Research And Design Of Fault Diagnosis Expert System For Vehicle

Expert System of Fault Diagnosis is a kind of software which is designed for improving the quality and the efficiency of vehicle repairing. According to the analysis and study on the structure of the vehicle and a variety of intelligent diagnostic technology, this essay proposed a new fault diagnosis expert system which combines heuristic knowledge and logical knowledge together, while building a knowledge base with relational database technique, finally to realize inference engine by using the abundant first floor data control technique of C#.The paper focuses on the following aspects.1. Build a knowledge base which based on the combination of heuristic and logical knowledge through analyzing the knowledge about the vehicle fault diagnosis. In detail, the logical knowledge realized by extending fault tree showing method means the structure and the principles. Heuristic knowledge prefers more on the expert's experience, adopting rule group(rule skeleton + rule body) to implement the knowledge representation.2. This paper mainly studies how to obtain the rule group. In the paper, to reduce the rule outline's pre-factor, the rough set attribute reduction algorithm was used, and finally can simplify the input. At the same time, to refine the rule body, we use an approach which is based on further learning of neural networks'structure.3. This paper proposed a new hybrid reasoning model which is based on the heuristic and logical knowledge, it request different knowledge models while using different reasoning methods...