The Research And Implementation Of Knowledge Base In The Blast Furnace Expert System

Well organized knowledge base is very important for the blast furnace expert system. We use knowledge base to storage many kinds of rule knowledge, causality, advices given by the domain experts and operators beside the blast furnace, and so on. And we adopt the technology related to knowledge base to provide data support and academic basis. This paper discussed the research and design of knowledge base which use oriented object method. According to the needs of the blast furnace inference, we put forward to use feature extraction method to extract the parameters of every transducer and regard the values as the fact knowledge in the knowledge base including the variation, fluctuant value, horizontal value and characteristic value. The advices which provided by the domain experts and operators were abstracted as rules. And the rules were looked as the foundation of judging the phenomenon which the blast furnace system would happen. The system analyzes and compares rule-implication method, graphic method and rule-comparison method, and it verifies and checks up the consistency and redundancy which the knowledge base may exist. Because the system is designed as a very important component of intelligent decision-making, so it advances software frame of the model and gives novel idea for operating decision in blast furnace material. At the same time, the whole system is modeled by the UML over again based on Object-Oriented idea. It especially improves the expansibility and maintainability of the system.