Mechanisechanism Modeling And Control Of The Temperature System Of Three Zone Furnace For Space Crystal Growth

Three-zone programmable furnace for space crystal growth is a new equipment for material experiment in the space. In this paper has done research on the mechanism modeling and control of the three-zone furnace for space crystal growth, and realized the temperature control of the three-zone furnace for space crystal growth. Also, this paper has applied the improved general predictive control algorithm to the simulation with mechanism model, to prepare for the realization of advanced temperature control of furnace for space crystal growth.It is very important to get the exact model of the three-zone furnace for space crystal growth both in design engineering and control engineering. In the design engineering, it is useful to determine the configuration and the component of the crystal furnace; in the control engineering, it is useful to select the model structure to identify the model for control. And also, it can be used to replace the plant while doing research on the control algorithm when the three-zone furnace for space crystal growth is unavailable, thus reducing the research cost. This paper has established the mechanism model by analyzing the heat transfer mechanism, and also analyzed the influence of the configuration and the component to the model.In this paper, the PID with feed-forward controller has been designed, and its parameters were obtained using Hooke-Jeeves algorithm. The feedforward compensation has been used to weaken the influence of the first order power voltage. Through a great deal of simulation, the validity of the controller has been proved, and was applied to the practical control experiment, showing satisfying result. Thus the temperature control task of the three-zone furnace for space crystal growth is achieved.This paper has done research on the decoupling predictive control. By predicting system's output using increment model, the control law with identity feed-forward gain matrix had been derived, which guarantees zero steady error without online adjustment. Meanwhile, this paper introduces gain compensation strategy into the predictive control and then analyzed its advantage. Previous research has pointed that, introducing mismatch filter to general predictive control(GPC) based on internal model control structure can improve the robust of GPC. Then on the basis of this viewpoint, the paper proposes a new method to design mismatch filter and test its validity through simulation. This paper uses mechanism model as the plant, introduces mismatch filter, and designs decoupling GPC with gain compensator. The simulation results show that, the proposed method in this paper works very well, thus making our research well prepared for the further application of advance control to the furnace for space crystal growth.