Data driven engine model identification and real-time adaptation

New model-based engine identification and adaptive control algorithms are introduced. In a first attempt to reduce the development time, control-oriented models are employed to represent engine processes. Model parameters are automatically identified for nominal engine operating conditions. To reduce model temperature sensitivity, such as during cold conditions, a scheme for real-time fast adaptation of model parameters is proposed. To maintain a high quality of control under all operating conditions, slow adaptation of model parameters is used to counteract the effects of engine-to-engine variations and at the same time to compensate for the effect of component aging and degradation. Experimental results for the implementation of the control algorithm performance in a vehicle with a V8 engine are presented and discussed.