Optimization of variable geometry regulation law for a Turboshaft turbofan conversion engine during mode transition using an LPV model

Abstract

To satisfy the requirements of high-speed vertical take-off and landing (VTOL) vehicles for multi-mode propulsion, a Turboshaft turbofan conversion engine is required to achieve smooth and efficient transition between turboshaft and turbofan operating modes. However, during the mode transition process, the coordinated regulation of variable geometry mechanisms induces significant variations in aerodynamic and thermodynamic parameters, resulting in strong nonlinearities, multi-variable coupling, and time-varying characteristics. These complex behaviors may lead to thrust fluctuations and degraded dynamic performance. To address these issues, an optimization method for variable geometry regulation of a Turboshaft turbofan conversion engine based on a linear parameter-varying (LPV) model is proposed.First, the importance of variable geometry control variables during the mode transition process is evaluated using a random forest algorithm, and the scheduling parameters of the LPV model are reduced accordingly. Subsequently, local linear state-space models are established within the selected scheduling parameter space, and an LPV model covering the entire mode transition process is constructed through interpolation. Based on the developed LPV model, a dynamic optimization problem considering actuator constraints and trajectory smoothness is formulated, where the minimum total thrust fluctuation during mode transition is selected as the optimization objective. The optimal regulation trajectories of the variable geometry mechanisms are obtained using a sequential quadratic programming (SQP) algorithm.Simulation results demonstrate that the proposed LPV model can accurately characterize the dynamic behavior of the Turboshaft turbofan conversion engine during mode transition. Compared with the original regulation strategy, the optimized variable geometry trajectories exhibit smoother and more continuous variations, reducing the total thrust fluctuation by 27.4 % and the maximum thrust variation rate by 58.8 %. The proposed method effectively improves the dynamic performance during mode transition and provides a reference for real-time optimization control of Turboshaft turbofan conversion engines.