METODO Y APARATO PARA CONDUCCION DEL FUNCIONAMIENTO DE UN ALTO HORNO

LA INFERENCIA SE EFECTUA MEDIANTE HIPOTESIS INTERMEDIAS QUE REPRESENTAN ESTADOS FISICOS DE UN ALTO HORNO QUE UTILIZA HG (GRADO HEURISTICO) PARA DIAGNOSTICAR DE MANERA TOTAL EL ESTADO DEL HORNO PARA FIJAR LAS ACCIONES OPTIMAS. SE CONTROLAN PARAMETROS ESPECIFICOS PARA RECONOCER INMEDIATAMENTE UNA TRANSICION DE CONDICIONES DENTRO DEL HORNO PARA EJECUTAR ADICIONALMENTE LA INFERENCIA. EN ESA INFERENCIA SE CUBREN VARIOS TIPOS DE ACCIONES COMO ACCIONES DEFENSIVAS Y ACCIONES OFENSIVAS. SE UTILIZA UN MODELO DE ESTIMACION DE DISTRIBUCION DE CARGA CONSIDERANDO EL COLAPSO DE UN LECHO DE COQUE PARA AYUDAR EN LA DECISION DE UNA ACCION OPTIMA CUANDO ES NECESARIA UNA ACCION PARA CAMBIAR LA DISTRIBUCION EN EL HORNO COMO RESULTADO DE LA INFERENCIA. LA CREACION Y ALTERACION DE UNA BASE DE CONOCIMIENTOS PARA LA INFERENCIA SE EJECUTAN SIN INTERRUMPIR LA INFERENCIA. A method for management of an operation of a blast furnace comprising the steps of: (a) preparing a data base of information related to said furnace and a knowledge base including rules for diagnosing the state of said furnace; (b) gathering said information in a first interval; (c) adding said information to the data base; (d) inferring the state of the furnace using the data base and knowledge base in a second interval longer than said first interval; and characterised by (e) forecasting distribution in the furnace under various control condition combinations to aid deciding the optimum action comprising (i) preparing said combinations by inputting present conditions and variously altering at least one of them, (ii) calculating the distribution using a burden distribution estimation model considering collapse of a coke bed under said combinations; and (iii) outputting the calculation results. An apparatus for said blast furnace operation is further claimed. The method pref. also comprises (a) defining a number of physical state intermediate hypotheses and final diagnoses; (b) deciding first causative relations between the information and the hypotheses and second causative relations between said hypotheses and said diagnoses according to heuristic knowledge; (c) establishing a condition and a wt. (W) in each group of related information and a threshold in a related hypothesis regarding each first causative relation; (d) establishing a wt. (Y) in each hypothesis and a threshold (Z) in a related diagnosis regarding each second causative relation; and (e) storing rules including the causative relations, the condition