摘要
针对直升机悬停时吊挂载荷所受拉力因下洗流扰动、吊挂载荷姿态变化及风扰等多源非线性耦合影响,难以实现实时高精度预测,影响直升机吊挂飞行安全的问题,提出了一种基于海洋捕食算法优化随机森林算法的拉力预测方法。首先,采集实验数据并构建随机森林回归模型;然后以训练集均方根误差为适应度函数,利用海洋捕食算法对树数和最大深度进行全局优化,获得最优模型参数;最后,将优化后的模型应用于测试集,实现高精度的吊绳拉力预测。仿真与实验结果表明,基于海洋捕食优化算法优化随机森林模型的预测准确率高达97.85%,验证了该方法的高效性和可靠性,为后续在线自适应控制策略提供了坚实的数据支撑。
Due to the influence of multi-source nonlinear coupling such as downwash flow,attitude change of the hanging load and wind disturbance,it is difficult to achieve real-time and high-precision prediction of the tension force of the hanging load when the helicopter is hovering,which affects the flight safety of the helicopter.To solve the problem,a random forest algorithm optimized by Marine Predators Algorithm is proposed to predict the tension force of the hanging load.Firstly,the experimental data are collected and the random forest regression model is constructed;then,the root mean square error of the training set is used as the fitness function,and the marine predators algorithm is used to globally optimize the number of trees and the maximum depth to obtain the optimal model parameters;finally,the optimized model is applied to the test set to achieve high-precision prediction of the sling tension.Simulation and experimental results show that the prediction accuracy of the random forest model optimized by the Marine Predators Algorithm is as high as 97.85%,which verifies the efficiency and reliability of the method and provides solid data support for the subsequent online adaptive control strategy.
作者
陆子怡
何建
刘毅臻
LU Ziyi;HE Jian;LIU Yizhen(Institute of Electronic and Electrical Engineering,Civil Aviation Flight University of China,Jianyang 641400,China)
出处
《机械与电子》
2025年第12期68-72,80,共6页
Machinery & Electronics
基金
大学生创新创业训练计划项目(S202410624165)。
