Fault Diagnosis using Fuzzy Deep Knowledge Model and Fuzzy Backward Reasoning
Fuzzy fault diagnosis techniques can be classified into two categories: shallow knowledge based and deep knowledge based. The former employs fuzzy sets that represent symptoms of faults and pattern recognition in which the observed systems are compared with the fuzzy sets. Most fuzzy fault diagnosis approaches belong to this category. The latter relies on knowledge about the normal operations of a system and about how faults affect the system’s operations, known as deep knowledge. The deep knowledge based fault diagnosis has the advantage of capture of faults even when symptoms are not obvious. This paper presents a fault diagnosis approach that uses a fuzzy deep model (FDM) and fuzzy backward reasoning (FBR). The FDM simulates both normal and faulty operation conditions of a system. FBR is performed to identify the faults. This approach was tested on a liquid tank system.
fault diagnosis fuzzy systems
Dayou Li Daming Jiang Yueqiao Li
Department of Computer Science and Technologies, University of Bedfordshire, Park Square, Luton, LU1 School of Electronic and Information, Beijing Jiaotong University, 3 Shangyuancun, Xizhimenwai Stree
国际会议
The International Conference on Electrical Engineering 2009(2009 电机工程国际会议)
沈阳
英文
1-5
2009-07-05（万方平台首次上网日期，不代表论文的发表时间）