Turbine Efficiency Estimation for Fault Detection Application

Riccardo Ceccarelli (1, 2) , Philippe Moulin (2) , Carlos Canudas de Wit (1)

1 NECS - Networked Controlled Systems
2 IFPEN - IFP Energies nouvelles

In nowadays Diesel engine, the turbocharger system plays a very important role in the engine functioning and any loss of the turbine efficiency can lead to drivebility problems and the increment of emissions. In this paper, a VGT turbocharger fault detection system is proposed. The method is based on a physical model of the turbocharger and includes an estimation of the turbine efficiency by a nonlinear adaptive observer. A sensitivity analysis is provided in order to evaluate the impact of different sensors fault, (drift and bias), used to feed the observer, on the estimation of turbine efficiency error. By the means of this analysis a robust variable threshold is provided in order to reduce false detection alarm. Simulation results, based on co-simulation professional platform (AMEsim© and Simulink©), are provided to validate the strategy.

Domaines: Automatique / Robotique

Riccardo Ceccarelli, Philippe Moulin, Carlos Canudas de Wit. Turbine Efficiency Estimation for Fault Detection Application. SAE 2010 World Congress & Exhibition, Apr 2010, Detroit, United States. pp.s/n, ⟨10.4271/2010-01-0568⟩. ⟨hal-00536623⟩