Modeling And Implementation Of The Box-type Resistance Furnace Control System

The resistance furnaces are very important heating equipments in modern industry and teaching areas. The temperature control process of resistance furnaces are very complicated, it has several characteristics like non-linear, heating unidirectional, large inertia, hysteresis and coupling. So it is very meaningful to research the modeling and control of resistance furnace system. The purpose of this paper is to provide the overall solution of box-type resistance furnace control system from modeling to control and optimization.First of all, according to the input and output characteristics of the system, this paper proposed to establish a second-order model of resistance furnace, which get rid of the traditional one-order model. And it depends on MATLAB/Simulink and curve fitting toolbox in MATLAB. Secondly, with the help of Simulink, we can get the simulation system which use of the newly second-order model. Then MATLAB can find the optimal parameters automatically with ITAE integral error indicators. And the optimal parameters can be transplanted to the actual control system. Finally, for the parameters which from the simulation system cannot always adapt to the actual system, system PID parameter self-tuning was needed. By calculating the controller parameters correction amount, fine-tune the system parameters all the time, which can improve the effect of system controlIn addition, according to the actual system requirements, KingView was used in PC which established communication with the smart meter, so that the PC can monitor the resistance furnace control system, and record the state of it.Experiments show that the second-order system model in this paper can fit the actual measured step response curve better. But the control effects between the simulation system and the actual system are still different. Although the simulation model is able to correspond to the trend of the actual system to a certain extent, but the simulation system still cannot simulate the actual system completely. In order to compensate for this shortcoming, the parameters on-line self-tuning is useful. It tuned the control parameters of actual system, made the control more accurately.