Rocket So The Inspection System Fault Diagnosis Expert System Design And Implementation

New generation Carrier Rocket System is a complicated system which requires high reliability. In principle, any fault is not allowed when the rocket is launching. Fault Diagnosis of every subsystem is very important to the reliability of the whole system whether in developing phase or in testing phase. This dissertation focuses on the design and implementation of fault diagnosis expert system for fault detection and management of carrier rocket, to incorporate the technology of fault diagnosis expert system into the whole process of fault diagnosis of rocket's producing and launching.In this paper, a hybrid knowledge representation approach, which makes full use of existing knowledge of fault diagnosis, is presented. It also can make fault diagnosis system continues to accumulate experience and repair cases. Using database to express repository, knowledge management will becoming more convenient. Based on the formation of repository, two reasoning modes-decision based reasoning and case based reasoning are designed to improve the capability of diagnosis and reasoning. This dissertation is mainly composed of several parts as follows.1. Aimed at the present diagnostic knowledge and continually extension of knowledge of fault detection system, this dissertation proposes a hybrid knowledge representation approach. It is proved in practice that this approach to represent knowledge not only can represent diagnostic experience and repair cases with the use of system effectively.2. Integrated inference mechanism based on decision trees and cases is implemented. Cases are triggered by the index which is gotten from knowledge represented by decision trees and diagnostic conclusions derived from the two approaches are output by synthesize. This integrated inference mechanism enhances the accuracy of diagnostic and improves the efficiency of reasoning.3. Present corresponding database representation method for diagnostic knowledge: aimed at knowledge based on decision trees, use the parent node representation adding node state column and fault column to implement the database storage for diagnostic knowledge; while for knowledge based on cases, the fault symptom matrix is available.4. Discuss the generation of test strategy for diagnosis based on information gain maximization and conditional entropy minimization. This approach can be used into the testability analysis and the optimization of test strategy for diagnosis. The logic management of test items and the dynamic schedule of test task are implemented by database technology.