基于支持向量机的直升机建模

Modeling for helicopter based on support vector machine

为建立高精度的直升机仿真模型,首次把支持向量机方法引入直升机智能化建模领域。对实际飞行数据进行野值剔除、高频滤波和微分平滑等预处理。在此基础上,利用支持向量机建立了直升机自转着陆过程的旋翼转速模型。与神经网络模型相比,该模型具有结构简单、运算速度快、泛化能力高等特点。理论分析和仿真结果表明,用支持向量机建立直升机的仿真模型是切实可行的。

To construct simulation model of a helicopter with high precision, the Support Vector Machine （SVM） method was introduced to the field of intelligent modeling of a helicopter. Based on pretreatment of practical flight data, such as eliminating outliers, filtering of high frequency noise, and differential smooth, the rotator speed model for landing process of a helicopter with rotator self-rotating was built up. Compared to the neural network model, the SVM simulation model of the helicopter was characterized by its simple structure, fast convergence speed and high generalization ability. It was revealed by theoretic analysis and simulation result that simulation model of a helicopter built by SVM method was feasible.