Simulation of hydraulic head using Particle Swarm Optimization Algorithm and Genetic Algorithm. (Case study: Debal khazaie sugarcane plantation)

Abstract

Farm experiments are useful in knowing the drainage systems but they have considerable limitations including the inability to use them as prediction tools. Application of simulation models can cover these deficiencies but it is necessary to use the field data to evaluate the accuracy of the model. In this study, Particle Swarm Optimization Algorithm and Genetic Algorithm is used to predict hydraulic head. For this purpose, field R9-11 of the Debal Khazaei sugarcane plantation is selected and number piezometers were installed in different depth (2/2,3,4 and 5 meters from the ground) and distance from collector.Piezometers. hydraulic load changes, the volume of irrigation water and drainage flow were measured from September 2013 to November 2014 on a daily basis. The results showed that the Particle Swarm Optimization Algorithm has a highest accuracy in predicting hydraulic head. So that the average RMSE in different depths between measured and predicted with Particle Swarm Optimization Algorithm and Genetic Algorithm obtained 0.098 and 0.114 , respectively and the average coefficient R^2 in different depths for Particle Swarm Optimization Algorithm and Genetic Algorithm models obtained 0.991 and 0.94 respectively. The test results of the comparison between measured and simulated data show that, between any of the values predicted by the models, measured data were not significantly different.

Keywords

• forecast
• artificial neural network
• drainage flow
• particle swarm optimization
• hydraulic head