GEOMETRIC ADAPTIVE CONTROL FOR ACCURACY AND STABILITY IN MACHINING CYLINDRICAL WORKPIECE (SELF-TUNING, ROUNDNESS, CHATTER, MULTIPROBE)

A new geometric adaptive control (GAC) system has been developed for improving the accuracy and stability in machining cylindrical workpiece. The system takes into consideration error generation process, machine tool dynamics, and metrology. The new stochastic model constructed in this thesis provides not only insight into the nature of machining processes, but also a means for the systematic design of geometric adaptive controllers. The adaptive control algorithms for the GAC system are derived by using the theory of self-tuning control (STC). The algorithms are simple, robust, and suitable for microprocessor implementation. Whitehouse\u27s multiprobe measurement is modified for use in the control system for the first time. Interesting properties of this measurement in control are analyzed and predicted. The results of simulation show that the GAC system can improve both the accuracy and the stability considerably. Through the theoretical analysis we are able to resolve the control problems imposed by multiprobe measurement, and to achieve near optimum control performance. A comparison with another GAC system which uses FCC technique is made to show that the proposed GAC system does have better performance than other existing GAC systems in precision machining