Temperature Control And Optimization Of Heating Furnace

Metallurgical industry is a large energy consumption of China ’ s industry,and the energy consumption of heating furnace accounts for a quarter of the whole metallurgical industry.In order to achieve the purpose of energy saving and emission reduction,we carried out research on this subject.This thesis studies the temperature control and optimization of heating furnace.Firstly,because the temperature object of the heating furnace has the characteristics of hysteresis,inertia and nonlinearity,the factory adopts manual adjustment to control the temperature of the furnace,but the effect of heating hurnace temperature control directly affects the quality of heating billet.Therefore,the temperature controller is designed in this thesis.By consulting the literature,the following two optimal control methods are proposed and compared,and chooses one to replace manual control.(1)Fuzzy PID control: According to the deviation and change rate of deviation between the measured value and the set value of the furnace temperature,the fuzzy rules are compiled through expert experience and the operation parameters of PID are adjusted by fuzzy reasoning.In the simulink module of MATLAB software,the fuzzy PID control structure is built,simulated,and the stability of the simulation results is analyzed.(2)Generalized predictive control: According to the error between the predicted output value and the actual value of the furnace temperature prediction model,the predicted output value is corrected immediately by feedback correction,and then the corrected predicted output is compared with the set value.The control increment is obtained by optimization calculation,so that the predicted output error is as small as possible to achieve stable control effect.The generalized predictive control is programmed and simulated by MATLAB software,and the simulation results of the two optimal control are analyzed and compared.Secondly,since most of the current temperature control optimization methods only focus on the single optimization of the heating furnace,and ignore the connection between the heating furnace and the roughing mill.Therefore,this thesis also presents the introduction of rough mill production information feed-forward to the furnace temperature control,the establishment of feed-forward compensation model to optimize the furnace temperature setting.The rolling force-billet temperature model is established by BP neural network,which uses rolling force to predict billet temperature.According to the billet temperature,the temperature setting value compensation is feed forward to the heating furnace,which can ameliorative the false high temperature of the heating furnace.With MATLAB neural network toolbox programming simulation,and analyze the effect of this model.The results show that compared with the fuzzy PID control,the generalized predictive control eliminates the overshoot,reduces the adjustment time,and effectively reduces the influence caused by the hysteresis of the heating furnace.The generalized predictive control can replace the manual control.In addition,through the rolling force feedforward compensation model,the set temperature of the heating furnace is reduced,the energy consumption of the heating furnace is reduced,and the comprehensive optimization of the heating furnace temperature is realized.