Performance monitoring and full diagnosis of a diesel engine turbocharger

Turbocharging systems are of high importance in the enhancement of engine power and efficiency. Faults in such systems may cause increased emissions levels beyond those set by legislation, and may also compromise fuel efficiency. This thesis investigates the application of system identification and vibration signal analysis techniques for the performance monitoring of turbocharged diesel engines by exploring turbocharger behaviour. The first technique is based on the use of system identification to build models representing the input and output relationship of an engine process. In this case, torque demand is the input, and the turbocharger speed is the output in a medium duty, turbocharged diesel engine. The proposition set here is that the model that can be derived does not have to reflect the complexity of the physical system. Hence, if simple models can be derived, any deviation from the model of normal operation, if adhering to some principles, could indicate the existence of a fault in a system. Dynamic linear auto-regressive moving averages with exogenous input (ARMAX) models were estimated to represent the relationship between input and output. The models have been used to demonstrate the capability of the proposed technique to diagnose faults.EThOS - Electronic Theses Online ServiceGBUnited Kingdo