Development of a dynamic predictor of battery cooling in automotive applications

Electrification of transport is central to achieving net-zero carbon emissions. Yet, this requires making significant improvements in battery technology. As an inherent characteristic, electrochemical batteries can operate optimally within a very narrow range of temperature and hence require constant cooling. In practice, the heat generation in batteries depends upon the applied load, while in electric vehicles such load is highly time dependent. Hence, unlike that in conventional cars using engines, battery cooling in electric vehicles needs to be an unsteady process. This, in turn, calls for employment of a controller to decide upon the temporal changes in the cooling rates. The controller should predict the future cooling requirements and adjust the coolant flow rate accordingly. Further, since the load and thus battery heat generation can change quickly, the prediction should be done over a short time. It follows that there needs to be a prediction capability that allows for the development of a dynamic thermal management of batteries. However, an essential restriction for this, is the limited computational power available onboard in a vehicle.
This project aims to develop a robust predictor of the heat generation and transfer in a typical battery pack used in electric vehicles. Models for electrochemistry of battery are coupled with the simulations of the cooling process to develop a comprehensive numerical model of heat generation and removal. This will be a high-order model which offers high accuracy with the expense of large computational power and cost. To reduce these, a low-order model is developed that requires significantly less computational power compared to high-order model and still provides acceptable predictions of the cooling process. The low-order model can be operated on a small computer in a short time and thus serves as a low-cost, dynamic predictor of battery cooling. The capabilities of the developed low-order model will be verified through comparison with the outputs of the high order model.
The aims and objectives of this project are:
1- To develop an accurate numerical model of heat generation and transfer in batteries in order to further understand the interactions between the underlying electrochemical and thermal processes.
2- To develop a low-order model of dynamic battery cooling and validate that against the high order model.
Potential applications and benefits:
This project has a direct relevance to the current technical issues faced by the electric vehicle industry. Better control of the battery temperature prolongs the battery life and thus helps bringing down the price of electric transportation while it also reduces the stress on the natural resources required for battery production.