摘要
针对柴油机故障部位与故障特征之间没有明确对应关系的问题,将信息融合技术引入柴油机故障诊断领域,采用多传感器采集信号、多故障特征提取方法、不同分类器处理结果获得的各种冗余互补信息,使用SVDD方法改进D-S证据理论,并建立两级融合模型进行验证,实现多等级、多层次的诊断。结果表明,该诊断方法正确率高达93%。
Aiming at the problem that there is no clear correspondence between the fault location and the fault characteristics of the diesel engine,the information fusion technology is introduced into the field of diesel engine fault diagnosis.It uses multi-sensor acquisition signals,multi-fault feature extraction methods,and various redundant complementary information obtained by different classifier processing results.The SVDD method is used to improve the D-S evidence theory,and a two-level fusion model is established for verification to achieve multi-level,multi-level diagnosis.The results show that this diagnostic method has a correct rate of up to 93%.
作者
张懿
崔佳
ZHANG Yi;CUI Jia(School of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang 212003,China;Changshu Rhett Electric Co.,Ltd.,Changshu 215500,China)
出处
《重庆理工大学学报（自然科学）》
CAS
北大核心
2019年第11期89-94,共6页
Journal of Chongqing University of Technology：Natural Science
基金
国家自然科学基金项目(51809128)
江苏省省重点研发计划产业前瞻与共性关键技术重点项目(BE2018007)
江苏省研究生科研与实践创新计划资助项目
关键词
柴油机
多传感器
多特征提取
D-S证据理论
两级融合
diesel engine
multi-sensor
multi-feature extraction
D-S evidence theory
two-level fusion
