Development of optimal parameter determination algorithm for two-dimensional flow analysis model

• Published in: Environmental modelling & software : with environment data news Vol. 185; p. 106331
• Main Authors: Shin, Eun Taek, An, Se Hyuck, Park, Sung Won, Lee, Seung Oh, Song, Chang Geun
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.02.2025
• Subjects: algorithms; Automated parameter optimization algorithm; automation; computer software; Eddy viscosity; fluid mechanics; prediction; roughness; Roughness coefficient; Two-dimensional flow model; viscosity; Automated parameter optimization algorithm; Two-dimensional flow model; Eddy viscosity; Roughness coefficient
• ISSN: 1364-8152
• DOI: 10.1016/j.envsoft.2025.106331

Accurate parameter selection is crucial for reliable predictions in fluid dynamics, environmental transport, and urban flood prediction. Traditional manual methods are time-consuming and prone to errors. This study introduces an automated algorithm to optimize roughness and viscosity coefficients in two-dimensional flow analysis models. Our algorithm automates the simulation process within specified parameter ranges, using Root Mean Square Error (RMSE) to compare results with experimental data. Applied to a diverging channel and an abruptly widening channel, the algorithm successfully identified optimal parameters, accurately matching experimental observations. Heatmaps visualize RMSE values, facilitating optimal parameter identification. This advancement enhances model efficiency and accuracy, streamlining the parameter determination process and offering a robust method for hydraulic modeling. •Automated Parameter Optimization: Developed an algorithm to automatically optimize roughness and viscosity coefficients in 2D flow models, enhancing simulation accuracy and efficiency.•Significant Improvement in Accuracy: The new algorithm improves prediction accuracy by up to 30% compared to traditional methods, effectively capturing complex flow dynamics.•Reduced Manual Effort: The algorithm minimizes the need for manual iterative simulations, streamlining the modeling process and reducing laborious effort in environmental and hydraulic studies.