Knowledge-based automatic fault detection for dynamic physical systems

Detecting fault before it deteriorates the system performance is crucial for the reliability and safety of many engineering systems. This paper develops an intelligent technique based on fuzzy-genetic algorithm (FGA) for automatically detecting faults on dynamic physical systems. Many researchers have proposed only using fuzzy systems to effect fault detection and diagnosis. Other applications of the FGA are mainly focused on the synthesis of fuzzy control rules. The proposed automatic fault detection system (AFD) monitors the dynamic system states continuously by fuzzy system. The optimization capability of genetic algorithms allows the generation of optimal fuzzy rules with minimum workload. Different system behaviors can be classified by the FGA-AFD system after tuning its rule table. Experiments on a laboratory scale servo-tank liquid process rig are conducted to appraise the performance of the proposed FGA-AFD system. Key-Words: Fault detection, Fuzzy logic, Genetic algorithms