A Study of Bank Line Shifting of the Selected Reach of Jamuna River Using Multi-Variant Regression Model

Journal of Soft Computing in Civil Engineering

Akramul Haque¹  Md. Matin² 

[1] M.Sc. Student, Department of Water Resources Engineering, BUET, Bangladesh;Professor, Department of Water Resources Engineering, Bangladesh University of Engineering and Technology (BUET), Dhaka-1000, Bangladesh;

关键词: hec-ras;    bank erosion;    google satellite images;    machine learning and deep learning technique;   

DOI  :  10.22115/scce.2022.319360.1387

来源: DOAJ

【 摘 要 】

Jamuna river is a morphologically very dynamic river. It carries a vast sediment load from the erosive foothills of Himalaya mountain. The length of the Jamuna River is 220 km. For this research work Jamalpur district is selected to assess morphological changes using hydrodynamic, Artificial intelligence and google satellite images. First, the hydrodynamic model was calibrated and validated at Kazipur station for the years 2018 and 2019 respectively. Then, left overbank maximum discharge, water level, velocity, the slope was extracted from HEC-RAS 1D at 300 m interval interpolated cross-section. Then, this cross-section was exported as a shapefile. In google earth, the erosion rate was measured corresponding to this interpolated cross-section. The results of the hydrodynamic model were given as input variable and erosion rate as an output variable in Machine learning and deep learning technique. Calibration and validation of the regression model was done for the years 2018 and 2019 respectively. This research work can be helpful to locate the area which are vulnerable to bank erosion.

【 授权许可】

Unknown