Determination of Pipe Diameters for Pressurized Irrigation Systems Using Linear Programming and Artificial Neural Networks

Pressurized irrigation systems have gained more attention lately, among other alternatives, in Mediterranean countries. Since the initial investment costs of pressurized are quite high, it is crucial to determine design parameters such as pipe diameter. Most current optimization techniques for diameter selection based on linear, non-linear, and dynamic programming models. The ultimate aim these produce solutions problems with less cost computation time. In this study, a novel approach determining was proposed using Artificial Neural Networks (ANN) an alternative existing For purpose, three were investigated. Different ANN architectures created tested hydrant level systems, irrigated area per hydrant, discharge, length, elevation. training algorithms, transfer functions, hidden neuron numbers tried best model each system. Using multilayer feed-forward architecture, highest coefficients determination (R2) found be 0.97, 0.93, 0.83 It concluded that diameters could determined by artificial neural networks planning systems.