Optimising and analysis of the hydraulic performance of a water dispersion needle sprinkler using RF-NSGA II and CFD

Abstract

The hydraulic performance of a needle dispersion sprinkler is related to the combination of its rotation speed and the parameters of the water dispersal needle, which impacts the water-saving and energy-saving efficacy of the sprinkler irrigation system. An experimental device was designed to provide data, and the random forest model (RF) and non-dominated sorting genetic algorithm II (NSGA II) were employed to optimise throw radius and combined uniformity. A CFD method was used to analyse the mechanism for improving hydraulic performance. The results indicated that the RF model predicted the throw radius and combined uniformity with high accuracy and the minimum fit value exceeded 0.96. The maximum deviation between the RF-NSGA II algorithm results and the experimental results did not exceed 8%. The optimisation results achieved the optimal matching of key parameters and improve the throw radius and uniformity. The decrease in entropy production and the increase in jet velocity were the primary reasons for the increase in the throw radius. The increased air entrainment rate and the enlarged liquid film distribution area, along with an increased number of ligaments and droplets, contribute to an improved combined uniformity. This study provides valuable insights and references for improving the hydraulic performance of sprinkler.