardware based on machine learning is characterized in that the structure comprises:
the acquisition module is used for acquiring historical data and latest data generated by the operation of different industrial hardware under the same type;
the setting and dividing module is used for setting a frequency threshold according to the frequency of the industrial hardware faults, taking industrial hardware historical data smaller than the frequency threshold as first-class fault data, and taking industrial hardware historical data larger than the frequency threshold as second-class fault data;
the first training module is used for constructing a class I fault model based on a DBSCAN algorithm and an isolated forest algorithm, training the class I fault model by using class I fault data, and completing parameter adjustment to obtain a trained class I fault model;
the clustering module is used for clustering the two types of fault data by using a clustering method;
the marking module is used for assisting a professional technician to classify faults according to the working state of the industrial hardware and the clustering result of the clustering module;
a second training module is constructed and used for constructing a classification model, taking the second-class fault data and the working state of the industrial hardware as input, taking the fault classification result as output and finishing the training of the classification model;
the trained fault model judges whether the latest data generated by the operation of the industrial hardware belongs to the first class fault data, if so, an alarm is sent, if not, the latest data is input into the trained classification model, and if the trained classification model outputs a specific classification fault, the latest data is judged to be the second class fault data and an alarm is sent.
7. The system for fault detection of industrial hardware based on machine learning according to claim 6, wherein the frequency threshold set by the setting and dividing module includes a minimum value and a maximum value;
and taking the industrial hardware historical data smaller than the minimum value as one type of fault data, and taking the industrial hardware historical data larger than the maximum value as two types of fault data.
8. The system for fault detection of industrial hardware based on machine learning according to claim 6, wherein the clustering module uses three clustering methods of Gaussian mixture model, spectral clustering and K-means to cluster two types of fault data respectively, and a professional carries out fault classification according to the working state of the industrial hardware by combining the results of the three clustering methods;
the working state of the industrial hardware comprises a normal working state and a fault working state, and for the clustering result of the two types of fault data, the professional technicians use the working state of the industrial hardware as the labels of the two types of fault data by means of the marking module, and the labeled two types of fault data are used as the input of the classification model.
9. The system for fault detection of industrial hardware based on machine learning of claim 8, wherein the second training module is configured to construct the classification model by using any one of logistic regression, random forest, support vector machine, linear discriminant analysis, quadratic discriminant analysis, naive Bayes, and K nearest neighbor methods.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111120733.5A CN114036998A (en) | 2021-09-24 | 2021-09-24 | Method and system for fault detection of industrial hardware based on machine learning |
| PCT/CN2022/105321 WO2023045512A1 (en) | 2021-09-24 | 2022-07-13 | Method and system for performing fault detection on industrial hardware on basis of machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111120733.5A CN114036998A (en) | 2021-09-24 | 2021-09-24 | Method and system for fault detection of industrial hardware based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114036998A true CN114036998A (en) | 2022-02-11 |
Family
ID=80134510
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111120733.5A Pending CN114036998A (en) | 2021-09-24 | 2021-09-24 | Method and system for fault detection of industrial hardware based on machine learning |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114036998A (en) |
| WO (1) | WO2023045512A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
| WO2023045512A1 (en) * | 2021-09-24 | 2023-03-30 | 浪潮集团有限公司 | Method and system for performing fault detection on industrial hardware on basis of machine learning |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118035779B (en) * | 2024-04-12 | 2024-06-18 | 山东海纳智能装备科技股份有限公司 | Torque measuring method and system for ultra-high pressure filter press |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
| CN108731923A (en) * | 2018-03-28 | 2018-11-02 | 中控技术（西安）有限公司 | A kind of fault detection method and device of rotating machinery |
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
| CN111325410A (en) * | 2020-03-13 | 2020-06-23 | 安图实验仪器(郑州)有限公司 | General fault early warning system based on sample distribution and early warning method thereof |
| CN112097365A (en) * | 2020-07-10 | 2020-12-18 | 珠海派诺科技股份有限公司 | Air conditioner fault detection and identification method and device based on prediction and classification model |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109753998A (en) * | 2018-12-20 | 2019-05-14 | 山东科技大学 | The fault detection method and system, computer program of network are generated based on confrontation type |
| CN111723519B (en) * | 2020-05-29 | 2022-11-15 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis device and method based on logistic regression and naive Bayes |
| CN111949429A (en) * | 2020-08-17 | 2020-11-17 | 山东超越数控电子股份有限公司 | Server fault monitoring method and system based on density clustering algorithm |
| CN114036998A (en) * | 2021-09-24 | 2022-02-11 | 浪潮集团有限公司 | Method and system for fault detection of industrial hardware based on machine learning |
-
2021
- 2021-09-24 CN CN202111120733.5A patent/CN114036998A/en active Pending
-
2022
- 2022-07-13 WO PCT/CN2022/105321 patent/WO2023045512A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
| CN108731923A (en) * | 2018-03-28 | 2018-11-02 | 中控技术（西安）有限公司 | A kind of fault detection method and device of rotating machinery |
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
| CN111325410A (en) * | 2020-03-13 | 2020-06-23 | 安图实验仪器(郑州)有限公司 | General fault early warning system based on sample distribution and early warning method thereof |
| CN112097365A (en) * | 2020-07-10 | 2020-12-18 | 珠海派诺科技股份有限公司 | Air conditioner fault detection and identification method and device based on prediction and classification model |
Non-Patent Citations (3)
| Title |
|---|
| 石鑫等: "电力变压器状态监测数据聚类研究", 《电力信息与通信技术》 * |
| 肖黎等: "基于半监督学习方法的磨煤机故障预警", 《热力发电》 * |
| 陈海彬等: "基于改进K-means聚类的kNN故障检测研究", 《沈阳化工大学学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045512A1 (en) * | 2021-09-24 | 2023-03-30 | 浪潮集团有限公司 | Method and system for performing fault detection on industrial hardware on basis of machine learning |
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
| CN115860714B (en) * | 2022-11-28 | 2023-08-08 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023045512A1 (en) | 2023-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114036998A (en) | Method and system for fault detection of industrial hardware based on machine learning | |
| CN110210512B (en) | Automatic log anomaly detection method and system | |
| US8868985B2 (en) | Supervised fault learning using rule-generated samples for machine condition monitoring | |
| CN110287552B (en) | Motor bearing fault diagnosis method and system based on improved random forest algorithm | |
| CN103914735B (en) | A kind of fault recognition method and system based on Neural Network Self-learning | |
| CN111505424A (en) | Large experimental device power equipment fault diagnosis method based on deep convolutional neural network | |
| KR102444442B1 (en) | System and method for diagnosing facility fault | |
| CN104021238A (en) | Lead-acid power battery system fault diagnosis method | |
| CN112905371B (en) | Software change checking method and device based on heterogeneous multi-source data anomaly detection | |
| CN108229553A (en) | A kind of OTDR curve datas analysis method | |
| CN116308304B (en) | New energy intelligent operation and maintenance method and system based on meta learning concept drift detection | |
| CN116166983A (en) | Train bow net fault detection method based on machine learning | |
| CN108334902A (en) | A kind of track train equipment room smog fireproof monitoring method based on deep learning | |
| CN112070180B (en) | Power grid equipment state judging method and device based on information physical bilateral data | |
| CN110620760A (en) | FlexRay bus fusion intrusion detection method and detection device for SVM (support vector machine) and Bayesian network | |
| Sun et al. | Air leakage detection of pneumatic train door subsystems using open set recognition | |
| CN117975372B (en) | Construction site safety detection system and method based on YOLOv and transducer encoder | |
| CN115310367A (en) | Mine air quality monitoring system based on LSTM neural network | |
| CN112199295A (en) | Deep neural network defect positioning method and system based on frequency spectrum | |
| CN112016597A (en) | Depth sampling method based on Bayesian imbalance measurement in machine learning | |
| CN116523722A (en) | Environment monitoring analysis system with machine learning capability | |
| CN113537360B (en) | Point-to-point classification fault detection method based on deep learning | |
| CN110490218A (en) | A kind of rolling bearing fault self-learning method based on two-stage DBN | |
| CN114372500A (en) | Intelligent factory control system based on big data | |
| TWI524724B (en) | Telecommunication network alarm correlation analysis 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220211 |