基于DE算法的共轴直升机模型辨识及精度分析

Coaxial helicopter model identification based on differential evolution algorithm and accuracy analysis

由于共轴直升机特有的旋翼布局引发了上下旋翼间强烈的气动干扰,采用传统的理论分析和风洞试验的方法难以获得适用于共轴直升机控制系统的飞行动力学模型.为此,设计了飞行扫频试验,根据飞行试验数据得到了悬停状态下包含共轴直升机飞行动力学模型耦合特性的非参数频率响应,运用仿生智能计算方法中的微分进化(DE,Differential Evolution)算法拟合频率响应建立了悬停状态下的共轴直升机状态空间模型.利用Cramer-Rao边界和不灵敏度的相关理论进行分析计算,说明辨识得到的参数具有较高的精度和可信度.通过比较辨识模型的输出和实际飞行数据的结果,说明辨识得到的模型能充分反映共轴直升机的飞行动力学特性,可用于飞行品质评估和飞控系统设计.

The coaxial helicopter exists intense aerodynamic interaction between the upper and lower rotor,and it is difficult to establish the accurate dynamic model for flight control systems using the theory analysis and wind tunnel experiment.Frequency sweep flight experiment data was used to extract the non-parametric frequency responses that fully characterizes the coupled helicopter dynamics.A nonlinear search based on differential evolution algorithm for a linear state-space model which matches the frequency-response data set was conducted.Parameter insensitivity and Cramer-Rao bound analysis results have low values,indicating very good reliability of the identified model.The accuracy of the identified model is verified by comparing the model-predicted responses with the responses collected during flight experiments,and the model can be used for flight quality analysis and flight control system design.