Search Results

Historically, electrical-equivalent modeling of battery systems has been the preferred approach of engineers when modeling hybrid and electric vehicles at the system level. This approach has provided modeling engineers good boundary conditions for batteries, with accurate terminal voltage and state of charge (SOC) calculations; however, it fails to provide insight into the electrochemical processes taking place in their Lithium-ion cells, necessary to optimize control algorithms and predict aging mechanisms within the battery. In addition, the use of predictive battery models that simulate electrochemical mechanisms empowers engineers with the ability to predict the performance of a Lithium-ion cell without requiring cells to be manufactured. If hardware is already available and tested, the use of physics-based battery models allows the simulation of the cell to be done well beyond the conditions at which the battery has been tested.

One promising solution for increasing vehicle fuel economy, while still maintaining long-range driving capability, is the plug-in hybrid electric vehicle (PHEV). A PHEV is a hybrid electric vehicle (HEV) whose rechargeable energy source can be recharged from an external power source, making it a combination of an electric vehicle and a traditional hybrid vehicle. A PHEV is capable of operating as an electric vehicle until the battery is almost depleted, at which point the on-board internal combustion engine turns on, and generates power to meet the vehicle demands. When the vehicle is not in use, the battery can be recharged from an external energy source, once again allowing electric driving. A series of models is presented which simulate various powertrain architectures of PHEVs. To objectively evaluate the effect of powertrain architecture on fuel economy, the models were run according to the latest test procedures and all fuel economy values were utility factor weighted.

A structured, semi-automatic method for reducing a high-fidelity engine model to a fast running one has been developed. The principle of this method rests on the fact that, under certain assumptions, the computationally expensive components of the simulation can be substituted with simpler ones. Thus, the computation speed increases substantially while the physical representation of the engine is retained to a large extent. The resulting model is not only suitable for fast running simulations, but also usable and updatable in later stages of the development process. The thrust of the method is that the calibration of the fast running components is achieved by use of automatically selected neural networks. Two illustrative examples demonstrate the methodology. The results show that the methodology achieves substantial increase in computation speed and satisfactory accuracy.