Condition monitoring of gasoline engine air intake system using second order sliding modes

This paper presents a novel strategy for the condition monitoring of Air Intake System (AIS) of a gasoline engine. The aim is to detect and identify manifold leakages and throttle valve fault. These two faults directly affect AIS performance that results in poor engine efficiency. Two critical and immeasurable engine parameters i.e., Throttle Discharge Coefficient (C _D) and Volumetric Efficiency (η_vol) are employed to monitor aforementioned faults. These parameters are estimated using Mean Value Engine Model (MVEM) and super twisting algorithm based Second Order Sliding Mode Observers (SOSMOs). Nominal values of the estimated parameters are used to generate residuals. These residuals are then evaluated to classify system health. A successful validation of the proposed scheme is conducted to diagnose and classify the health status of On Board Diagnostics version-II compliant commercial vehicle engine. The presented scheme is computationally cheap for online implementation and is readily applicable to cater for production line spread of single make.