A rotary machinery fault diagnosis approach based on rough set theory

Rotary machinery fault diagnosis can be treated as a pattern classification task. Based on the vibration characteristic spectrum, a rotary machinery fault diagnosis approach using the rough set theory is proposed. Accurate diagnostic results can be obtained directly from a set of complete fault spectrum samples; satisfactory diagnostic results can also be derived from a set of incomplete fault spectrum samples using the approach. The inherent redundancy in the spectrum information is revealed. The approach presents a new idea to rotary machinery fault diagnosis based on incomplete characteristics. Examples show that the proposed approach is very effective.