Trends on modelling techniques applied on ship's propulsion system monitoring

Abstract

The aim of the work deals with some aspects regarding modelling techniques using analytical redundancy usually applied in fault detection, fault isolation, decision making and system recovery when monitoring severe or critic control systems such as ship propulsion systems, in order to achieve fault tolerant control systems. To get such modelling objectives, back-propagation neural networks are used as universal functional approximation devices or soft sensors whose outputs are compared with real-time data to achieve residuals.