Hybrid software architecture of knowledge-based signal inspection systems

Abstract

This thesis deals with a software architecture of the intelligent signal inspection systems and the design methodology for building them. To combine multiple knowledge sources in signal inspection domain, a hybrid knowledge representation and processing architecture, so called ISIS(Intelligent Signal Inspection System), is proposed as a solution to problems encountered in the development of knowledge-based signal inspection system. To modularize the process of knowledge elicitation and to make knowledge base maintainable, the task of inspection is delegated to two subsystems each of which has a proper knowledge processing scheme to match the properties of its own task. The front-end subsystem which detects the signal patterns(events) for any harmful flaw is built by integrating a fuzzified syntactic pattern recognition concepts and a neural network concepts. The back-end subsystem which evaluates the characteristics of the events is built based upon object-oriented rule base system concepts. A design methodology for applying the proposed architecture has also been proposed in the thesis. The design methodology suggests an abstraction mapping concepts and generic task structure for conceptual modeling of domain knowledge, and defines detail functions should be performed in design and detail design modeling steps of the development life-cycle. The proposed architecture has been verified by evaluating a prototype which was developed to automatically interpret non-destructive evaluation signals to inspect health of tubes used in nuclear power plants. Performance analysis results for the prototype show that the proposed architecture can enhance the productivity and quality of the signal inspection task, and architectural analysis results show that the ISIS architecture can reduce the complexity of the problem space for developing a knowledge-based signal inspection. The proposed design methodology can be used as a generalized guideline for applying the proposed arch...