Integration of physical and statistical models for automotive engine control
Akira Ohata, Katsuhisa Furuta · 2006
To encounter the complexity and the time consuming problem of engine control system developments, almost all automotive manufacturers want to realize Model Based Development (MBD) process. Rapid plant modeling is one of the most essential issues to establish MBD. In this paper, two types of physical and statistical model integration have been proposed as the model simplification methodologies based on the error function between physical model and experimental data and nonlinear ARX identification with balanced realization by using the generated data from physical model.