Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Cost-sensitive support vector machine locomotive wheel detecting system and detecting method thereof

A support vector machine and cost-sensitive technology, applied in the field of cost-sensitive support vector machine model locomotive wheel detection system, can solve the problem of data sample misclassification cost, inequality, etc.

Active Publication Date: 2017-03-08
HUNAN UNIV OF TECH
View PDF6 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the problems of misclassification of data samples and unequal costs existing in the traditional cost-sensitive support vector machine, and then considering that the adaptive mutation particle swarm optimization algorithm is a group-based intelligent optimization algorithm with strong robustness, the population Advantages such as diversity and global search characteristics

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
  • Cost-sensitive support vector machine locomotive wheel detecting system and detecting method thereof
  • Cost-sensitive support vector machine locomotive wheel detecting system and detecting method thereof
  • Cost-sensitive support vector machine locomotive wheel detecting system and detecting method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0085] Such as figure 1 As shown, a cost-sensitive support vector locomotive wheel state detection system includes a data preprocessing module 1, a cost-sensitive support vector machine training module 2, a parameter optimization module 3, an optimal cost-sensitive support vector machine classification module 4, and a discrimination Module 5 and wheel state output module 6;

[0086] The output terminal of the data preprocessing module 1 is connected with the input terminal of the training module 2, the output terminal of the cost-sensitive support vector machine training module 2 is connected with the input terminal of the parameter optimization module 3, and the output terminal of the parameter optimization module 3 is connected with the optimal The output end of the cost-sensitive support vector machine classification module 4 is connected, the output end of the optimal cost-sensitive support vector machine classification module 4 is connected to the input end of the discrimina...

Embodiment 2

[0097] Data imbalance refers to the large difference in the number of samples of the two categories involved in the classification, which will cause the skew of the classification hyperplane.

[0098] Figure 4 The dots in the middle circle represent the positive category, and the square dots represent the negative category. H, H1, and H2 are classification surfaces calculated based on a given sample set. Since there are very few negative class samples, some sample points that are originally negative classes are not provided, such as Figure 4 For the two square points on H4, if these two points are provided, the corresponding classification planes should be H1, H3 and H4, which is obviously very different from the previous results. Now because of the skew phenomenon, a large number of positive classes can "push" the classification toward the negative class, which affects the accuracy of the results. When the standard support vector machine is used to solve the problem of unbalan...

Embodiment 3

[0162] In order to verify the effectiveness of the method, the present invention selects a heavy-duty locomotive wheel state data set to conduct experiments to demonstrate the effectiveness and reliability of the method. In the experiment, the kernel function needs to be selected first to confirm the effectiveness of the kernel function in dealing with the binary classification problem of unbalanced data, so as to determine that the adaptive mutation particle swarm algorithm can better optimize the parameters of the cost-sensitive support vector machine classification model .

[0163] 1. Choice of kernel function

[0164] At present, in engineering practice, there are four main types of kernel functions commonly used in SVM:

[0165] 1) Linear kernel function

[0166] K(x,x i )=x T x i ;

[0167] 2) Polynomial kernel function:

[0168] K(x,x i )=(δx T x i +r) d ,δ>0;

[0169] 3) Gaussian radial basis kernel function:

[0170] K(x,x i )=exp(-||x-x i || 2 / (2δ 2 )),δ>0;

[0171] 4) Two-laye...

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 discloses a cost-sensitive support vector machine locomotive wheel detecting system and a detecting method thereof. The system comprises a data preprocessing module, a cost-sensitive support vector machine training module, a parameter optimizing module, an optimal cost-sensitive support vector machine classifying module, a determining module and a wheel state output module. The detecting method comprises eight steps totally. The parameter optimizing step is an adaptive mutation particle swarm optimization algorithm and has advantages of high robustness, global searching characteristic, etc. A space which continuously reduces in iteration is enlarged, and searching is performed in a larger space. Furthermore high population diversity is kept. Additionally, a possibility of finding out an optimal value by the algorithm is improved.

Description

Technical field [0001] The invention belongs to the technical field of measurement engineering, and more specifically, relates to a cost-sensitive support vector machine model locomotive wheel detection system and method. Background technique [0002] Support vector machine is a practical method developed in statistical theory. It is a theory that specializes in the study of machine learning rules in the case of small samples. Based on the principle of structural risk minimization, it occupies a very important position in the field of pattern recognition and machine learning. Because SVM can effectively overcome the shortcomings of other machine learning methods such as over-learning, under-learning, low generalization ability, and local minima, now SVM has been widely used in state recognition, fault detection and other fields. The prior art proposes a high-speed train wheel idling prediction method based on a support vector machine, but this method manually modifies the support...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G01M17/013G06K9/62
CPCG01M17/013G06F18/2411
Inventor 何静刘林凡张昌凡谭海湖赵凯辉孙健豆兵兵刘光伟
Owner HUNAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com