Predicting Turbulent Buoyant Jet Using Machine Learning Techniques

Abstract

In this paper, machine learning techniques are utilized to predict the temperature distribution in a vertical buoyant turbulent jet. Experimental results for five cases with different flow rates are reported. The results show that temperature behaves linearly along the vertical axis of the jet. Also, the thermal stratification phenomenon has been observed. Different machine learning techniques have been used to predict the temperature distribution in the induced vertical buoyant turbulent jet. The used machine learning including k-nearest neighbor algorithm (k-NN), artificial neural networks (ANNs), Support Vector Regression (SVR), and random forest (RF). It was found both SVR and RF methods are the best machine learning techniques to predict the temperature distribution in a vertical buoyant turbulent jet.