Optimization of well flow rate's pumping cost with various metaheuristic algorithms

4th International Civil Engineering & Architecture Conference, Trabzon, Türkiye, 17 - 19 Mayıs 2025, ss.1079-1085, (Tam Metin Bildiri)

The study’s aim is to determine the optimum well coordinates and flow rate values in a well problem,

encountered optimization problem in the management of water resources, with minimum pumping cost through

the application of ten metaheuristic optimization algorithms - Particle Swarm Optimization (PSO), Harmony

Search (HS), Cuckoo Search (CS), Symbiotic Organisms Search (SOS), Teaching Learning Based Optimization

(TLBO), Light Spectrum Optimizer (LSO), Electric Eel Foraging Optimization (EEFO), Geyser Inspired

Algorithm (GEA), Newton-Raphson-Based Optimizer (NRBO), and Puma Optimizer (PO). For this purpose, a

problem suite was developed consisting of two different problems with four existing, two, and three new wells. In

this problem, the objective function is the pumping cost, while the design variables are the coordinates of the two

new wells added for the case with six wells, the coordinates of the three wells added for the case with seven wells,

and the flow rate values in all wells. The optimum solution, which determines the well locations and flow rate

values that minimize the pumping cost of a certain total well flow rate value from an infinite aquifer with two

zones of different transmissivities, was sought by means of these algorithms in this problem. The solutions

obtained from independent simulations conducted on this problem suite were statistically assessed. The

performances of the algorithms were evaluated with Friedman which is a non-parametric statistical test and

Wilcoxon paired comparison test. According to the findings, it was seen that the LSO algorithm had the highest

performance among the ten algorithms examined.