Fault prediction method based on synthetic minority class oversampling and deep learning

A fault prediction and deep learning technology, applied in neural learning methods, character and pattern recognition, instruments, etc., can solve problems such as the inability to overcome the problem of unbalanced data set data distribution

Inactive Publication Date: 2019-06-11
BEIJING AEROSPACE MEASUREMENT & CONTROL TECH
View PDF5 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, SMOTE has a certain blindness when selecting neighbors, and this method cannot overcome the

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Fault prediction method based on synthetic minority class oversampling and deep learning
  • Fault prediction method based on synthetic minority class oversampling and deep learning
  • Fault prediction method based on synthetic minority class oversampling and deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] In order to solve the problems in the existing technology, such as the lack of a solution to the inability to carry out deep learning predictive analysis in the case of few fault samples, the blindness of neighbor selection in the traditional SMOTE method, and the reduction of the marginalization of the distribution of unbalanced data sets, the electromechanical equipment is in operation. When an abnormality occurs, it is impossible to effectively predict the problematic equipment or components. This embodiment provides a fault prediction method based on synthetic minority class oversampling and deep learning. see figure 1 , which is a flowchart of a K-Means-SMOTE improved resampling method for unbalanced data classification of electromechanical equipment provided in this embodiment. This method first uses the K-Means method to cluster the minority class samples in the sample set, and deletes the noise sample class whose centroid of each cluster is closest to the majori...

Embodiment 2

[0085] In the following, the above process will be described in detail in combination with specific examples.

[0086] Specifically, taking a certain electromechanical equipment failure ("pump speed output failure prediction") as an example, the failure prediction based on deep learning is carried out.

[0087] 1) Preprocess the original data of the fault, clean the data to remove invalid data and duplicate data, import the processed data, and divide them into majority class samples and minority class samples, and establish parameters in normal state samples; use the K-Means method Clustering minority class samples;

[0088] 2) Find out the noise clusters, and remove new noise clusters in the minority class sample set P;

[0089] 3) Reclassify each remaining cluster of the minority class samples, and delete the noise class samples in each remaining class cluster of the minority class;

[0090] 4) Synthesize new samples and merge data, the merged data is as follows image 3 ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a fault prediction method based on synthetic minority class oversampling and deep learning. The Means method is used for clustering a few types of samples in the sample set; deleting the noise class cluster after clustering; dividing the class cluster into noise class samples in each class cluster by using a KNN method; fault samples and risk samples, deleting the noise samples; and finally, inputting a random number into each class cluster, and selecting a certain sample as an output sample according to a proportional relation between the random number and the fault class sample and the risk class sample in the class cluster;realizing oversampling of the SMOTE method ; and then increasing the number of a few types of samples through multiplication operation, so thatthe types of the samples in the finally obtained fusion sample are more balanced, and the acquired feature data are balanced, thereby facilitating model training, maximally mining the law behind thedata, and realizing a better fault prediction effect.

Description

technical field [0001] The invention belongs to the field of fault prediction and detection, in particular to a fault prediction method based on synthetic minority oversampling and deep learning. Background technique [0002] Complex equipment such as aircraft and rail transit equipment operate in various environments for a long time. Due to the influence of multiple factors such as complex environments, working conditions, and loads, their functions and performance may undergo abnormal changes, and these abnormal changes can usually be reflected by fault prediction methods come out. However, in practical applications, the obtained original data objects are often unbalanced, that is, the number of samples of a certain category is much larger than that of other categories. When the fault data is uneven, it is difficult to effectively predict it using traditional data-driven methods. [0003] In unbalanced data, the class with a relatively large number is generally called th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/04G06N3/08
Inventor 房红征任帅杨浩熊毅隋景峰余家豪罗凯樊焕贞王菲
Owner BEIJING AEROSPACE MEASUREMENT & CONTROL TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap