Author(s): Minhyeok Lee; Seoeun Kwak; Yunhwan Kim; Moon Jeong; Meeyoung Park; Yong-Gyun Park

Linked Author(s):

Keywords: Optimal water intake layer; Drinking water treatment; Water quality; Deep learning; LSTM

Abstract: Managing the quality of drinking water sources is essential to ensure the high operational and treatment efficiency of water treatment plants. Poor water quality in drinking water sources can lead to overloading of water treatment plants and high treatment costs. In this study, we utilized LSTM (Long Short-Term Memory), a deep learning algorithm suitable for analyzing and predicting time series data, to predict the seasonal and annual water quality of Juam Lake in South Korea and select the optimal water intake point. To ensure the high performance of the deep learning model, we used water quality network data from Lake Juam, South Korea, collected between January 2013 and June 2023. Statistical techniques were applied to the water quality measurement network data to more accurately analyze the influence of water quality factors. The results of this study suggest that deep learning-based algorithms can improve the treatment and operational efficiency of water treatment plants and can be used to monitor the performance of water intake and treatment facilities.

DOI: https://doi.org/10.3850/iahr-hic2483430201-42

Year: 2024