AI and ML Techniques for Machine Health Condition Monitoring

Wilson Wang (Wilson.Wang@Lakeheadu.ca) received his Ph.D. in Mechatronics Engineering from the University of Waterloo (Waterloo, Ontario, Canada) in 2002. From 2002 to 2004, he was employed as a senior scientist at Mechworks Systems Inc. He joined Lakehead University in 2004, and now he is a professor and director in Mechatronics Engineering. His research interests include mechatronics, smart sensors, artificial intelligence, machine learning, diagnostics and prognostics of engineering systems, intelligent control, and signal processing. He is a Lakehead Research Chair. He has serves to several journal boards (e.g., IEEE Transaction on Fuzzy Systems, IEEE/ASME Transactions on Mechatronics, IEEE Transactions on Instrumentation and Measurement), and organizations, for example, the Chair of Expert Committee for CFI (Canada Foundation for Innovation). Reference: Reliable machine condition monitoring systems are critically needed in industries to recognize equipment defects at their earliest stage so as to improve production quality, operation efficiency and safety. An AI and machine-learning (ML)-based monitoring system consists of modules such as data acquisition, signal processing, fault diagnosis and prognosis. Smart sensor-based data acquisition systems are used to collect signals wirelessly. Signal processing is a procedure to extract representative features from measurement for system analysis and fault detection in machinery systems. Diagnosis is a procedure to classify features/patterns into different categories corresponding to different equipment health states. Prognosis is a process to predict the remaining useful life of the damaged equipment to schedule predictive maintenance operations. New AI tools such as adaptive neurofuzzy and evolving fuzzy techniques are used in automatic diagnostic and prognostic operations. Appropriate ML algorithms can be used to improve decision-making convergence and adaptive capability to accommodate different machinery conditions.