基于Morlet小波改善飞行颤振试验参数识别精度的方法

Investigation on Improving the Accuracy of Parameter Identification Based on Morlet Wavelet in Flight Flutter Testing

为了解决激励能量有限和现场测试数据量较少、噪声大,系统参数识别的准确度差的问题,采用Morlet小波时频滤波和频域参数识别相结合的方法进行参数识别来提高精度。基于Morlet小波函数建立特性滤波器组进行时频域滤波,讨论滤波参数的选取方法,采用有理正交多项式(RFOP)拟合算法进行频域参数识别,基于欧洲航空界广泛采用的GARTEUR飞机模型数据建立密频模态系统,进行飞行颤振的试验数据仿真。结果表明该方法在信号噪声较大时,可以有效地提高系统参数识别的精度。

Since the limited stimulated power during the testing and the testing data with bigger noise, it is difficult to get accurate parameter identification. To improve the accuracy of parameter identification we investigated the filtering method based on Morlet wavelet to get fine Frequency Response Function （FRF） and then identified parameter by means of Rational Fraction Orthogonal Polynomials （RFOP） method which was quick in calculation. The feature filter bank was built to denoise in temporal frequency domain and the filter parameter selection was discussed. Numerical simulation was conducted using GARTEUR aircraft model excited by sweeping input. Results show that accuracy of the estimated FRFs is improved. Error of mode damping identified from de-noised signals is decreased comparing with those obtained from noise-corrupted signals.