Machinery fault diagnosis based on fuzzy measure and fuzzy integral data fusion techniques

Fuzzy measure and fuzzy integral theory are an outgrowth of classical measure theory. Fuzzy measure and fuzzy integral theory take into account the importance of criteria and interactions among them, and have excellent potential for applications such as classification. This paper presents a novel data fusion approach for machinery fault diagnosis using fuzzy measures and fuzzy integrals. The approach consists of a feature-level data fusion model and a decision-level data fusion model. The fuzzy c-means analysis method was employed to identify the relations between a feature set and a fault prototype to establish mappings between features and given faults. Rolling element bearing and electrical motor experiments were conducted to validate the models. Different features were obtained from recorded signals and then fused at both feature and decision levels using fuzzy measure and fuzzy integral data fusion techniques to produce diagnostic results. The results showed that the proposed approach performs very well for bearing and motor fault diagnosis.