A New Discrete Differential Evolution Algorithm Coupled With Simulation–optimization Model for Groundwater Management Problems

Abstract

Discrete differential evolution (DDE) is a promising algorithm specifically developed to solve discrete problems. In this study, we aim to apply DDE to groundwater management problems and to compare its performance and discrete space search capabilities with the well-known genetic algorithm (GA) techniques. Local search process was used to enhance the performance of GA algorithm. Metaheuristic algorithms are used for finding location of wells as a hybrid optimization procedure. Two examples from the groundwater management literature were selected to test the performance of the algorithm. The main novelty and objective of this study lie in the comparison of the discrete space search capabilities of the mentioned metaheuristics algorithms using the groundwater management problems. In the first test example, discrete space search performances of algorithms are 15% and 93% for GA and DDE, respectively. In the second test example, DDE exhibited a significantly higher test results (77%) compared to GA (1%). The analysis revealed that GA often prematurely converged and was insufficient to produce the optimum result. DDE reaches the solution considerably faster than the other algorithms. The results showed the superior performance of DDE in the discrete space. As the problem becomes more discrete, the performance of the DDE algorithm in finding the optimum solution increases considerably. Thus, it can be revealed that DDE can also be applied to a wider range of water resource management problems as an effective discrete optimization algorithm. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.