A fuzzy decision system based on statistical learning for fault classifications

A fuzzy decision system (FDS) is proposed for condition monitoring of machining processes. The membership functions are established through a learning process based on test data, rather than being selected as a priori. The optimal partition and information gain weighting functions are also introduced in order to improve the robustness and reliability of this method. Experiment verification with an optimistic success rate of 97.5% was achieved.<
>