Due to the perennial problems associated with the gas hydrate phenomenon in natural gas and oil industries, it is crucial to avoid hydrate formation. If a natural gas passes through a throttling process, hydrate formation may occur. In this study, novel methodologies were developed to determine the hydrate formation limits to isenthalpic Joule-Thomson (JT) expansions for the first time. First, a new empirical correlation is introduced for predicting the required minimum initial temperature of a natural gas passing through a JT expansion without the risk of hydrate formation. Next, state-of-the-art algorithms including adaptive neuro-fuzzy inference system (ANFIS), least squares support vector machine (LSSVM), and radial basis function artificial neural network (RBF-ANN) were utilized for presenting the models capable of predicting the gas initial temperature. According to the results of the performed error analysis, it is found that the proposed methods provide excellent estimations. The proposed models show R-2 to be more than 0.99 and AARD% to be less than 0.5. However, the most accurate estimations are obtained from the presented RBF_ANN model. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.