Application of Self-adaptive Vision-Correction Algorithm for Water-Distribution Problem

Abstract

Water is one of the essential elements in human life, and the water-distribution system is an important infrastructure that distributes water to the people. The design of the water-distribution system was done manually in the past, but recently, it was conducted by applying optimization algorithms. Various optimization algorithms were developed for the optimal design of the water-distribution system, but there was a disadvantage to perform sensitivity analysis for parameter estimation manually. A vision-correction algorithm (VCA) was developed to emulate the conventional manual vision-correction procedure, and it exhibited a good performance in many mathematical benchmark and civil engineering problems. However, the VCA has limited usefulness because of its large number of parameters, many of which require estimation. In this study, a self-adaptive VCA (SAVCA) was developed to overcome these shortcomings by modifying the parameters of the VCA to be self-adaptive or fixed. The Balerma network — a water-distribution system — was selected as a civil engineering problem. The results of SAVCA were better than those of other methods for the design of Balerma network. The SAVCA exhibited good usability and performance and can be applied to various fields in civil engineering including the water-distribution system.