Knowledge Based Expert System for Defect Identification and Rectification in Engine and Steering Control Systems of Fast Attack Crafts

Abstract

Cognitive systems deal with symbolic manipulations on knowledge and it’s stored as rules, theories etc. State-of-the-art fault detection methods are equipment and domain specific and non-comprehensive. However, possessing domain knowledge and human reasoning can be applied for fault detection by having a thorough understanding of the associated system and its surroundings. This study introduces a complete semantic framework for fault detection and diagnostics (FDD) in system simulation and control of an indigenously designed engine and steering control system for Fast Attack Crafts (FAC) by the Sri Lanka Navy. The suggested technique includes the construction of knowledge base for FDD purposes using rules and offers increased functionality of such systems using inferencebased reasoning to extract information about operational anomalies. Hence, an Expert System (ES) has been designed as a solution for defect identification and rectification (DIDR) challenge for the indigenously designed Naval Propulsion and Steering Control (NPSC) System onboard FACs.