Application of Probabilistic Causal-effect Model based Artificial Fish-Swarm Algorithm for Fault Diagnosis in Mine Hoist

This paper presents an intelligent methodology for diagnosing incipient faults in mine hoist. As Probabilistic Causal-effect Model-Based diagnosis is an active branch of Artificial Intelligent, in this paper, the feasibility of using probabilistic causal-effect model is studied and it is applied in artificial fish-swarm algorithm (AFSA) to classify the faults of mine hoist. In probabilistic causal-effect model, we employed probability function to nonlinearly map the data into a feature space, and with it, fault diagnosis is simplified into optimization problem from the original complex feature set. And an improved distance evaluation technique is proposed to identify different abnormal cases. The proposed approach is applied to fault diagnosis of friction hoist with many steel ropes, and testing results show that the proposed approach can reliably recognise different fault categories. Moreover, the effectiveness of the method of mapping hitting sets problem to 0/1 integer programming problem is also demonstrated by the testing results. It can get 95% to 100% minimal diagnosis with cardinal number of fault symptom sets greater than 20.