A Data-Driven MPC Energy Optimization Management Strategy for Fuel Cell Distributed Electric Propulsion UAV

With the development of green aviation technology, distributed electric propulsion aircraft has been the focus of research topic in the field of aviation technology due to its high flight efficiency, low pollutant emissions, high energy efficiency, and diverse aerodynamic layouts design. Compared with gas oil fuel, hydrogen-based fuel cell has the advantages such as zero emissions, low noise and high energy density. In order to improve the overall performance of the fuel-cell distributed electric propulsion UAV, Research on adaptive energy management strategies was conducted in this paper to improve the dynamic response of power system according to variation of propulsion power load. In order to deal with the uncertainty of the electric power load changes during different flight conditions of the UAV, the propulsion power demanding prediction method is presented under different flight conditions based on the flight data obtained from the real electric propulsion UAV flight testing. Based on the data-driven neural network, a distributed electric propulsion power load forecasting model was established. Based on the modeling of the distributed hybrid electric propulsion power system, three energy management strategies are proposed for comparison and verification in this paper. In view of the problem that the uncertainty of propulsion power demand under different flight conditions of UAV affects the performance of distributed electric propulsion system, an energy optimization management strategy based on deep neural network propulsion power demand forecasting combined with model predictive control is proposed. The performance evaluation of the proposed EMS is conducted via digital simulation studies using the data obtained from real-world UAV flighting experiments and its performance is compared with two benchmark schemes