نوع مقاله : مقاله پژوهشی
نویسندگان
1 Department of Physics, Faculty of Physical sciences, Ambrose Alli University, Ekpoma Edo State, Nigeria
2 Department of Applied Geophysics, Federal University of Technology, Akure, Nigeria
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Groundwater is a significant driver of water supply considering its constant accessibility, intrinsic quality, and ease of immediate diversion to disadvantaged areas. The bulk of the Ogbomoso population relies on surface water due to normative groundwater investigation, which is a laborious and resource-intensive process. Over the last decade, the use of an adaptive neuro-fuzzy inference system (ANFIS) has garnered enthusiastic acceptance in a variety of research domains. This study aimed to evaluate groundwater potential in Ogbomoso, Nigeria, using ANFIS in a geographic information system coupled with three metaheuristic optimizing algorithms: the genetic algorithm (GA), particle swarm optimization (PSO), and the firefly algorithm (FA). To facilitate groundwater potential mapping (GPM) in the research area, a database of 165 wells with 8 predictive parameters was created. Thirty percent (49) of the 165 well locations were designated for the validation set, while the remaining seventy percent (116) were designated for the training set. The slope, lineament density, lithology, overburden thickness, bedrock relief, coefficient of anisotropy, hydraulic conductivity, and transmissivity were the eight (8) groundwater variable parameters generated for modeling. The findings showed that each of the models had good prediction capability; nonetheless, the ANFIS-GA has the strongest predicting effectiveness, having a correlation level of r = 0.8 (80%), followed by both the ANFIS-PSO and the ANFIS-FA with r = 0.77 (77%). The groundwater potential index developed for the study region was zoned into low (0.77–1.63 (30%), moderate (1.63–1.74) (25%), and high 1.74–2.50 (45%) using the ANFIS improved by the GA method. The linear correlation method was used to validate the model using 110 water columns from wells in the study area. The findings of this study demonstrate that ANFIS models paired with metaheuristic algorithmic optimization can be an invaluable tool for making decisions for groundwater utilization and monitoring. To facilitate groundwater potential mapping (GPM) in the research area, a database of 165 wells with 8 predictive parameters was created. Thirty percent (49) of the 165 well locations were designated for the validation set, while the remaining seventy percent (116) were designated for the training set. The slope, lineament density, lithology, overburden thickness, bedrock relief, coefficient of anisotropy, hydraulic conductivity, and transmissivity were the eight (8) groundwater variable parameters generated for modeling. The findings showed that each of the models had good prediction capability; nonetheless, the ANFIS-GA has the strongest predicting effectiveness, having a correlation level of r = 0.8 (80%), followed by both the ANFIS-PSO and the ANFIS-FA with r = 0.77 (77%). The groundwater potential index developed for the study region was zoned into low (0.77–1.63 (30%), moderate (1.63–1.74) (25%), and high 1.74–2.50 (45%) using the ANFIS improved by the GA method. The linear correlation method was used to validate the model using 110 water columns from wells in the study area. The findings of this study demonstrate that ANFIS models paired with metaheuristic algorithmic optimization can be an invaluable tool for making decisions for groundwater utilization and monitoring.
کلیدواژهها [English]
)