Effectiveness of Several Metaheuristic Methods to Analyze Hydraulic Parameters in a Drinking Water Distribution Network

Abstract

The reliability and ease of applying metaheuristic methods in solving large and complex equation systems make it interesting to be applied as an alternative solution to solving problems in various fields. This article proves the effectiveness of an optimization model based on the metaheuristic method for the analysis of hydraulic parameters of drinking water distribution pipes. The metaheuristic methods explored are Differential Evolution (DE) algorithm, Particle Swam Optimization (PSO) algorithm and CODEQ algorithm. The effectiveness of the three methods is measured relative by comparing the results of the analysis of the three models with the results from Newton Raphson method and Monte Carlo simulation method. The analysis shows that the optimization model based on the DE, PSO and CODEQ algorithms is very effective for solving problems on a simple network that has 6 pipe elements and 5 service nodes. The results obtained have a level of accuracy as good as Newton Raphson method. In the case of complex networks that have 32 pipe elements and 21 service nodes, there is an indication of performance degradation which is indicated by a decrease in fitness value. In this case, Newton Raphson method still shows its consistency. The optimization model based on the metaheuristic method is still far more effective than the Monte Carlo simulation method, although it is not as effective as Newton Raphson method. The Monte Carlo simulation method is not recommended for hydraulic pipe network analysis, even for simple networks.