The application of data mining for marine diesel engine fault detection

Abstract

Early fault detection for marine diesel engines is very important to ensure reliable operation throughout the course of their service. An early fault detection method is introduced in this paper based on the thermal parameters, and a fault predication system for marine diesel engine is developed based on abnormal data mining technology, which acquires the thermal parameters of the diesel engine from the local safety, alarm and control system through field bus, manages the data by database, and predicts the operation condition through statistic and data miming technology. It is found that the abnormal data mining is effective to fault detection at the early stage.