Abstract :
The analytical design of a control system for a highly complex, event-based system, the internal combustion engine (ICE), is a task much simplified by the use of a structured design methodology in a simulation-based environment. This task is further complicated in the next generation of ICEs, direct injection spark ignition, with new interactions and new hardware to control, and a new torque-based architecture. The use of hand-written code for this whole process is too time-consuming and error-prone. The benefits of simulation allow a platform for rapid algorithmic iterations, real-time implementation of rapid control prototyping (RCP) and, ultimately, autocoding for production. Once the control system architecture has been decided, it can be broken down into version-controlled components called features. When these have been designed and implemented, they can be executed in a test harness and modifications made. An application builder combines these control system components into a final build which is executed in real time, controlling the engine. For further diagnosis, real-time test data can be executed as a trajectory through the test harness to replicate problems. Example control system components are studied. These show the crank-based and time-based subcomponent breakdown. Both vehicle and simulation data are presented
