Prediction of Battery Remaining Capacity in Space Power System Using Extreme Learning Machine

Shenggen Zheng,Wentao Huang,Qinqin Fan
DOI: https://doi.org/10.1109/icpes53652.2021.9683837
2021-01-01
Abstract:State of a storage battery in the space power system is the most factor for the spacecraft safety. Therefore, how to estimate the remaining capacity of the battery is a vital task. To end this, a prediction model of battery remaining capacity is proposed in the current study. In the proposed model, the extreme learning machine (ELM) is used and a grid search is applied to find its optimal hyperparameters. Experimental results show that the prediction accuracy of the proposed model is better than that of other competitors. Moreover, the proposed algorithm uses less time.
What problem does this paper attempt to address?