Abstract
Metal cutting operations constitute a majority of all manufacturing activities. The detection of tool wear is rudiment to the smooth functioning of the metal cutting operation. In this thesis, the complex yet essential task of deriving a model for the detection of tool wear is done using a neuro-fuzzy system. The neural network captures the steady state relationship between the condition of the tool and sensor values, in the weights of the connections between neurons. The fuzzy rules and the entire mechanism are used to provide a linguistic model for the detection of tool wear. However the fuzzy membership functions need to be tuned so that they reflect the true meaning of the process variables. This is done by using an error-based, density-driven adaptation scheme. The successful prediction of tool wear in a milling process by using the neuro-fuzzy system with the intrinsic adaptation scheme is demonstrated.
Mesina, Omez Samoon (1993). A neuro-fuzzy system for tool condition monitoring in metal cutting. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1993 -THESIS -M578.