摘要
结合粗糙集理论和神经网络在信息处理方面的优势,建立了一个基于粗糙集理论和BP神经网络相结合的发动机失火故障诊断模型。通过对EQ6102型发动机的实际试验表明,模型简化了网络训练样本,优化了神经网络结构,提高了系统运行效率。
Low burning quality of engine mixture gases will cause engine power descending, fuel consumption increasing and pollution of exhaust emission aggravation. Analyzing the relationship between change of exhaust emission＇s content and the burning quality, a new fault diagnosis model of misfire in engines based on rough sets theory and neural network technology was presented. The model reduced the sample size, optimized the neural network, decreased the computation and increased the diagnosis correctness. It has been testified by the experimentation of EQ6102 engine and results showed that it could optimize neural network technology and increase the running efficiency of the system.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2005年第8期118-121,共4页
Transactions of the Chinese Society for Agricultural Machinery
基金
高等学校优秀青年教师教学科研奖励计划资助项目(项目编号:02411)
浙江省自然科学基金资助项目(项目编号:301270)
浙江省自然科学基金人才基金资助项目(项目编号:RC02067)
浙江省教育厅资助项目(项目编号:JK2002316)
关键词
发动机
故障诊断
粗糙集
神经网络
Engine, Fault diagnosis, Rough set, Neural network
