Proposal Criterion & Method to Estimate Battery SOC in Spacecraft Application

Abstract

In low earth orbit mission (LEO), the Battery Module is an obvious solution for spacecraft's (S/C) design. It supports the power demand during eclipse period in orbit and meet peak power during sunlit. The batteries integrated in LEO-S/C has to undergo approximately 15 charge/discharge cycles in a day, used continuously without any rest or open circuit duration. In order To guarantee a good heath of battery and predict any damage during integration, environmental tests phases and also in orbit, this Module should be fallowed through indicator parameters; those observed in battery such as terminal voltage, current and temperature could not give instantly an accurate idea about its state of charge (SOC)and state of health (SOH). That is why a large number of techniques and algorithms have been used to identify, Off/On-l ine, the real available energy and the degradation caused in a battery. In our approach, we propose to integrate the effects of the unknown evolution of certain battery internal parameters, and the induced noises as a part of the battery model. In parallel, we propose to tune the estimated available energy at each debited current. In this case, the nominal battery capacity $(\text{C}_{\text{n}})$ is interpreted as energy losses depending on debited or injected current of our battery. This tuning solution omits the perturbation of any fluctuation induced by a fast-current change. By imbricate the solutions proposed previously, a global method was developed. In this global solution, we couple the Adaptive Extended Kalman Filter (AEKF) estimator to a Fuzzy-Logic model. The obtained results show that the proposed approach contributes accurately to estimate the battery state of charge SOC, and fallows its state of heath.