基于聚类SVM瓦斯传感器故障预测研究

Based on clustering support vector machine of gas sensor fault prediction

基于支持向量机的SVM(Support Vector Machine)的基本原理,讨论了一种基于聚类的支持向量机CSVM(clustering support vector machine)的学习算法,并应用于建立瓦斯传感器的动态预测模型,介绍传瓦斯感器故障预测的原理,阐述了基于该算法的瓦斯传感器故障预测的实现过程。仿真结果表明:与标准支持向量机算法相比,该算法可以大大减少支持向量个数,提高瓦斯传感器故障预测精度,为瓦斯传感器故障预测提供了一种新方法。

Based on Support Vector Machine（SVM） the basic principles of,Discussed which is based on clustering support vector machine learning algorithm,Which is used to build Gas sensors dynamic forecasting mode Introduction of sensor failure prediction theory and described gas sensors based on the algorithm implementation process failure prediction,Simulation results show that,With the standard support vector machine algorithm,the algorithm can greatly reduce the number of support vectors to improve the gas sensor failure prediction accuracy,For the gas sensor fault prediction provides a new approach.