Air Leakage Diagnosis in Heavy Duty Truck Engines with EGR and VGT

Sammanfattning Scania CV AB is a leading company within development and production of buses, trucks as well as industrial and marine engines. New environmental and safety legislations continuously demand higher quality from the products. An upcoming European legislation, Euro 6, implies that gas leakages from truck engines should be detected while driving. If the source of the leakage is not only detected, but also isolated, that is separated from other faults, the adjustments in the workshop goes faster since there is no need for leakage localisation. A faster reparation increases the up-time, i.e. the amount of time that the truck can be used. This master thesis work uses current methods developed at Scania for residual generation to perform model-based leakage diagnosis. In this work, measurements are gathered for different sensor faults and two leakages. The measurements are used to evaluate the actual performance of the resulting diagnosis system. The result, based on the residuals generated by the method, shows that leakages on the boost-side and the exhaust-side can be detected, and isolated from faults in the pressure sensors on the boost-side and the exhaust-side. The isolation of these four faults is considered the hardest to achieve among sensor faults and leakages why the full isolation performance is promising. Further measurements are needed to determine the full isolation performance of the diagnosis system. The resulting system is reasoned to be suitable for execution in real time on-board the truck.