A new supervisory control system for cutting force in milling operations using neural networks and fuzzy logic

A supervisory control system for cutting force in machining with the proposed new NURBS (Non-uniform Rational B-spline) neural-fuzzy and the neural network approaches is presented in this dissertation. A high precision neural-fuzzy system is developed using the NURBS membership functions. In the supervisory level, the proposed NURBS neural-fuzzy system and the knowledge based system are used in the supervisory control level to optimize the machining operations by providing the cutting force reference. In the process control level, two sets of neural networks have been developed to control the desired average resultant cutting force together with the contour error in multi-dimensional end milling operations. The first set of the neural network is used to specify the feed rate to maintain the desired cutting force. This feed rate is then resolved along the feed axes using a parametric interpolation algorithm so that the desired geometric shape is obtained. The second set of the neural network is used to make corrections to the feed rate components specified by the parametric interpolation algorithm to minimize the contour error caused by the dynamic lag of the closed-loop servo systems controlling the feed drives. With the proposed method, a complete supervisory control system for cutting force in machining can be achieved.