Servo parameter tuning for a 5-axis machine center based upon GA rules

This paper addresses the use of the genetic algorithm (GA) to tune the serv o controller in a multi-axis CNC machine center. Because the number of para meters in a CNC controller is usually very large and their effects are high ly coupled, tuning these parameters is a task requiring great skill. In thi s paper, the genetic algorithm is applied to tune the servo parameters in a commercial CNC machine controller. The controller is a PC-based open archi tecture system, thus allowing the GA to be implemented on a personal comput er. In this study, a dynamic mutation amplitude process is also introduced to improve the quality of evolution. Experimental results obtained on a Sie mens 840D CNC controller indicate that the proposed method is able to autom atically identify the effect of each parameter and to tune the controller s o as to achieve expert level performance. (C) 2001 Elsevier Science Ltd. Al l rights reserved.