The State of the Art in Research into the Condition Monitoring of Industrial Machinery

Abstract

Machinery is adopted for a variety of functions, ranging from simple fans to complex ships. One of the fundamental challenges currently faced in a wide range of industries is how to identify faults in machinery before reached a critical level, so as to avoid system degradation, malfunction, and catastrophic failure. The condition monitoring of machines has long been accepted as a most effective solution in avoiding sudden shutdown and for detecting and preventing failures in complex systems. This paper describes the basic concepts of condition monitoring, and introduces the necessary background information about the various condition monitoring technologies used for different types of machines. Also, it discusses the potential benefits through utilizing the artificial intelligence techniques, such as artificial neural networks (ANN), fuzzy logic system (FLS), genetic algorithm (GA), and support vector machine (SVM), in developing robust condition monitoring systems to address issues of fault detection and diagnosis. Finally, rapid developments in electronics technology have opened new aspect in building monitoring systems using embedded devices. Therefore, in the last section of this paper the principles of embedded systems and their application in condition monitoring are further elaborated.