Adaptive maximum available energy evaluation for lithium battery in hydrogen-electric hybrid unmanned aerial vehicle applications considering dynamic ambient temperature and aging level

Abstract

Accurate evaluation of the maximum available energy (MAE) of the lithium battery is crucial for energy management and optimization in hydrogen-electric hybrid unmanned aerial vehicle (UAV) applications. However, uncertainties in ambient temperature and cell aging level would affect the MAE of the lithium battery, posing challenges for its accurate state evaluation. This paper presents an adaptive MAE evaluation method based on an optimized sliding mode observer (OSMO) for the lithium battery. The study begins by conducting comprehensive long-term experimental tests to analyze the dependencies of the lithium battery’s MAE on different ambient temperatures and cell aging levels. Subsequently, an OSMO with a Kalman filter is developed to estimate the variations of the battery’s open circuit voltage (OCV) on the basis of a dynamic lithium battery model. Furthermore, the real-time OCV is successfully used to achieve accurate adaptive MAE evaluation for the lithium battery. The originality of this paper is that the model parameters and proportional coefficients can be updated adaptively based on the OSMO with the Kalman filter. In this case, the OCV can converge to their actual value such that accurate MAE evaluation can be guaranteed, even with uncalibrated model parameters under dynamic ambient temperature and uncertain aging levels. Finally, experimental and simulation results verify that the proposed method exhibits a convergence speed four times higher than the traditional method. The mean absolute percentage error (MAPE) of MAE evaluation is less than 1.5 % under dynamic ambient temperature and uncertain aging levels.