Real-time Estimation of Two-Dimensional Temperature Distribution in Lithium-ion Pouch Cells

Thermal management is an integral part of battery management systems due to the effect of temperature on safety, life-time and efficiency of batteries. Therefore, a reliable real-time estimation algorithm is required to estimate the temperature distribution in battery cells based on available measurements. Temperature estimation in pouch type cells is especially challenging due to the non-uniform distribution along length and breadth. Motivated by this issue, we study effective sensor placement and estimation algorithm design for pouch cells in this paper. Specifically, we explore two scenarios: Scenario 1 where multiple temperature sensors are available, and Scenario 2 where only one temperature sensor is available. For Scenario 1, we find the minimum number of sensors required and their effective locations whereas for Scenario 2 we find the effective location of the single sensor which maximize the state observability. We employ the Gramian observability analysis for this study. Subsequently, we design sliding mode observer based real-time algorithms for distributed temperature estimation in both scenarios. Finally, we illustrate the performance of the proposed estimation algorithms through extensive experimental and simulation studies.