DONATE

IAHR Document Library


« Back to Library Homepage « Proceedings of the 31st IAHR World Congress (Seoul, 2005)

Influence of Water Quality Forecasting Due to Realtime Monitoring Data Handling

Author(s): In-Sung Yeon; Sang-Jin Ahn; Gyu-Bang Yeon; Ung-Yong Kim

Linked Author(s):

Keywords: Modular neural network; Water quality; Smoothing; Real-time forecasting; GUI

Abstract: Neural network model is used to research the water quality and runoff forecasting at river and lakes (Maier and Dandy, 1996; Zhang and Govindaraju, 1998). Modular neural network shows better results when using time factor, which is qualitative data trained with quantitative data. The real time data exchange to the hourly data by data edit process for the analysis with discharge hourly data. Data edit process, in order of preference, was done median, 1st smoothing, 2nd smoothing. The neural network model trained the hourly data and forecast water quality due to data edit process. Using the smoothing was better than using the median generated by raw data, and the best result was acquired by using the 2nd smoothing. In order for the operational convenience, GUI was constructed using MATLAB. The main page is designed to control the connected data processing model, water quality forecasting model, and warning model. We were easily able to perform the forecasting process through the water quality forecasting system implemented by GUI.

DOI:

Year: 2005

Copyright © 2024 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions