FPGA implementation of an observer for state of charge estimation in lithium-polymer batteries

On-line estimation of battery parameters is a strong requirement for many current applications where energy storage plays a critical role, e.g. electric vehicles and smart grids. In the last few years, different models have been proposed to represent the behavior of the battery with enough accuracy for precise estimation of state of charge. Using these models to estimate battery parameters in real-time requires a significant amount of memory resources and computational power to be available. In addition, the ability to concurrently analyze the behavior of individual cells or battery modules in a battery pack would add a significant value to any battery management system. In this context, FPGAs can be a very advantageous alternative to microcontrollers. This paper reports on the design and validation of an observer for state of charge estimation, which is the first step in authors' efforts toward the development of a whole FPGA-based on-line battery parameters identification and state estimation system.