Search

There are complex coupling relationships between flight motion and energy management for hybrid electric fixed wing UAVs with photovoltaic/fuel cell/battery. Aiming at realizing high-energy-efficiency long endurance autonomous flight, this paper proposes a coordinated optimization method of trajectory optimization and energy management for hybrid electric UAVs. Numerical simulation and experiment are carried out. Separated by surplus demand energy except that produced by PV, this method includes an optimal energy flight trajectory optimization layer and an energy management layer. For the former, fuzzy neural network sequential convex optimization (FNNSCP) is proposed to handle the flight trajectory optimization problem for maximizing utilization solar energy. For the latter, an online energy management strategy based on convex quadratic programming MPC (CQPMPC) is proposed to realize the power allocation. Numerical simulation results show that FNNSCP reduced 2% energy requirement compared with Radau pseudo-spectral method (RPM) from flight trajectory optimization level. From energy management level, CQPMPC saves hydrogen consumption by 3.1% and 16.3% compared with nonlinear model predictive control (NMPC) and fuzzy logic state machine (FLSM), respectively. In the experiment, less hydrogen fuel is consumed by proposed method by 5.2% and 15.6% compared with NMPC and FLSM, respectively. Experimental results show that the proposed method has better energy-saving performance and excellent real time performance. It indicates that optimizing flight trajectory and improving energy management efficiency can realize the high-energy-efficiency flight of hybrid electric UAVs. This method is conducive to promoting the application of hydrogen energy in the field of UAVs. [Display omitted] • A coordinated optimization method is proposed for the high-energy-efficiency flight. • Fuzzy neural network adaptive sequential convex optimization method is proposed. • Online energy management strategy with convex quadratic programming MPC is proposed. • The proposed method is verified by simulation and energy management experiment. [ABSTRACT FROM AUTHOR]

Energy management is crucial to realize efficient operation for hybrid electric UAVs. Adaptive hierarchical energy management strategy (AHEMS) is proposed to improve energy management efficiency for hybrid electric UAVs. AHEMS includes an optimization layer and a tracking layer. The former optimizes battery state of charge (SOC) reference trajectory and demand power by considering hybrid energy system characteristics, flight mission requirement and UAV model constraints. Sequential convex optimization method is proposed to solve battery SOC global reference trajectory. The latter realizes online power allocation by tracking the battery SOC reference trajectory through proposed model predictive control based on convex optimization. Meanwhile, fuzzy logic neural network is proposed to adaptively adjust prediction horizon for balancing the computational cost and optimality. AHEMS takes battery health management into account and prevents battery from over charging or over discharging. Numerical simulation results demonstrate that AHEMS can improve the energy management efficiency. Less hydrogen is consumed and battery is used more efficiently. Compared with nonlinear model predictive control (NMPC) and fuzzy logic state machine (FLSM), AHEMS saves 19.2% and 29.7% hydrogen fuel, respectively. Experimental results indicate that AHEMS is applicable for online energy management. It has good energy management effect. AHEMS is conducive to promoting the application of hybrid electric UAVs. [ABSTRACT FROM AUTHOR]