论文部分内容阅读
基于内模原理设计了涡轴发动机功率涡轮转速控制器.针对主旋翼扭矩变化对功率涡轮转速的干扰,提出了一种基于极端学习机的扭矩预测方法.极端学习机训练基于动态仿真数据,其输入通过相关分析获得.基于内模原理的功率涡轮转速控制器采用极点配置的设计方法,将输入信号的内模直接加入控制器,实现鲁棒跟踪.扭矩前馈采用比例微分(PD)控制策略,实现对发动机负载变化干扰的有效补偿.数字仿真结果表明:极端学习机扭矩预测精度高,扭矩相对误差小于1.5‰,与不加前馈控制相比,所提出的控制方法减小了机动飞行过程中功率涡轮转速的超调或下垂30%以上.
Based on the internal model principle, a turbo-shaft power turbine speed controller is designed.Aiming at the disturbance of the main rotor torque to the power turbine speed, a torque prediction method based on extreme learning machine is proposed.Extended learning machine training is based on dynamic simulation data, The input is obtained through correlation analysis.The power turbine rotational speed controller based on the internal model principle adopts the pole configuration design method to directly add the internal model of the input signal to the controller for robust tracking.The torque feedforward adopts proportional-derivative (PD) control strategy , So as to realize the effective compensation for the interference of engine load variation.The numerical simulation results show that the precision of the torque of extreme learning machine is high and the relative error of torque is less than 1.5 ‰. Compared with the control without feedforward control, the proposed control method reduces the maneuver Power turbine speed overshoot or sagging more than 30%.