CN109635008A - A kind of equipment fault detection method based on machine learning - Google Patents
A kind of equipment fault detection method based on machine learning Download PDFInfo
- Publication number
- CN109635008A CN109635008A CN201811572223.XA CN201811572223A CN109635008A CN 109635008 A CN109635008 A CN 109635008A CN 201811572223 A CN201811572223 A CN 201811572223A CN 109635008 A CN109635008 A CN 109635008A
- Authority
- CN
- China
- Prior art keywords
- data
- primary data
- equipment
- machine learning
- detection method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- 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
Abstract
The invention belongs to industrial equipment data collection and analysis technical fields, disclose a kind of equipment fault detection method based on machine learning.The technical solution of the present invention is as follows: S1. obtains the primary data of the equipment of the failure of acquisition；S2. cleaning operation and postsearch screening are carried out to primary data；S3. operation is optimized to pre- data and exports search result；S4. data are classified to obtain initial results collection；S5. gradually classify to initial results collection；S6. by contrast weight average result and initial results collection, whether the equipment fault for judging that the primary data represents belongs to known fault；S7. fault message is exported by man-machine interface or Mishap Database is added in the primary data.The present invention can be realized the self-teaching function of machine, while establish the Mishap Database of equipment, can accurately and timely locking equipment failure and solution really, can reduce machinery maintenance number, improve overhaul efficiency significantly, be suitable for promoting the use of.
Description
Technical field
The invention belongs to industrial equipment data collection and analysis technical fields, and in particular to a kind of setting based on machine learning
Standby fault detection method.
Background technique
In industrial processes, plant maintenance is essential work.However at present in the mistake safeguarded to equipment
Cheng Zhong is usually to pass through manually to examine industrial equipment in real time again until equipment quotes failure or discovery irregular working
Look into and verify, the method for this artificial detection cannot the failure to industrial equipment detected automatically, the degree of automation is low.
There is the rolling bearing fault detection based on machine learning in the prior art, but the technology is mainly in the axis of rolling
Holding occur in process of production various, there is characteristic data and its representative information to be analyzed, therefore cannot be extensive
Suitable for the data type of the equipment such as various machining tools, and then guarantee the safety in production of equipment.
Summary of the invention
In order to solve the above problems existing in the present technology, it is an object of that present invention to provide a kind of setting based on machine learning
Standby fault detection method, the present invention can be realized the self-teaching function of machine, while establish the Mishap Database of equipment, energy
Enough accurately and timely locking equipment failure and solution, the detection work of equipment can prevent the generation of mechanical breakdown really, can
Machinery maintenance number is reduced, improves overhaul efficiency significantly, so that the production efficiency for promoting enterprise is very big, Shi Sheng enterprise warp
Ji benefit gets a promotion, therefore is extremely important for the production practices of enterprise.
The technical scheme adopted by the invention is as follows:
A kind of equipment fault detection method based on machine learning, comprising the following steps:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
Preferably, in the step S1, data with existing interface of the primary data from current device and/or currently set
Standby external sensor.
Preferably, cleaning operation includes that column data editor, cell editor and grouping item are compiled in the step S2
Volume, it further include checking respectively for data storage format, length, coded format and reference field.
Preferably, optimizing operation, specific step is as follows in the step S3:
S301. initial time t=0 randomly chooses initial population P_0；
S302. individual adaptation degree functional value F (P_t) is calculated；
If S303. fitness function value corresponding to optimum individual is sufficiently large in population or algorithm continuous operation mostly generation
And the optimal adaptation degree of individual goes to step S306 without significantly improving；
S304. current time t '=t+1 selects P_t using selection operator method from P_ (t-1)；
S305. P_t is intersected, mutation operation, goes to step S302 later；
S306. optimal kernel functional parameter and penalty factor are provided, and with its training dataset to obtain multiple startings
Point.
Preferably, in the step S3, using support vector machines (SVM) classifier, according to structural risk minimization
Principle constructs optimal classification surface, so that the error in classification to primary data is minimum.
Preferably, every after a classifier, weighting classification calculated result will do it phase in the step S5
The weighted value answered updates.
Preferably, summed according to the weighted results exported in each classifier in circle in the step S5,
Obtain final output result.
Preferably, the calculation formula of weighted value is as follows:
Hfinal=sign=(∑ ai·Hi),
Wherein, HiFor the basic classification device obtained according to the minimum threshold value of error rate, aiIndicate HiIn final classification device
Significance level.
The invention has the benefit that
1) as mechanical equipment is more and more accurate complicated, equipment fault detection work seems especially difficult, and the present invention can
The self-teaching function of realizing machine, while establishing the Mishap Database of equipment, being capable of accurately and timely locking equipment failure really
And solution, the detection work of mechanical equipment can prevent the generation of mechanical breakdown, can reduce machinery maintenance number,
Overhaul efficiency is improved significantly, so that the production efficiency for promoting enterprise is very big, so that Business Economic Benefit gets a promotion, therefore right
It is extremely important in the production practices of enterprise；
2) in model construction process, the present invention replaces traditional single model syndrome check method and artificial fault detection
Method is joined with genetic algorithm tune instead of traditional parameter adjustment method using the combination of genetic algorithm and AdaBoost Meta algorithm, can
To improve the accuracy rate of algorithm, improve efficiency, using AdaBoost Meta algorithm can support Various Classifiers on Regional (support vector machines, with
Machine forest, BP neural network) customized addition, carry out the detection of failure, training simultaneously makes the present invention have autonomous troubleshooting
The ability of generation, without carrying out situation judgement by artificial；
3) during machine learning, collected fault message is stored in the corresponding Mishap Database of current device
In, to carry out related situation record to passing break down, realize the process of machine learning, it is similar when occurring for the second time
Failure of the same race can prompt Trouble cause and solution, administrative staff that can pass through people in first time in man-machine interface
Machine interface carries out arrangement update to the information in Mishap Database；
4) after training, the present invention, which can have all kinds of machine tooling equipment, production equipment, considerable degree of generally to be fitted
With property, can be widely used in the industrial production application scenarios of the various various machinery production failures being likely to occur, suitable for pushing away
It is wide to use.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is flow diagram of the invention.
Specific embodiment
Hereinafter reference will be made to the drawings, by way of example describe in detail it is provided by the invention it is a kind of based on deep learning and
The pedestrian traffic statistical method of multiple target tracking.It should be noted that being used to help for the explanation of these way of example
Assistant solves the present invention, but and does not constitute a limitation of the invention.
The terms "and/or", only a kind of incidence relation for describing affiliated partner, indicates that there may be three kinds of passes
System, for example, A and/or B, can indicate: individualism A, individualism B exist simultaneously tri- kinds of situations of A and B, the terms
"/and " it is to describe another affiliated partner relationship, indicate may exist two kinds of relationships, for example, A/ and B, can indicate: individually depositing
In A, two kinds of situations of individualism A and B, in addition, character "/" herein, typicallying represent forward-backward correlation object is a kind of "or" pass
System.
Embodiment 1:
As shown in Figure 1, the present embodiment provides a kind of equipment fault detection method based on machine learning, including following step
It is rapid:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
In the present embodiment, in step S1, data with existing interface and/or current device of the primary data from current device
External sensor.
Embodiment 2
On the basis of embodiment 1, as shown in Figure 1, the present embodiment provides a kind of, the equipment fault based on machine learning is examined
Survey method, comprising the following steps:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
In the present embodiment, in step S2, cleaning operation includes column data editor, cell editor and grouping item editor, is gone back
Including checking respectively for data storage format, length, coded format and reference field.
In the present embodiment, in step S3, the classification results of primary data are obtained using random forest.Concrete operations are as follows:
It makes definitions first to all data of appearance, i.e. N: training primary data number, M: number of features m: specified number of features,
For determining the result of decision (m≤M) of a node on decision tree；For each sorting item, there is the slave instruction put back at random
Practice to concentrate and extract N number of trained primary data, as the training set of the sorting item, the primary data not being extracted into gives a forecast, and assesses it
Error.For the node that each branch occurs, m feature is randomly choosed, each node on decision branch item is by these
Feature determines.It selects to violate the primary data being drawn into each branch, is defined and be classified as oob primary data.It is right
Each primary data, when calculating and analyzing it as oob primary data, the classification situation of sorting item.It is voted and is made with simple majority
For the classification results of the primary data, use accidentally point number account for the ratio of primary data sum as the oob false segmentation rate of random forest.
According to m feature therein, calculates best branch and mode occur.To each primary data, it is calculated as oob primary data
Branch.For the classification situation of primary data, using simple majority ballot as the classification results of the primary data.
Embodiment 3
On the basis of embodiment 1, as shown in Figure 1, the present embodiment provides a kind of, the equipment fault based on machine learning is examined
Survey method, comprising the following steps:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
In the present embodiment, optimizing operation, specific step is as follows:
S301. initial time t=0 randomly chooses initial population P_0；
S302. individual adaptation degree functional value F (P_t) is calculated；
If S303. fitness function value corresponding to optimum individual is sufficiently large in population or algorithm continuous operation mostly generation
And the optimal adaptation degree of individual goes to step S306 without significantly improving；
S304. current time t '=t+1 selects P_t using selection operator method from P_ (t-1)；
S305. P_t is intersected, mutation operation, goes to step S302 later；
S306. optimal kernel functional parameter and penalty factor are provided, and with its training dataset to obtain multiple startings
Point.
Embodiment 4
On the basis of embodiment 1, as shown in Figure 1, the present embodiment provides a kind of, the equipment fault based on machine learning is examined
Survey method, comprising the following steps:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
In the present embodiment, in step S3, using support vector machines (SVM) classifier, according to structural risk minimization
Optimal classification surface is constructed, so that the error in classification to primary data is minimum.Such as: by a Nonlinear Mapping, by it is non-linear can
The data divided are converted into the data of the linear separability in feature space, it is assumed that one group of primary data linear separability, then by looking for
A linear classifier classifies to it out.At this point, the feature vector of input primary data is the vector of a low-dimensional, and
Lower dimensional space linearly inseparable then can use the vector that kernel function is translated into higher-dimension, this mapping can be low-dimensional sky
Between in linear inseparable two classes point be converted to linear separability.
In model construction process, the present invention replaces traditional single model syndrome check method and artificial fault detection side
Method, can be with genetic algorithm tune ginseng instead of traditional parameter adjustment method using the combination of genetic algorithm and AdaBoost Meta algorithm
The accuracy rate for improving algorithm, improves efficiency, and can support that Various Classifiers on Regional is (support vector machines, random using AdaBoost Meta algorithm
Forest, BP neural network) customized addition, carry out the detection of failure, training simultaneously makes the present invention have autonomous troubleshooting hair
Raw ability, without carrying out situation judgement by artificial.
Embodiment 5
On the basis of embodiment 1, as shown in Figure 1, the present embodiment provides a kind of, the equipment fault based on machine learning is examined
Survey method, comprising the following steps:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained
To pre- data；
S3. operation is optimized to pre- data by genetic algorithm, the pre- data operated by analysis are randomly generated multiple
Then starting point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, to be added
Weight average result；
S6. by contrast weight average result and initial results collection, judge whether is equipment fault that the primary data represents
Belong to known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding failure letter of primary data
Breath, and by man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition failure
Database, and emergency prompt is exported by man-machine interface.
In the present embodiment, in step S5, every after a classifier, weighting classification calculated result be will do it accordingly
Weighted value updates.
In the present embodiment, in step S5, is summed, obtained in circle according to the weighted results exported in each classifier
Final output result.
In the present embodiment, the calculation formula of weighted value is as follows:
Hfinal=sign=(∑ ai·Hi),
Wherein, HiFor the basic classification device obtained according to the minimum threshold value of error rate, aiIndicate HiIn final classification device
Significance level.
As mechanical equipment is more and more accurate complicated, equipment fault detection work seems especially difficult, and the present invention can be real
The self-teaching function of existing machine, while establishing the Mishap Database of equipment, can accurately and timely really locking equipment failure with
And solution, the detection work of mechanical equipment can prevent the generation of mechanical breakdown, can reduce machinery maintenance number, show
It lands and improves overhaul efficiency, so that the production efficiency for promoting enterprise is very big, so that Business Economic Benefit gets a promotion, therefore for
The production practices of enterprise are extremely important；After training, the present invention can be to all kinds of machine tooling equipment, production
Equipment has considerable degree of general applicability, can be widely used in the industry of the various various machinery production failures being likely to occur
In production application scene, it is suitable for promoting the use of.
It should be noted that, in this document, such as first and second etc relational terms are used merely to an entity
Or operation is distinguished with another entity or operation, is existed without necessarily requiring or implying between these entities or operation
Any actual relationship or order.Moreover, the terms "include", "comprise" or its any other variant be intended to it is non-
It is exclusive to include, so that the process, method, article or equipment for including a series of elements not only includes those elements,
It but also including other elements that are not explicitly listed, or further include solid by this process, method, article or equipment
Some elements.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including
There is also other identical factors in the process, method, article or equipment of the element.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can store in computer-readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed.
Finally, it should be noted that the above description is only an embodiment of the present invention, it is not intended to limit patent of the invention
Range, it is all using equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content, directly or indirectly
Other related technical areas are used in, are included within the scope of the present invention.
Claims (8)
1. a kind of equipment fault detection method based on machine learning, it is characterised in that: the following steps are included:
S1. the primary data of the equipment of the failure of acquisition is obtained；
S2. cleaning operation is carried out to primary data, and postsearch screening is carried out to the primary data after completion cleaning operation, obtained pre-
Data；
S3. operation is optimized to pre- data by genetic algorithm, multiple startings are randomly generated in the pre- data operated by analysis
Then point retrieves starting point in synchronization, and exports search result；
S4. by AdaBoost Meta algorithm Weighted Searching as a result, and classify to data, obtain initial results collection；
S5. according to the setting weight of sample in each classifier, gradually classify to initial results collection, so that it is flat to obtain weighting
Result；
S6. by contrast weight average result and initial results collection, whether the equipment fault for judging that the primary data represents belongs to
Known fault；
S7. the judging result such as in step S6 is yes, then transfers from Mishap Database and change the corresponding fault message of primary data, and
By man-machine interface export fault message, as the judging result in step S6 be it is no, then by the primary data addition fault data
Library, and emergency prompt is exported by man-machine interface.
2. the equipment fault detection method according to claim 1 based on machine learning, it is characterised in that: the step
In S1, primary data is from the data with existing interface of current device and/or the external sensor of current device.
3. the equipment fault detection method according to claim 1 based on machine learning, it is characterised in that: the step
In S2, it further includes checking respectively for data storage lattice that cleaning operation, which includes column data editor, cell editor and grouping item editor,
Formula, length, coded format and reference field；When postsearch screening, make to keep one to the primary data reference system completed after cleaning
It causes, and guarantees the uniqueness of coding.
4. the equipment fault detection method according to claim 1 based on machine learning, it is characterised in that: the step
In S3, optimizing operation, specific step is as follows:
S301. initial time t=0 randomly chooses initial population P_0；
S302. individual adaptation degree functional value F (P_t) is calculated；
If S303. fitness function value corresponding to optimum individual is sufficiently large in population or algorithm continuous operation mostly generation and a
The optimal adaptation degree of body goes to step S306 without significantly improving；
S304. current time t '=t+1 selects P_t using selection operator method from P_ (t-1)；
S305. P_t is intersected, mutation operation, goes to step S302 later；
S306. optimal kernel functional parameter and penalty factor are provided, and with its training dataset to obtain multiple starting points.
5. the equipment fault detection method according to claim 1 based on machine learning, it is characterised in that: the step
In S3, using support vector machines (SVM) classifier, optimal classification surface is constructed according to structural risk minimization, so as to first
The error in classification of beginning data is minimum.
6. the equipment fault detection method according to claim 1 based on machine learning, it is characterised in that: the step
Every after a classifier in S5, weighting classification calculated result will do it corresponding weighted value and update.
7. the equipment fault detection method according to claim 6 based on machine learning, it is characterised in that: the step
In S5, is summed according to the weighted results exported in each classifier in circle, obtain final output result.
8. the equipment fault detection method according to claim 6 based on machine learning, it is characterised in that: the meter of weighted value
It is as follows to calculate formula:
Hfinal=sign (∑ ai·Hi)
Wherein, HiFor the basic classification device obtained according to the minimum threshold value of error rate, aiIndicate HiIt is important in final classification device
Degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811572223.XA CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811572223.XA CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109635008A true CN109635008A (en) | 2019-04-16 |
| CN109635008B CN109635008B (en) | 2020-11-24 |
Family
ID=66076229
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811572223.XA Active CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109635008B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837898A (en) * | 2019-10-25 | 2020-02-25 | 袁茂银 | Underground cable fault repairing method and device based on model selection |
| CN113240140A (en) * | 2021-06-04 | 2021-08-10 | 海尔数字科技(青岛)有限公司 | Fault detection method, device, equipment and storage medium of physical equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617576A (en) * | 2013-11-21 | 2014-03-05 | 南京大学 | General fault detecting and maintenance method for equipment |
| CN108520080A (en) * | 2018-05-11 | 2018-09-11 | 武汉理工大学 | Automatic system of marine diesel-generator failure predication and health status online evaluation system and method |
| CN108681747A (en) * | 2018-05-11 | 2018-10-19 | 武汉理工大学 | Rotary machinery fault diagnosis based on deep learning and condition monitoring system and method |
-
2018
- 2018-12-21 CN CN201811572223.XA patent/CN109635008B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617576A (en) * | 2013-11-21 | 2014-03-05 | 南京大学 | General fault detecting and maintenance method for equipment |
| CN108520080A (en) * | 2018-05-11 | 2018-09-11 | 武汉理工大学 | Automatic system of marine diesel-generator failure predication and health status online evaluation system and method |
| CN108681747A (en) * | 2018-05-11 | 2018-10-19 | 武汉理工大学 | Rotary machinery fault diagnosis based on deep learning and condition monitoring system and method |
Non-Patent Citations (3)
| Title |
|---|
| RUIHU WANG: "AdaBoost for Feature Selection, Classification and Its Relation with SVM", 《SCIVERSE SCIENCEDIRECT》 * |
| XUEWEN LI 等: "GA-AdaBoostSVM classifier empowered wireless network diagnosis", 《EURASIP JOURNAL ON WIRELESS COMMUNICATIONS AND NETWORKING》 * |
| 李胜 等: "基于GA-PLS和AdaBoost的故障诊断方法", 《数理统计与管理》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837898A (en) * | 2019-10-25 | 2020-02-25 | 袁茂银 | Underground cable fault repairing method and device based on model selection |
| CN113240140A (en) * | 2021-06-04 | 2021-08-10 | 海尔数字科技(青岛)有限公司 | Fault detection method, device, equipment and storage medium of physical equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109635008B (en) | 2020-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111428054B (en) | Construction and storage method of knowledge graph in network space security field | |
| WO2022135265A1 (en) | Failure warning and analysis method for reservoir dispatching rules under effects of climate change | |
| Capozzoli et al. | Fault detection analysis using data mining techniques for a cluster of smart office buildings | |
| Goh et al. | Incorporating the rough sets theory into travel demand analysis | |
| McGovern et al. | Identifying predictive multi-dimensional time series motifs: an application to severe weather prediction | |
| CN108470022B (en) | Intelligent work order quality inspection method based on operation and maintenance management | |
| CN106067094A (en) | A kind of dynamic assessment method and system | |
| CN108985380B (en) | Point switch fault identification method based on cluster integration | |
| CN110866819A (en) | Automatic credit scoring card generation method based on meta-learning | |
| CN109657947A (en) | A kind of method for detecting abnormality towards enterprises ' industry classification | |
| CN112735097A (en) | Regional landslide early warning method and system | |
| CN105426441B (en) | A kind of automatic preprocess method of time series | |
| CN110636066B (en) | Network security threat situation assessment method based on unsupervised generative reasoning | |
| CN109635010B (en) | User characteristic and characteristic factor extraction and query method and system | |
| CN105955951A (en) | Message filtering method and device | |
| CN104881735A (en) | System and method of smart power grid big data mining for supporting smart city operation management | |
| Mao et al. | Anomaly detection for power consumption data based on isolated forest | |
| CN104156403A (en) | Clustering-based big data normal-mode extracting method and system | |
| CN115794803B (en) | Engineering audit problem monitoring method and system based on big data AI technology | |
| Gu et al. | Application of fuzzy decision tree algorithm based on mobile computing in sports fitness member management | |
| CN109635008A (en) | A kind of equipment fault detection method based on machine learning | |
| CN109594967A (en) | A kind of lock of tool drilling detection method for early warning based on well logging big data | |
| El Wakil et al. | Data management for construction processes using fuzzy approach | |
| CN116541755A (en) | Financial behavior pattern analysis and prediction method based on time sequence diagram representation learning | |
| CN116450704A (en) | Automatic generation method and generation device of flow model |
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 |