ABSTRACT
Pipeline pressure drops can be decreased by various methods. Utilizing small amounts of additives that named drag reducing agent (DRA) in pipelines with crude oil may reduce the friction caused by fluid. It highlights the importance of accurate approximation of drag reduction (DR). In this work, the performance of artificial neural network (ANN) in forecasting DR in crude oil pipelines was enhanced using artificial bee colony (ABC) and particle swarm optimization (PSO) algorithms. To this end, we considered Reynolds number, concentration of DRA, type of DRA, temperature, and kind of pipe as the DR influential factors. Using 80:20 ratios for determining the training and testing data, each model performed with its optimal parameters. Three accuracy criteria of coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE) were used to evaluate their efficiency. Based on the results, applying ABC and PSO evolutionary algorithm leads to increasing R2 from 0.9487 to 0.9743 and 0.9806 in the training phase, and from 0.9584 to 0.9795 and 0.9835 in the testing phase. Moreover, a considerable decrease was observed for the RMSE (30% and 36%) and MAE (31% and 38%) in the testing stage.
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ueso.