Diesel engine modeling with multi-model correction of manifold pressure for development of control algorithms

Abstract

A simulation model for diesel engines is widely applied to evaluate a diesel engine controller before a vehicle test because controller error can be indicated by an engine model. For the diesel engine modeling, intake manifold pressure is generally derived as an adiabatic model that describes intake manifold pressure dynamics precisely. However, this model is vulnerable to steady state error as it neglects the heat exchange. In order to solve this issue, this thesis proposes a multi-model correction algorithm for the diesel engine model. The proposed diesel engine model consists of two submodels to calculate accurate intake manifold pressure. The first is the adiabatic model designed as a mean value model that includes exhaust gas recirculation (EGR) and variable geometry turbocharger (VGT) systems. The second is a pressure ratio model between the intake and exhaust manifold to compensate the steady state error of the adiabatic model. This pressure ratio model is calculated with input variables such as engine speed, fuel injection quantity, EGR valve, and VGT vane positions. The proposed pressure ratio model based correction algorithm is designed to adjust only the steady state error to preserve the pressure dynamics of the adiabatic model. The proposed model is validated in 216 steady state conditions with a root mean square error of the intake manifold pressure less than 3.17 %. The transient performance of the model is also demonstrated by a comparison with the engine experimental results. The proposed model in this test describes the exact transient dynamics of the engine while the steady state error is reduced by virtue of the multi-model correction algorithm.