Dual-driven model for predicting micro differential pressure at converter mouth
The micro-differential pressure control at the converter mouth is of great significance to improving the quality of gas recovery.In order to reasonably regulate the furnace mouth pressure,reduce the amount of air inhala-tion,and increase the calorific value of the recovered gas,a converter gas prediction model was established based on chemical reactions in the converter,historical oxygen blowing data,and neural network algorithms.The accuracy of the model can reach 96.3％.Based on the relationship between the micro-differential pressure and the fan speed,annular seam opening,hood height,hood opening and converter gas volume,a converter mouth micro-differential pressure balance model was established,and coupled with the converter gas prediction model to form a micro-differential pressure prediction dual drive model.The converter gas prediction model provides the predicted con-verter gas volume,and the micro-differential pressure balance model provides the micro-differential pressure balance relationship.The two are combined with the set fan speed,ring gap opening,and hood height to form a micro-differential pressure prediction dual-drive model.When the predicted differential pressure exceeds the set value,the fan speed,ring gap opening,hood opening,hood height and other influencing factors are adjusted to control the dif-ferential pressure within a reasonable range.Through the analysis of actual production data,the relative error between the prediction and control results of the dual-drive model and the set average value of micro-differential pres-sure is 2.7％.The average converter mouth pressure during blowing is reduced from 13.42 Pa to 4.88 Pa.The average suction per furnace The air volume is reduced from the original 40.26 km3 to 14.64 km3.The calorific value(standard state)of coal gas increased from 7 422 kJ/m3 to 8 812 kJ/m3,an increase of 18.73％,indicating that the model can accurately predict and regulate the micro-differential pressure,improve the shortcomings of the time lag of previous prediction methods,and provide a way to improve converter gas calorific value provides an effective method.
micro differential pressureconverter gasprediction and regulationdual-driven modelconverter mouth
NETL
NSTL
万方数据
