Data-driven model-free adaptive sliding mode control for melt surface temperature of Czochralski silicon monocrystal growth process

Heat is the driving force of solid-liquid phase transformation in Czochralski silicon monocrystal growth process, and melt surface temperature directly affects the quality of silicon monocrystal. Due to the complex physical changes, multi-field and multi-phase coupling, uncertain model and time-varying and nonlinear characteristics of the silicon monocrystal growth process, which makes the mechanism model difficult to obtain. This paper presents a model free adaptive discrete sliding mode control (MFASMC) method for melt surface temperature control. Firstly, the thermal system model between heater power and melt surface temperature is established based on data using ELM. Secondly, design of controller with compact dynamic linearization by I/O data of the system. Meanwhile, the idea of discrete sliding mode is to enhance the tracking performance and robustness of model free adaptive control (MFAC). Finally, The effectiveness of the proposed method is verified by simulation analysis.