Parameter Estimation for Physical Based Air Path Models of Turbocharged Diesel Engines — An Experience Based Guidance

Physical based air path models lead to a substructuring of the highly complex engine systems into several interacting submodels of low order. They offer detailed process information, support advanced control system design and allow to significantly reduce the calibration effort. Hence, physical approaches are predestinated to cope with the rise in system complexity and with the increasingly challenging demands concerning air system performance. Whereas the basic model equations are known a general methodology to obtain the model parameters is lacking. The purpose of this paper is to shed light on the identification procedure and to offer the automotive engineer helpful advice to gain well calibrated simulation models. Analysing the air path equations the determining factors on the parameter quality are investigated. Based on the results sensible modifications of the test bed setup and the measurement strategy are presented. In addition the need for careful data processing is discussed. The comparison of simulation results and measurement data from the engine test bed proves the potential of the developed methodology.