On condition monitoring of exhaust valves in marine diesel engines

Abstract

The feasibility of noninvasive characterisation of exhaust valve conditions in large marine diesel engines were experimentally investigated on a four cylinder 500 mm bore 2-stroke marine diesel engine at MAN B&W Diesel's Research Center in Copenhagen, Denmark. The experiments comprised three different valve conditions, where two were concerned with artificially induced valve burn-through situations. The primary potential monitoring measurements were vibration and structure-borne stress waves, also known as Acoustic Emission (AE). Results have shown that the AE signals have a major advantage over other involved sensors, indicating sensitivity to both mechanical and fluid-mechanical combustion related activity. Recorded data has been preprocessed and features extracted using principal component analysis (PCA). From a number of applied heuristics and statistics, a search for the optimal sub-space of principal components to use, have been carried out. The chosen feature-space has been used for classification of involved exhaust valve conditions by applying both regularized feedforward neural classifiers and linear discriminators. The complexity of the neural networks have further been optimized by optimal brain damage pruning, leading to increased generalization.