Urban Drainage Networks Rehabilitation Using Multi-Objective Model and Search Space Reduction Methodology

摘要: The drainage network always needs to adapt environmental and climatic conditions provide best quality services. Rehabilitation combining pipes substitution storm tanks installation appears be a good solution overcome this problem. Unfortunately, the calculation time of such rehabilitation scenario is too elevated for single-objective multi-objective optimization. In study, methodology composed by search space reduction whose purpose decrease number decision variables problem solve optimization optimize process represent Pareto fronts as result urban networks proposed. A comparison between different model results made. To obtain these results, Storm Water Management Model (SWMM) first connected Pseudo Genetic Algorithm (PGA) then Non-Dominated Sorting II (NSGA-II) are designed investment costs instead flood damage costs. applied real in city Medellin Colombia. show that provides models with considerably reduced variables. shows models’ used after better outcomes than complete terms optimality solutions.