<p>To solve the disturbance suppression problem of the flexible output of the engine under a large range of nonlinear and irregular changes in the required power during the conversion phase of a tiltrotor aircraft, an integrated aircraft/engine control method is proposed. Firstly, based on the comprehensive simulation model of the tilt-rotor aircraft and turboshaft engine, the tilt trajectory is optimized based on the concept of dynamic optimization. Then, based on the rapid prediction of the required power of the tilt rotor, an aircraft/engine integrated control method is designed by using advanced forward control. At the same time, to reduce the accuracy loss in the process of neural network pruning, an improved pruning method based on BP neural network is proposed, which is combined with stepwise regression fitting method, and random forest method is used for feature selection. The results show that compared with the traditional total distance feedforward control method, the proposed comprehensive control method can reduce the overshoot and droop of the power turbine speed by 50.68% and 48.03%, respectively. The light power prediction model reduces the calculation time of the power model by 42.6%. Compared with the total distance feedforward, the overshoot and droop suppression effects are still improved by 42.81% and 37.28%. The proposed integrated control method effectively improves the power supply quality of the turboshaft engine.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

An Integrated Control Method for Tiltrotor Aircraft/Turboshaft Engine based on Lightweight Power Prediction Model During the Conversion Phase

  • Chen Yu,
  • Guo Haoran,
  • Zhang Haibo

摘要

To solve the disturbance suppression problem of the flexible output of the engine under a large range of nonlinear and irregular changes in the required power during the conversion phase of a tiltrotor aircraft, an integrated aircraft/engine control method is proposed. Firstly, based on the comprehensive simulation model of the tilt-rotor aircraft and turboshaft engine, the tilt trajectory is optimized based on the concept of dynamic optimization. Then, based on the rapid prediction of the required power of the tilt rotor, an aircraft/engine integrated control method is designed by using advanced forward control. At the same time, to reduce the accuracy loss in the process of neural network pruning, an improved pruning method based on BP neural network is proposed, which is combined with stepwise regression fitting method, and random forest method is used for feature selection. The results show that compared with the traditional total distance feedforward control method, the proposed comprehensive control method can reduce the overshoot and droop of the power turbine speed by 50.68% and 48.03%, respectively. The light power prediction model reduces the calculation time of the power model by 42.6%. Compared with the total distance feedforward, the overshoot and droop suppression effects are still improved by 42.81% and 37.28%. The proposed integrated control method effectively improves the power supply quality of the turboshaft engine.