Development and Validation of a Control-Oriented Model of a Spark-Ignition Direct-Injection Engine

Due to the increasing demands on improved fuel economy and stringent government regulations on tailpipe emissions, many automotive industries and research institutes have been looking for alternative solutions, such as diesel engines, hybrid-electric vehicles, and fuel cell technologies, over conventional port fuel injection (PFI) gasoline engines to meet the demands. On the other hand, many people in the automotive community also realize that there are still a lot of room for improvements in gasoline engine technologies, such as utilizing direct injection and/or variable valve actuation. In order to fully realize the potential benefits of such advanced technologies in gasoline engines, a well-coordinated complex control system design is essential. This paper describes the development and validation of a control-oriented mean-value model for a spark-ignition direct-injection (SIDI) engine to assist and accelerate such coordinated control system design and calibration processes via use of an engine model. The performance and accuracy of the dynamic engine model are evaluated and validated against a set of data for an engine running on a transient driving cycle.