论文部分内容阅读
本文分析和研究了低通滤波、最小二乘法、频率法和卡尔曼滤波方法的优点和缺点,将利用卡尔曼滤波求取直升飞机气动导数的方法作了进一步的修改,卡尔曼滤波的输入数据由低通滤波数据改用频率法的输出数据,卡尔曼滤波时的初始导数由单纯的最小二乘法所获得的导数改用由频率法得出的导数。计算结果表明,改进的卡尔曼滤波方法消除了原方法由于低通滤波而造成的相位漂移,振荡频率误差显著降低,复合相关系数提高到0.9以上,特征根已十分接近真值,是一种切实可行的方法,适合于各种直升飞机的导数识别。
In this paper, the advantages and disadvantages of low-pass filtering, least-squares method, frequency method and Kalman filtering method are analyzed and studied. The Kalman filter is used to obtain the aerodynamic derivatives of helicopters. The Kalman filter input The data is converted from the low-pass filtered data to the output of the frequency method. The initial derivative of the Kalman filter is derived by the simple least-squares method using the derivative derived from the frequency method. The results show that the improved Kalman filtering method can eliminate the phase drift due to low-pass filtering, the oscillation frequency error is significantly reduced, the complex correlation coefficient is increased to above 0.9, and the eigenvalue is very close to the true value. Feasible method, suitable for a variety of helicopter derivative identification.