Energy management optimization in stand-alone power supplies using online estimation of battery SOC

摘要

With the aim to improve the energy management of battery packs in stand-alone power supply applications, a simple but effective model-based algorithm for battery state estimation has been developed. The method can be classified into the category of PI-based observers and exploits a Thevenin-based equivalent circuit to model the battery behavior associated with a simple start-up identification process. The algorithm provides a real-time accurate SOC estimation which can be used to evaluate the actual power capability and to predict the amount of energy flows in a long term time horizon. Moreover, useful information about battery aging such as SOH value can be obtained. Thanks to its straightforward implementation, the proposed algorithm can be conveniently integrated in the battery management systems of charge regulators and other power converters which are part of stand-alone power supplies. In such a way, it is possible to increase the harvested energy without increase the hardware requirements. A comprehensive validation has been carried out by performing several experimental comparisons between the battery state estimation performed by suggested approach and standard methods.