An optimal gain adaptive control system for CNC metal cutting machine tools

Abstract

The optimal gain adaptive control system described in this dissertation is a nonlinear, sampled data system which is designed to regulate the cutting force during turning operations. The controller is designed to keep the system stable under large variations in process parameters to limit the force overshoots and provide fast transient response. The stability analysis of the system is performed via Liapunov's indirect method and it was optimized by minimizing the Integral Squared Error index of performance. As a result, a data base for optimal gains as functions of process parameters was generated. Since in most cases the depth-of-cut might change, the use of this data base is limited. To overcome this problem an on-line adaption scheme of the gains was designed to achieve optimal response without the information of process parameters. Using this control scheme a series of computer simulations were performed which demonstrate excellent response under wide variations of the process parameters. Based on the optimal gain scheme, the adaptive control system was implemented on a CNC lathe. Experimental results have shown similar responses to those obtained by computer simulation. In addition to increasing the productivity, the optimal gain system was applied to contouring systems to improve contouring accuracy.