观察一维紊据的一种有效之新拆解法.pdf

第十二屆全國計算流體力學學術研討會 高雄，中華民國九十四年八月 The 12th National Computational Fluid Dynamics Conference Kaohsiung, August,2005 CFD12-2501 觀察一維紊流數據的一種有效之新拆解法 A New and Effective Tool to Look into Details of a Turbulent Data String Yih-Nen Jeng, Professor Department of Aeronautics and Astronautics National Cheng-Kung University, Tainan, Taiwan, 70101 Email: Z6208016@email.ncku.edu.tw Chi-Tsung Chen, Ph.D Candidate Department of Aeronautics and Astronautics National Cheng-Kung University You-Chi Cheng No. 111, Tongpin Rd. Tainan, Taiwan ABSTRACT The continuous wavelet (Morlet) transform is modified by adding a Gaussian window to the Fourier spectrum corresponding to the scale function. The Jeng and Cheng strategy of effectively suppressing the non-periodic and low frequency error of the spectrum together with an iterative filter are employed so that the windowing procedure on the spectrum domain can be accurately done. The existence of the inverse transform of the present modified transformation can be shown. The bandwidth of the resulting two-dimensional wavelet coefficient plot of a single sine function is much narrower than that generated by the original Morlet transform. Subsequently, the visibility of the wavelet coefficient plot of several waves with frequencies closing to each other is significantly improved. The application of the proposed wavelet transform to the velocity data string of a low speed turbulent wake flow after a blunt body clearly shows many details which are not known before. On the res