Abstract
Discharge voltage is an important indicator to alarm end-of-discharge of lithium-ion batteries. Therefore, prediction of the discharge voltage when the battery is in use is helpful in preventing issues caused by running out of power. For many real applications, the battery is working under unknown and dynamic loads, which makes the prediction difficult. In this paper, we propose a novel method to predict the discharge voltage under unknown future loads. This method firstly establishes the relationship between the discharge voltage and the loads, then predicts future loads based on a framework consisted of wavelet analysis and polynomial neutral network with group method of data handling. Finally, the discharge voltage is predicted using the battery model with particle filter-based updating procedure and the predicted future loads. The effectiveness of this method is demonstrated by a real flight dataset coming from experiments conducted on a plant protection UAV. The results show that our method can achieve good prediction accuracy and outperforms some other benchmark methods.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call