Dynamic Coordinated Control Strategy for Aircraft Hybrid-Electric Propulsion Systems Based on Hybrid Data-Driven Methods

Hybrid-electric propulsion systems (HEPS) based on turboshaft engines are an effective power solution for vertical takeoff and landing (VTOL) aircraft due to their high efficiency and high power-to-weight ratio. To ensure real-time tracking of the power target and to coordinate the stable and efficient operation of each component, a dynamic coordinated control strategy is essential. This paper proposes a nonlinear model predictive control (NMPC) strategy based on a hybrid datadriven framework to address the coordinated control problem of HEPS. First, a component-level model is developed for the variable-cycle power generation unit (PGU) that includes dualmotor electric regulation and variable guide vanes. Then, based on deep learning neural networks and mechanism analysis, a hybrid data-driven nonlinear prediction model of the HEPS is established. Finally, a nonlinear model predictive controller is designed, considering both the engine safety constraint cost and the battery state of charge (SOC) boundary cost. The results show that the proposed coordinated control strategy achieves better tracking performance in system power and voltage, and the fuel consumption is reduced by 0.63% compared with the linear MPC strategy.