Evaluating the application of metaheuristic approaches for flood simulation using GIS: A case study of Baitarani river Basin, India. (2022)

• Record Type: Journal Article
• Authors: Samantaray, Sandeep; Das, Sambit Sawan; Sahoo, Abinash; Satapathy, Deba Prakash
• Is Part Of: Materials today. Volume 61:Part 2(2022)
• Page Start: 452
• Page End: 465
• Publication Date: 2022
• Subjects: Flood -- Baitarani river -- GIS -- ANFIS-WOA -- ANN; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.11.561
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Flood is considered one of the major concerns worldwide, resulting in increasing economic losses and mortality rates. Different techniques have been proposed and applied to analyze this natural hazard. In last two decades, data-mining techniques like ANN (artificial neural network) models have been utilised increasingly for flood modelling. The main aim of this research is the development of a flood hazard map considering six flood causative parameters with geographic information system (GIS) and study its effect on prediction of flood magnitude utilising ANN, ANFIS (adaptive neuro-fuzzy Inference system), and hybrid ANFIS-WOA (whale optimisation algorithm). The case study is conducted in Baitarani River basin flowing mostly through Keonjhar district located in northern part of Odisha, India. Six factors, namely slope, elevation, flow length, rainfall, landuse/ land cover (LULC), and distance from rivers, were utilized. For assessing performance of models, three statistical criteria, i.e., R 2 (coefficient of determination), RMSE (root mean square error), and MAE (mean absolute error), are used. Verification outcomes presented acceptable agreement amid actual hydrological records and predicted models. Validation results revealed that hybrid ANFIS-WOA model (with an accuracy of 98.08%) was better at predicting flood than ANFIS and ANN models with 96.41% and 93.05%, respectively. The findings of present study can be utilized for helping national and local government plans for future development and build suitable (to the local environmental conditions) new infrastructures for protecting the life and property of inhabitants in the study area.