The modeling of oxygen transfer efficiency in gated conduits by using genetic expression programming

Abstract

Oxygen transfer is the process by which oxygen is transferred from the gaseous to the liquid phase. The oxygen transfer efficiency depends almost entirely on the amount of surface contact between the air and water. This surface contact can be increased by gated conduits that involve air-water mixture flow. In reality, the physical structure of the air-water interface is complex and still awaits clarification. In recent years, different soft computing systems have been successfully employed for the solution of complex problems. Genetic expression programming (GEP) is an example of soft computing systems. This study presents the use of GEP based on a genetic algorithm to predict oxygen transfer efficiency in high-head and free-surface gated conduits. The comparison of experimental results with the results of GEP models revealed that correlation coefficients (R2) are very high and mean square errors (MSE) are very small. Therefore, GEP models are a fairly promising approach for the prediction of oxygen transfer efficiency in gated conduits.