作者: Kok Yeng Chen , Chee Peng Lim , Weng Kin Lai
DOI: 10.1007/S10845-005-4371-1
关键词:
摘要: In this paper, the fuzzy min–max (FMM) neural network is integrated with a rule extraction algorithm, and resulting applied to real-world fault detection diagnosis task in complex industrial processes. With capability, FMM able overcome “black-box” phenomenon by justifying its predictions using if–then rules that are comprehensible domain users. To assess effectiveness of network, real sensor measurements collected used for detecting diagnosing heat transfer tube blockage conditions circulating water (CW) system power generation plant. The parameters systematically varied tested, results explained. Bootstrapping employed quantify stability performance statistically. extracted found be compatible information as well opinions experts who involved maintenance CW system. Implications facility an intelligent useful tool discussed.