عنوان مقاله [English]
Two-phase flow systems analysis data are rear and cause major problems as efficient prediction in operation and control of the system or delay in identifying system parameters behavior.Analysis of the pressure waves through transient conditions can be an effective method to achieve precise results.In the present article, the bubble formation was studied using computational fluid dynamics (CFD) and the acoustic finite element method (FEM). The simulation conducted to obtain the aerodynamic noises generated by initiation and separation of bubbles in a vertical column. A Transient three-dimensional model is developed based on VOF method for tracking interface of two phases and large eddy simulation (LES) for an accurate calculation of pressure fluctuations. The model is simulated with the aid of the commercial software ANSYS FLUENT based on the finite volume method. To obtain convergence the implicit body force treatment is taken into account PISO algorithm is used to calculate a consistent result between velocity and pressure field. Pressure is discretized with PRESTO scheme. The RANS solution is used to calculate the initial flow field. In this case momentum is discretized with the second order upwind scheme and the geometric reconstruction scheme has been used as the volume fraction discretization scheme.Due to the importance of interface tracking in sound sources recognition and the problems which occur during adding LES model in the simulation, different interface reconstruction methods has been compared. High Resolution Interface Capturing scheme (HRIC) is selected as the best method.The Fofwcs Williams and Hawking (FWH) equation is used as an extension of the classical Lighthill aerodynamic acoustics for predicting far field noises. Sound sources are extracted by taking samples of pressure fluctuations from Wall penetration and virtual plane that is defined parallel to flow. These sources are investigated in the moment of separation and formation of bubble. In this paper are concluded that the largest fluctuation occur during necking of bubbles, while the smallest bubble produced the biggest peak in acoustic fluctuation curves.Besides time analysis, the pressure fluctuation is transferred from time domain to frequency by Fourier transform method and spectral analysis is performed on the data. The spectral analysis are compared to experimental results to better understand the effects of turbulence models, flow rate and methods of taking sample on the acoustic frequency response. The spectra of the filtered acoustic data (at airflow rate of 240 lit/min) shows a peak at about 40 Hz and is shown frequency response increases by the decrement of flow rate in bubbly flow. In this research frequency merge of two bubbles is demonstrated at 800 Hz and frequency response is used to estimate the bubble size.