Fault Detection and Diagnosis of IC Engine based Power Trains by IAS Analysis

International audienceInternal combustion (IC) engine based power trains are common transmissions in different industries such as trains, standby power supply and marine drives. Conventional vibration based monitoring often needs a considerable number of sensors because of large layout of the system, which is cost-intense and difficult to operation. This study focuses on developing a cost-effective approach to monitoring such a large system using one or two channels of instantaneous angular speed (IAS) measurements. It models the powertrain system as a flexible nonlinear rotor and characterises IAS responses at different measurement positions when the system is under different faults such as misfires, coupling defects, lubricant deterioration and impeller defects. The results show that it is feasible to use a fewer of IAS channels to monitor the entire system, but its performance depends largely on the sensor locations and data processing methods. The simulation studies based on the developed torsional model with 8 degree of freedom demonstrates that the IAS signals at the pulley wheel achieves the best detection results for the misfires in different cylinder of the 4 cylinder engine powertrain