New Discrete Particle Swarm Optimization Applied to the Design of Pressurized Irrigation Networks

Abstract

AbstractPipe size of pressurized irrigation networks that usually have a branched layout had been optimized by assuming a predetermined layout. Few researches have considered simultaneous layout and pipe size optimization. In this study, a hybrid approach that is a combination of pipe size optimizer and layout optimizer is adapted for simultaneous layout and pipe size optimization. Once the solutions (specific layouts) are developed, a modified version of Labye’s iterative discontinuous method (LIDM) is applied to optimize pipe sizes. After the layout optimizer rearranges, the solutions and the process continues. Two versions of a new discrete particle swarm optimization (PSO) are used as tree layout optimizer, namely, original jumping PSO (JPSO) and tree growing JPSO (TG-JPSO). In the original JPSO, a loop model network is considered first, and a spanning tree is produced by using JPSO for eliminating certain branches. Tree growing JPSO without producing a loop model directly deals with branched layout. ...