International Journal of Science, Engineering and TechnologyAuthors: Mr. Prabhu M, Sanjeev Kumar. K, T.S. RameshBabu B.E
Abstract: This paper aims to assess the application of Artificial Intelligence (AI) in the development of stage-discharge relationships. Stage-Discharge Relationships often known as rating curves are a crucial tool for hydrological modelling and water resource management. This study uses AI, specifically Large Language Models (LLMs) in the development, calibration, and validation of stage-discharge curves. The methodology involves utilizing LLMs to generate python code. The hydrological data i.e stage and discharge from September 1971 to May 2021 (30 years) at Biligundulu gauging station on the Cauvery River was used to derive SD Curve. The developed equation was then applied to hydrological data from June 2021 to May 2023. The predicted discharge values were subsequently compared to the actual observed values. The outcomes are evaluated using R-squared, Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), demonstrating a strong fit to the observed data (R² = 0.9995, RMSE = 17.29 m³/s, MAE = 10.91 m³/s). The process and results indicate that AI-driven approaches can offer a robust alternative to conventional methods. This study demonstrates AI can empower hydrologists without extensive coding expertise to conduct complex data analysis. By leveraging LLMs, complex hydrological models, data pre-processing, etc., can be automated, enabling more researchers and practitioners to conduct advanced analyses.
