Spectral analysis of pressure fluctuations procuced by numerical simulations of a fluidized bed

Tulio Charão de Oliveira¹; Flavia Zinani¹; Maria Luiza Sperb Indrusiak¹

¹ Universidade do Vale do Rio dos Sinos - UNISINOS

doi:10.20906/CPS/COB-2015-0649

Resumo

Fluidization is a process in which a particulate solids flow is induced by the flow of a gas or a liquid. Fluidized bed reactors are extensively used in chemical and petrochemical industries. The technology allows for more efficient and clean burning of solid fuels and also for the conversion of solid fuels into syngas by gasification. The study of hydrodynamics of fluidization is challenging and spectral analysis of pressure fluctuation measured in fluidized beds is a tool employed for the understanding and characterization of such systems. In this work, pressure fluctuation signals generated in Computational Fluid Dynamics (CFD) simulations of a fluidized bed were analyzed using Power Spectral Density (PSD) method and discrete wavelet transform method (DWT). The PSD method is based on Fast Fourier Transform (FFT) and is able to identify dominant frequencies. The DWT decomposes the signal in frequency intervals and allows the investigation of the behavior of each frequency interval in time. Three different superficial velocities were simulated in order to obtain pressure fluctuations for different fluidization regimes. It was found that 32 s of simulation with a sampling frequency of 250 Hz was sufficient to provide a signal representing all flow features. The PSD analysis showed that the dominant frequency of the process increases when the fluidization regime goes from bubbling to turbulent. The wavelet transform showed that the main flow structures occur at very low frequencies and are well captured by decomposition in six levels.

Palavras-chave: fluidized bed; wavelets; computational fluid dynamics; pressure fluctuations