Observability analysis for surface sensor location in encased battery cells

Compact battery pack design and cooling requirements present significant challenges for the placement of temperature sensors. Typically, temperature sensors are located near the edge of the battery, and away from the hottest regions, which leads to slower response times and increased errors in the prediction of interior cell temperature. New sensor technology allows for sensor placement between cells to improve sensor performance. With the ability to place sensors anywhere on the exterior of the cell, an observability analysis is necessary to determine the optimal locations for these sensors. The analysis is performed using a spatial discretization of a validated electrothermal model. This model describes a 5 Ah Li-ion cell harvested from a Ford C-max 2013 pack. Given that the spatial discretization of the heat partial differential equation (PDE) governing the system results in singular values that are very small (numerically zero), two methods are presented in this paper to quantify observability and address the issue of optimal sensor placement. The optimal sensor placement between cells yields a 240% improvement (for 3 sensors on the surface) and 15% improvement (for one sensor on the surface) in observability over existing sensor placement near the top edge of the cell. The pack geometry and airflow conditions impact sensor placement. It is preferable to place the sensors towards the outlet side of the airflow as opposed to the inlet side, resulting in a 13% improvement in observability.