Research On Model Parameter Calibration And Temperature Optimization Method Of Reheating Furnace

As one of the main energy consumption equipment in the hot rolling production line of iron and steel enterprise,the main function of reheating furnace is to heat slabs to the temperature that can comply with rolling mill's requirements to slabs.The complex industrial process of slab heating is large delay,strong coupling,nonlinear,multi-variable,and slab temperature detection inside the furnace is a kind of typical industrial process quality parameters are difficult to measure.Therefore,the effective ways to improve the heating quality of billets and save the energy is taking advanced modeling method and control technique,building the billet temperature forecast model to implement the slab temperature soft measurement and get the optimal furnace temperature settings from model.The reheating furnace is the background of this thesis.First of all,based on the heating furnace heat transfer mechanism of steel billet temperature forecast mathematical model is established.In view of the difference between slab upper surface and lower surface such as gas temperature and pressure,this thesis ameliorates the model based on the segmented total furnace heat exchange factor model,and build upside and downside segmented total furnace heat exchange factor model,enhance the rationality of model.Total furnace heat exchange factor is significant for the accuracy of model and this thesis take particle swarm optimization(PSO)algorithm to search and get it,because of the complexity of theoretical calculation.Because of the time varying of total furnace heat exchange factor,it will change when the furnace thermal parameters change.This thesis adopts the fuzzy reasoning approach to compensate total furnace heat exchange factor online to improve self-adaptive capability of model,also make model calculation more accurate.At the same time,the thesis compares and analyses two compensation value distribution methods,which are distribution ration based on process monitoring method and terminal correction.At last,the thesis further discusses furnace temperature optimization setting method on the base of slab temperature forecast model.This article employs iterative learning control method to calculate furnace temperature optimal settings.Because of the thermal disturbance in slab heating,the furnace temperature settings should be compensated.There are lots of slabs heating in furnace.Each slab has a group of optimal furnace temperature settings.In order to make every slab heat to follow with ideal heating curve,furnace temperature settings should be comprehensive compensated.Simulation results indicate that compared with the ameliorated slab temperature prediction model based on segmented total heat exchange factor,the model built in this thesis can calculate the slab temperature quite accurately.The average absolute errors of slab temperature predicted by the modified model are within the 6? limits,which can meet the technological requirements.The optimal furnace temperature settings in this thesis can improve the slab heating quality,and be helpful for high quality production.