Machine recognition of void fraction for two-phase flow measurements

An investigation was made of various histogram shapes as a means of determining the void fraction of two-phase flow of steamwater mixtures. Machine recognition techniques based on curve fitting procedures were implemented as a means of computing the void fraction. The third and fourth central moments associated with each histogram were used as the key elements in the void fraction identification. This method of void fraction prediction was found to be easily implementable, as the information necessary for the construction of the histograms could be computed and stored during the execution of the experiment. The calculations for the void fraction identification could then be completed when processing time was available after the conclusion of the experiment.