Journal of Materials and Engineering Structures
Volume 13, Numéro 3, Pages 505-516
2026-09-30
Modeling Non-linear Time Of Concentration In Heterogeneous Catchments: A Comparative Study Of Svr And Ann Machine Learning Approaches
Authors : Suljic Nedim . Prskalo Gordan .
Abstract
The main objective of this research is to improve the accuracy of estimating the precipitation accumulation time (tc) in catchments with heterogeneous morphological and hydrological characteristics. The work aims to address the unreliability of traditional empirical models that often ignore the nonlinearity of the runoff process and do not take into account variable parameters such as previous soil moisture content (AMC) and precipitation intensity. The research is based on a comparative analysis of traditional methods (Kirpich, Kinematic Wave) and modern artificial intelligence models. Machine learning algorithms were used: Support Vector Machines (SVR) and Artificial Neural Networks (ANN). The models were trained and tested on datasets that include parameters such as terrain slope, stream length, surface roughness, and variable hydrological factors. Validation was conducted on the Sava River sub-basins, using statistical accuracy indicators. The results of the study show that traditional methods significantly underestimate the accumulation time under conditions of nonlinear hydrological processes, which can lead to erroneous estimates of peak flows. It has been found that soil moisture and terrain saturation dramatically alter the dynamic response of the basin, which fixed empirical models cannot adequately quantify. The scientific contribution of the work is manifested in the introduction of a dynamic approach to determining the concentration time, which abandons the static concept of traditional formulas.
Keywords
Rainfall accumulation time ; Concentration flow ; Machine learning ; Hydrology
Les articles similaires
Chu Tien Dung
.
Nguyen Quang Phuc
.
Luong Xuan Chieu
.
pages 253-265.
عينوس رضوان
.
ص 333-352.
Chemingui Fatima Zohra
.
Benrachi Fatima
.
pages 51-56.
Mameri Y
.
Debbache N
.
Seraghni N
.
Sehili T
.
pages 83-89.
