Implementation of Monitoring Algorithms for the Battery Pack Optimal Design and Control in an Electric Vehicle

Abstract

Battery monitoring algorithms and exclusively state of charge (SOC) estimation is the key function for successful operation of the battery management system (BMS). Enabling high storage utilization, safe operation environment for the battery pack and individual cells, cost reduction by avoiding over engineering, fault detection and lifetime enhancement are among all other benefits provided by monitoring algorithms. In this work, up-scaled dynamic voltage simulation and SOC estimator are described for the high power lithium ion cell (63Ah) used in parallel-series connection (2X108) to form the battery pack used in concept electric vehicle (EV) taxi ‘‘EVA‘‘ developed at TUM CREATE Singapore. Models are developed in offline platform with matlab/Simulink based on the aging experiments. Models are evaluated in realtime hardware in the loop (HIL) system and the interaction between HIL and experimental set up are described. Verified models are implemented on target microcontroller for small scale evaluation and BMS of the EV for large scale implementation in the final stage. Each individual cell affects the overall performance of the pack, to deal with this, single cell SOC estimation within the battery pack is considered critical feature of the BMS. Battery pack is the most expensive component of the EV and conducting aging experiments on the pack is very expensive and requires specific laboratory equipment, hence extensive experimental tests are performed on the cells level at various controlled ambient temperature to simulate different conditions for the cells which they will go through in real application. These experimental tests are also used to create reliable reference data under controlled laboratory conditions, cell parameterization, diagnostics and aging studies on the cells among all. Not all the cells in the battery pack are identical; external/internal influences lead to cell to cell discrepancies. These influences can be from slight manufacturing process differences, hot spots due to cell locations inside the pack, sudden failures among other influences causes different behavior between cells. Weakest cell in the pack dictates the charging and discharging limits and safety level of the battery pack. This is due to the fact that the weak cell charges faster among other cells and reaches it critical voltage/temperature values before other cells being charged and discharges faster due to less capacity compared to other cell. This work demonstrates the simulation method to estimate the SOC for individual cell among the battery pack and the overall battery pack SOC considering the cell to cell variations, surface temperature differences and aging influences of the cell. This method leads to battery utilization of the high storage system and safe and reliable operation of the cells and is beneficial as well in cell balancing applications as it will be shown that these variations can be as high as 5-10% capacity difference between two cells in the battery pack. Figure 1