Abstract: In view of the characteristic of the temperature control system, which is subsystem of the lithography tools, with big inertia, multiple time-delay and multiple perturbation, a dynamical mathematical model of the temperature control system is presented by using theory modeling and system identification technology. High dimensional model is decoupled by using large-scale system decoupling theory, and the model parameters are estimated by using system identification method. This model is used in the temperature control system of lithography tools. The experiment shows that the model is strong robust and self-adaptive. The temperature of the lithography tools can be achieved very high temperature stability within ±0.006 ℃ by using this model.

Key words: Gray identification, System modeling, Temperature control, Health evaluation, Performance degradation, Principle Components Analysis, Random Forest, Tool wear