A robust predictive technique for the pressure drop during condensation in inclined smooth tubes

The pressure drop during condensation in inclined tubes at different saturation temperatures is one of the most important design parameters in many applications. Due to need of huge investments for providing a highly equipped laboratory and difficulties in data collection over challenging situations, developing a high performance predictive model is helpful to design and optimize condensers with lower pumping costs as a result of accurate estimation of pressure drops. In this communication, the potential of four different universal intelligent models, particle swam optimization-artificial neural network (PSO-ANN), genetic algorithm-least square support vector machine (GA-LSSVM), hybrid approach-adaptive neuro fuzzy inference system (Hybrid-ANFIS), and genetic algorithm-power law committee with intelligent systems (GA-PLCIS) are evaluated for precise estimating the pressure drop (ΔP) and frictional pressure drop (ΔPfric). The comparative results demonstrated that the developed GA-LSSVM, Hybrid-ANFIS, and GA-PLCIS models could be implemented to establish favorable predictions for the application of interest. Nevertheless, the GA-PLCIS models by combining the merits of the single developed models indicate higher performance by introducing a R2 = 0.9990752581, MSE = 0.0140, and RRMSE = 2.4983 for the ΔP and a R2 = 0.9990960793, MSE = 0.0126, and RRMSE = 2.2414 for the ΔPfric. Based on the results, the GA-PLCIS can be taken into account as a practical and easy-to-use model with high accuracy predicting performance, which is highly helpful for engineers to monitor the precise results under different conditions, even in the challenging situations such as low mass fluxes and low qualities.