Knowledge Base Construction Of Auto Braking Expert System & Its Reasoning Mechanism Research

How can we reduce the burden of calculation in auto braking system efficiently,shorten the development periods of braking products, improve the design efficiencyof braking system, and obtain the same design level as experts? It is an excellent ideato use Automotive Braking Expert System (BES) to calculate performance anddesign of braking system. The basic concept is illuminated, and the KnowledgeExpression Method (KEM) for the conventional expert system is analyzed in thepaper firstly. On the basis of analyzing all kinds of KEM, the Object-Oriented theoryis chosen as the design project to construct knowledge base of expert system.The Object-Oriented theory was popularized in the 1980's, and become oneimportant analysis idea and programming method. The method of Object-Orienteddescribes things being studied doubly, namely static description and dynamicdescription of object. The former expresses the object's classified attribute, the latterexpresses the object's action attribute, and both of them are interactional and alsointerdependent. The static description expresses the object's state, and the dynamicdescription is divided into two parts: one is the operation to modify the class's ownattribute state, and the other is the operation to output the result. Therefore, the"object" is the entity which consists of data and the code to operate the data, and hasspecial attribute and action mode.The fundamentals concept is introduced, and the Object-Oriented theory forKEM to resolve complicated object is analyzed in the paper. And the 23 items in theautomotive braking system are abstracted into different classes and saved in theknowledge base. The development tool of Visual C++ to construct knowledge baseof BES support KEM of Object-Oriented theory well, and can divide the constructionand maintenance of knowledge base, reasoning process from the development ofHuman-Computer Interface (HCI) efficiently. Therefore the programmingdevelopment efficiency is improved. Furthermore,the reasoning mode of expertsystem is discussed, and the reasoning mode and search strategy in common use areintroduced in the paper. And the reasoning mode and search strategy used in BES areexplored elementarily which provide a good basis for construction of effectivereasoning mechanism.At last, the BES is tested by some samples, and the paper gives summarizationsand expectations, which afford the excellent referenced value to continue making theBES more applicable and perfect.