FUZZY CLUSTERING FOR AUTOMATED TOOL CONDITION MONITORING IN MACHINING

A strategy for unsupervised classification for automated tool condition monitoring by using fuzzy neural networks is proposed. This approach is based on a newly developed classification algorithm by the authors, namely the multiple principal component fuzzy neural network for tool condition monitoring. This algorithm uses three major components of soft computation; fuzzy logic, neural network, and probability reasoning. The proposed network functioned successfully in clustering the data obtained from cutting tests performed within a reasonable range of cutting conditions. Experiments in turning were implemented to test the performance of the proposed algorithm. Several sensors were used for monitoring feature selection. The results showed approximately 80 to 94% success rates in these tests.