The use of first principle modelling for faulty behaviour prediction in the design stage

摘要

The use of intelligent condition monitoring systems in ship engine rooms, if any, is today mostly limited to single parts of equipment, such as the diesel engine or a generator set. Obtaining the knowledge required to build an intelligent monitoring system is difficult, the more so for systems that include equipment delivered through various suppliers. Many of the systems on board ships belong to this category, such as the fuel oil treatment system, the cooling systems and some of the refrigeration systems. These systems also include ship specific equipment such as tanks and piping. Taking into consideration the important role these systems play in the safety, reliability and economy of the ship, intelligent condition monitoring could be a useful asset. This paper describes an approach, using first principle modelling, that enables the development of diagnostic knowledge already in the design stages of the ship. Using faulty behaviour prediction, fault symptom matrices are defined, related to different operating conditions.