UTILIZATION OF FOURIER DECOMPOSITION FOR ANALYZING TIME-PERIODIC FLOWS

The analysis of complex time-dependent flows is difficult because of t he immense multi-dimensional data sets involved. The use of Fourier de composition of time-periodic hows can alleviate the situation in certa in cases. For many laminar flows, the fast convergence of the Fourier series allows a compact representation of the time-dependent solution. In typical cases less than 10 modes (amplitude and phase angle) can a ccurately reconstruct the solution in the physical domain. The resulti ng saving in storage is more than one order of magnitude and it can be utilized in the graphical as well as physical analysis of the flow fi eld. Significant nonlinear interactions are manifested in the Fourier domain by strong steady streaming or by the generation of high modes. Propagating flow structures can be identified by their pattern of ampl itudes and phase angles. Yet, the identification of the nature of the flow structure (e.g., vortices) requires more elaborate procedures. Tw o distinct flow examples are employed to test these notions: the exter nal self-excited vortex shedding behind a circular cylinder and the in ternal pulsatile flow in a constricted channel. These flows represent a wide range of laminar time-periodic flows characterized by propagati ng vortices.