A Desired Modeling Environment for Automotive Powertrain Controls

Abstract

AbstractA desired modeling environment for automotive powertrain control development is proposed in this paper to timely develop the required models for control system developments. Our aim is to systematically derive a grey box model consisting of physical and empirical models with the minimum order and number of the parameters. The environment consists of physical and empirical modeling, model simplification, system modeling, model/data managements, optimization methodologies and physical law libraries. In this paper, physical model is defined as the one satisfying the considered conservation laws and empirical model is defined as the one having adjusted parameters. In this paper, two approaches from a physical model to the target and a pure empirical model to the target are introduced. Physical modeling based on the constraints and the considered conservation laws is also introduced.KeywordsEmpirical ModelModeling EnvironmentParameter DistanceElectronic Control UnitIntake ValveThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.