Sound and image process techniques
The microphone used was an electret type one, cooperated with a digital voice recorder whose sample frequency was set at 8000 Hz. Being able to convert a discrete signal expressed in a series of time to one expressed in a series of frequencies, the Fourier transform is widely used in the analysis of fluctuation, vibration and sound. One generally implemented form is Fast Fourier Transform (FFT) algorithm. By FFT, the time domain acoustic waves were converted to frequency domain ones, so the dominant frequencies and its amplitudes can be clearly seen. According to the Nyquist sampling theorem, only the fluctuations whose frequencies are blow 4000 Hz (half the sample frequency) can be well depicted by those discrete digital signals while the remaining frequencies are meaningless. The 8000 Hz frequency here is high enough because the results show no frequencies exceed 3000 Hz.
A HX-6 high speed camera made by NAC Company was used to capture the process of steam-air jet plume condensation. HX-6 has a CCD of 5 mega-pixels and could take photos as high as 650,000 frames per second. In this experiment, images were taken at a rate of 3000 frames per second. Image size is 640 pixels width and 1280 pixels height with each pixel representing a distance of 0.06383 mm.
Void fraction or gas volume fraction distribution in the flow field was obtained by images process using MATLAB. The process procedure of a single image can be found in Qu et al. (2015), by which an image turns into a matrix whose dimension equals to that of the image. In the matrix, an element marked one represents the occupation of gas, while an element marked zero represents the occupation of water. After adding a series of these matrixes converted from images shot instantaneously, and dividing the resulted matrixes with the number of matrixes, we get the probability field of whether or not there’s gas, which is the gas volume fraction.