A knowledge base system for rotary equipment fault detection and diagnosis

This paper studies the fault detection and diagnosis for the most common faults in the rotary equipment. Large amount of experiments are carried out on the machinery fault simulator for simulating different types of rotary machine faults. The study covers from different type of data acquisition sensors, different signal processing and feature extraction techniques. A hierarchical rule-based fault detection system which comprises of a knowledge base coupled with an inference engine is proposed. The knowledge-base that maps the fault mode to signal processing and detection methods is built up. The rule-based fault detection system capable of assisting mechanics and engineers to deal with fault diagnosis of the rotary equipment is presented.