摘要
将支持向量机方法用于某大型液体火箭发动机稳态试车数据的挖掘,建立了多故障分类器,采用23次试车数据对上述挖掘结果进行了测试,将测试结果与人工神经网络方法等所得结果进行了比较。并利用28类仿真稳态故障数据对该方法进行了进一步验证。结果表明,支持向量机方法是一种可基于小样本的、有效的液体火箭发动机故障检测与诊断方法。
A novel method of fault detection and diagnosis of liquid rocket engine based on support vector machine（SVM） is proposed to solve the problem of deficiency of fault data samples. A multi-fault classifier is constructed. Data got from 23 tests are used to testify obtained models. The results were compared with those of ANN method. 28 kinds of simulated data are also used to verify the performance of SVM method. It shows that the SVM method is an efficient approach based on small samples for the Fault detection and diagnosis of liquid-propellant rocket engine.
出处
《导弹与航天运载技术》
北大核心
2007年第4期54-58,共5页
Missiles and Space Vehicles
基金
国家自然科学基金资助项目(50376073)
关键词
液体火箭发动机
故障检测
故障诊断
数据挖掘
支持向量机
Liquid rocket engine
Fault detection
Fault diagnosis
Data mining
Support vector machine
