SPECIAL ISSUE RESEARCH ARTICLE Steam consumption prediction of a gas sweetening process with methyldiethanolamine solvent using machine learning approaches Meisam Moghadasi | Hassan Ali Ozgoli | Foad Farhani Department of Mechanical Engineering, Iranian Research Organization for Science and Technology, Tehran, Iran Correspondence Hassan Ali Ozgoli, Department of Mechanical Engineering, Iranian Research Organization for Science and Technology, Sh. Ehsani Rad St., Enqelab St., Parsa Sq., Ahmadabad Mostoufi Rd., Azadegan Highway, PO Box: 33535111, Tehran, Iran. Email: a.ozgoli@irost.org Summary This study proposes a comprehensive data processing and modeling framework for building high-accuracy machine learning model to predict the steam con- sumption of a gas sweetening process. The data pipeline processes raw historical data of this application and identifies the minimum number of modeling vari- ables required for this prediction in order to ease the applicability and practical- ity of such methods in the industrial units. On the modeling end, an empirical comparison of most of the state-of-the-arts regression algorithms was run in order to find the best fit to this specific case study. The ultimate goal is to lever- age this model to identify the achievable energy conservation opportunity in such plants. The historical data for this modeling was collected from a gas treating plant at South Pars Gas Complex for 3 years from 2017 to 2019. This data gets passed through a multistage data processing scheme that conducts multicollinearity analysis and model-based feature selection. For model selec- tion, a wide range of regression algorithms from different classes of regressor have been considered. Among all these methods, the Gradient Boosting Machines model outperformed the others and achieved the lowest cross- validation error. The results show that this model can predict the steam con- sumption values with 98% R-squared accuracy on the holdout test set. Further- more, the offline analysis demonstrates that there is a potential of 2% energy saving, equivalent to 24 000 metric tons of annual steam consumption reduc- tion, which can be achieved by mapping the underperforming energy consump- tion states of the unit to the expected performances predicted by the model. KEYWORDS energy conservation opportunity, energy consumption prediction, machine learning, natural gas sweetening 1 | INTRODUCTION The natural gas consumption as an energy carrier has been rising steadily over the last few decades. In the past 10 years, 22% of the total energy consumed globally was attributed to natural gas. 1 Due to this significant energy use, energy performance improvement of natural gas treating plants is an important area of focus, where many classical approaches have been proposed based on simu- lation and numerical modeling. The implementation of such techniques requires an accurate description of the first-principles engineering equations. This requirement, Received: 12 June 2020 Revised: 17 August 2020 Accepted: 20 August 2020 DOI: 10.1002/er.5979 Int J Energy Res. 2020;1–15. wileyonlinelibrary.com/journal/er © 2020 John Wiley & Sons Ltd 1