Two-Step Energy Management Optimization Based on Anti-Disturbance Model Predictive Control for Hybrid Power System in Unmanned Aerial Vehicle Applications

Unmanned aerial vehicles (UAVs) have attracted increasing attention in civil and military applications. To achieve long endurance and large load for UAVs, the proton exchange membrane fuel cell (PEMFC) and Li-ion battery (LIB) hybrid power system (HPS) is an attractive scheme. In this paper, we first improve a fully-active PEMFC/LIB HPS by integrating an additional low-power consumption switch. This improvement enables the PEMFC to directly supply power through the low-power consumption switch, thereby effectively minimizing energy conversion loss. Furthermore, a two-step energy management optimization strategy including low-pass filter-based operation mode switch and model predictive control (MPC) is proposed for the PEMFC/LIB HPS in UAV applications. The low-pass filter can mitigate the disturbances caused by power fluctuation and sampling noise, while the MPC would reduce the hydrogen consumption and performance degradation. Compared with the conventional MPC-optimization scheme for the HPS in UAV applications, the improved anti-disturbance MPC-optimization scheme can reduce hydrogen consumption by 6.4%, extend the PEMFC service life by 7.7%, and double the computation rate of energy management implementation. Moreover, with the improved anti-disturbance MPC-optimization scheme, the LIB state of health (SOH) is improved by 2.2 and 7.1 times compared with the rule-based scheme and the conventional MPC-optimization scheme, respectively.