WATER LEVEL PREDICTION AT CHAU DOC MONITORING STATION, AN GIANG PROVINCE

Tóm tắt

In the Mekong Delta, flooding is an annual natural phenomenon that occurs on a large scale, affecting human life, agriculture, infrastructure, the economy, and society. On the other hand, during the dry season, water levels drop, leading to water shortages and drought. This study investigates the application of machine learning to predict water levels at Chau Doc monitoring station in An Giang province. Specifically, this study uses LSTM algorithms combined with random forests to predict water levels. The model was trained using water level data collected from 2019 to 2022 and is designed to predict levels until 2031. The observed data was used to implement the model and assess its accuracy, efficiency, and reliability. The results show that  the random models achieved a Nash-Sutcliffe Efficiency (NSE) greater than 0.7 and a Root Mean Square Error (RMSE) between approximately 5 and  10 cm, showing that the model is both reliable and effective in simulating water levels. Local water quality management authorities have applied various measures, and early predictions of water level changes in the river help management agencies respond proactively to upcoming situations related to the water environment in the area. Moreover, this research serves as a useful reference for further studies and in management of water quality.

Từ khóa: Water Level, Water Level Prediction, Machine Learning, LSTM, Chau Doc Monitoring Station

Tải bài báo