DIAGNOSIS OF MARINE DIESEL ENGINE FAULTS BY PATTERN RECOGNITION OF ACOUSTIC SOUND

The fact that 50-90% of faults in marine diesel engines in the MO classification ship (unmanned ship) are detected by the crew, implies that human hearing plays a crucial role in the diagnosis of the engine. The ability to diagnose by hearing is improved by experience in hearing the fault sounds. Recently, crew have had little opportunity to hear faulty sounds. This research aims at establishing a new diagnosis system, based on the acoustic signal for four-stroke marine diesel engines, in order to assist or relieve the engineers diagnosis efforts. This paper describes how to discriminate between exhaust valve gas leaking operations and normal operations. Distinctive signals in primary recorded sound waves are recognised by means of linear regression and multivariate analysis (discriminant analysis) of diesel engine combustion cycles. It is proposed that, by means of linear regression, it is possible to distinguish differences in the patterns of primary waves between normal and fault operations of diesel engines. This system could be developed into a more general system in future.

• Supplemental Notes:
  • ICMES 93, Marine System Design and Operation; 6th Intl Conf on Marine Engng Systems; 29 Sept - 1 Oct 1993; Hamburg, Germany. Sponsored by Inst Marine Engrs, UK, et al. Procs. Publ by Marine Management Holdings, London, UK, ISBN 0-907206-53-0. Ppr 16, p 161 [10 p, 4 ref, 4 tab, 6 fig]
• Authors:
  • Hikima, T
  • Katagie, T
• Publication Date: 1993

Language

• English

Subject/Index Terms

• TRT Terms: Diesel engines; Fault location; Signals; Sound
• Old TRIS Terms: Sound signals
• Subject Areas: Marine Transportation; Vehicles and Equipment;

Filing Info

• Accession Number: 00708083
• Record Type: Publication
• Source Agency: British Maritime Technology
• Files: TRIS
• Created Date: Aug 14 1995 12:00AM