A hybrid expert system combined with a mathematical model for blast furnace operation.

For the establishment of completely automatic and most intelligent control system, a hybrid expert system composed of both a mathematical model (the TS model) and experiential rules was developed as the application of artificial intelligence.The selection of control method either by the TS model or by experiential rules is appropriately performed by furnace state judgment rules which are actually experiential rules.In the normal furnace state, the control by the TS model is performed to minimize hot metal temperature for the reduction of energy cost. In the abnormal furnace state, the control by experiential rules is performed to achieve stable furnace operation for the prevention of significant reduction of hot metal production.Further, diagnosis rules are provided in the system to determine the cause of abnormality and to select necessary actions for the long term control. The necessary actions indicated by diagnosis rules are performed manually through the final decision of operators or staffs.The system was introduced into Kashima No. 1 blast furnace of Sumitomo Metal Industries, Ltd. in October of 1988. The hybrid expert system is successfully applied in actual short term control. The establishment of the complete diagnosis rule for long term control is one of the main future tasks of the system.