Adaptive control of wafer temperature in RTP

This paper presents a wafer temperature control method in RTP using a stable nonparametric adaptive control approach based on a neural network referred to as a piecewise linear approximation network (PLAN). Numerous studies have addressed the RTP temperature control problem and most research on this problem require exact knowledge of the system dynamics. Most approaches do not guarantee the desired performance in practical application when there exist some modeling errors between the model and the actual system or the system dynamics are time-varying. In this paper, a stable nonparametric adaptive control scheme is applied to RTP without an exact information on the dynamics. The system dynamics are assumed to be an affine nonlinear form, i.e., x/spl dot/=f(x)+g(x)u, and the unknown portion of the dynamics are estimated by PLAN. The controller architecture is based on adaptive feedback linearization. The performance of the proposed control approach is validated through experiments on the RTP system developed by Kornic Systems Corporation in Korea.