Name: Dr. Wilson Wang
Department: Mechanical Engineering
Title of Lakehead University Research Chair: Lakehead University Research Chair in Intelligent Systems
Start date: March 1, 2019
Website address of researcher: http://flash.lakeheadu.ca/~wwang3
Keywords Describing Chair's Areas of Research: Artificial intelligence; Machine learning, Signal Processing; Fault diagnostics; System state prognostics; Intelligent control, Mechatronics, Smart sensors
Research relevance (importance of the research and how it will benefit Canadians): The objective of Dr. Wang's research is to develop new technologies and intelligent tools for real-time machinery condition diagnostics and prognostics. The purposes are to improve operation accuracy and reliability, but reduce costs by reducing machinery downtimes.
Description of the project
Title: Intelligent Diagnostics and Prognostics of Electric Machines
One of the fundamental problems facing a wide range of industries is how to identify a machinery fault before it reaches critical states in order to avoid machinery performance degradation, malfunction, or even catastrophic failure.
The current strategy in dealing with these problems is called preventive maintenance, which is to periodically shut down machines for manual inspection. If no defect can be detected (or the equipment is healthy), this downtime will add significant costs to the operation of the machine. The goal of Dr. Wang’s research program is to develop new technologies and intelligent tools for real-time machinery condition monitoring. It is a multidisciplinary research area including mechanical, electrical and computer engineering.
The intelligent monitor will consist of the following modules. 1) Data Acquisition: Smart sensor networks will be developed and used to collect signals wirelessly, in different forms such as vibration, electric current and sound. 2) Signal Processing: new signal processing techniques will be proposed to recognize representative features from the collected signals to detect defects in machinery systems. 3) Intelligent Diagnostics: intelligent tools will be proposed to integrate the features generated by the related signal processing techniques for automatic fault diagnostics. 4) Prognostics: knowledge-based predictors will be developed to estimate the future states of the faulty machinery system, and estimate the remaining useful life for repair and maintenance operations without periodically shutting down machines for the preventive maintenance operations.
In addition, machine learning methods will be proposed to improve the adaptive capability of the intelligent monitor to accommodate different machinery and operating conditions.