In-Situ Temperature Estimation in Rapid Thermal Processing Systems using Extended Kalman Filtering

An Extended Kalman Filter (EKF) is developed to provide an improved pyrometer-based estimate of temperature in rapid thermal processing systems. Models of the heat transfer characteristics, emissivity variations, and temperature measurement equipment, are used in formulating an EKF as a closed-loop observer to filter out noise, lessen the effects of processing variations, and to infer temperature at locations where no measurements are available. Extensive simulations are conducted to analyze the robustness of the EKF.