Master of Science in Civil Engineering (MSCE)
Civil and Environmental Engineering
The City of Chandler, Arizona lies in semi-arid area and the water resources management has been a major concern by the State citizen. There is always a need of optimizing the available water resources when the rate of demand constantly beats the rate of replenishment. In this study we have developed a management model to link the Chandler water distribution model (EPANET) with the Chandler groundwater flow model (MODFLOW) under an optimization framework (Genetic algorithm). The EPANET model supplied by the Chandler city municipal authority has several improper setting (pumping schedules, valves settings, pressure zones etc.). The original model has several warnings about system being disconnected and unbalanced since it lacks efficient set of constraints. The pumping values present in the well package for the Chandler water planning area do not commensurate with their corresponding values in the EPANET model. In the study when the models are linked it is made sure that the sub-surface and surface models have a proper connection and accurate data transfer, this is the main objective for linking the two models. The overriding goal of the management model is to optimize the well pump operations such that the pump energy cost is minimized. The optimization problem was solved using a genetic algorithm (GA) to search for the optimal 24-hour real-time pumping patterns for all well pumps while conducting 7-day EPANET and MODFLOW simulations. The main optimization function has several objectives like energy, pressure violation, drawdown, and reliability etc., each having a weight or penalty (w or r) in the function. These weights are first determined by trial-and-error approach. Using Parallel Genetic Algorithm (PGA) we have managed to determine the optimized weights and penalties for the function by running several simulations in a very short time, which was not possible otherwise with a normal PC. In the later part of the study we have also developed a user interface in EXCEL to display the optimization results for different cases, with the weights and penalties as user input from EXCEL worksheet.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Katiyar, Vineet, "Production well operations optimization in water distribution system using genetic algorithm" (2007). LSU Master's Theses. 3225.
Frank T-C. Tsai