Fault detection of marine diesel engines using vibrations and noise

Attempts to solve the predictive maintenance problem in reciprocal engines have been already presented with promising results. A basic characteristic of these methods is the assumption of the existence of large amounts data with various faults for the fault diagnosis procedure. On the other hand little work has been performed for the in-situ predictive maintenance of marine diesel engines. At the present paper a new method is presented which combines vibration and acoustic signals, recorded at or near the operating engine and employs Wavelets and Neural Networks tools, towards the marine diesel engines' fault detection.