During the last decade a number of data gathering systems have been placed in service aboard ship to provide information on diesel engine condition. Currently available sensors and display instrumentation now offer a wide range of real-time engine performance and condition information. However, data interpretation and evaluation still require considerable experience and good judgement on the part of on-board personnel. Incorporating engineering and technical knowledge into condition monitoring systems could substantially reduce the workload aboard currently operating ships and offers the promise of reduced onboard experience levels required in the future.
This paper describes one approach to embedding substantial cause-effect knowledge into current and future diesel engine data acquisition systems aboard ship. The resulting “smart” system could offer interpretation assistance to onboard personnel, supplementing their ability to deal with large volumes of information and learning from them as the importance of various symptom-fault relationships is quantified. The approach incorporates a procedure by which a knowledge matrix is initially filled with best available information.