A Hybrid Pattern Recognition Architecture for Cutting Tool Condition Monitoring

One of the important developments in modern manufacturing industry has been the trend towards cost savings through stuff reductions whilst simultaneously improving the product quality. Traditional tool change strategies are based on very conservative estimates of tool life from past tool data and this leads to a higher tool change frequency and higher production costs. Intelligent sensor based manufacturing provides a solution to this problem by coupling various transducers with intelligent data processing techniques to deliver improved information relating to tool condition. This makes optimization and control of the machining process possible. Many researchers have published results in the area of automatic tool condition monitoring. The research work of Scheffer C. etc. showed that proper features for a wear monitoring model could be generated from the cutting force signal, after investigating numerous features. An approach was developed to use feed force measurements to obtain information about tool wear in lathe turning (Balazinski M. etc.). An analytical method was developed for the use of three mutually perpendicular components of the cutting forces and vibration signature measurements (Dimla D. E. etc.). A tool condition monitoring system was then established for cutting tool-state classification (Dimla D. E. etc.). In another study, the input features were derived from measurements of acoustic emission during machining and topography of the machined surfaces (Wilkinson P. Etc.). Li, X etc. showed that the frequency distribution of vibration changes as the tool wears (Li X. etc.). Tool breakage and wear conditions were monitored in real time according to the measured spindle and feed motor currents, respectively (LI X. L. Etc. ). Advanced signal processing techniques and artificial intelligence play a key role in the development of tool condition monitoring systems. Sensor fusion is also found attractive since loss of sensitivity of one of the sensors can be compensated by other sensors. A new on-line fuzzy neural network (FNN) model with four parts was developed (Chungchoo C. etc.). They have the functions of classifying tool wear by using fuzzy logic; normalizing the inputs; using modified least-square back propagation neural network to estimate flank and crater wear. A new approach for online and indirect tool wear estimation in turning using neural networks was developed, using a physical process model describing the influence of cutting conditions on measured process parameters (Sick B.). Two methods using Hidden O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m