Vehicle Cabin Thermal Multi-Zone Modelling for Control

Modern vehicles require thermal control systems that are capable of considering continuously varying environmental effects in order to guarantee high thermal comfort. Understanding the dynamics within a vehicle cabin is thus crucial to improving thermal comfort and ultimately overall vehicle comfort. Despite prior research within this field, the problem of generating an accurate control capable model with inter-zonal relations has not been solved. In this work, a cabin model is designed based on experimental data derived from real vehicle tests. The cabin is excited through multi-sine signals, in order to identify its nonlinear dynamic behavior. The model is structured on first principles, using a grey box method coupling the numerous measurements across a vast domain of vehicle operation. Parameter optimization is used to distinguish the various vehicle-operating points. The resulting nonlinear model has a validated coefficient of determination of 0.70 and a squared error within 0.03. Thus, it is suitable for control.