1. Water Quality Prediction Based on Data Mining and LSTM Neural Network
- Author
-
Liqin Zhao, Senjun Huang, and Jiange Jiao
- Subjects
Water resources ,Artificial neural network ,Environmental engineering ,Environmental science ,Grey correlation analysis ,Economic shortage ,Water quality ,Turbidity ,Water pollution ,Grey relational analysis - Abstract
Water pollution exacerbates water shortages affecting human health and quality of life. Water quality prediction is of great significance in the future water quality management. In this thesis, the internal relations among dissolved oxygen, temperature, pH and turbidity were revealed by using grey correlation analysis method. Furthermore, the LSTM neural network was used to predict dissolved oxygen in water. The results showed that dissolved oxygen is closely related to temperature and pH. Temperature and dissolved oxygen are negatively correlated, and pH is positively correlated with dissolved oxygen. Dissolved oxygen, which affects the key indicators of water quality, has a good prediction effect, with an accuracy of more than 90%. The research results provided valuable references in water pollution control and water resources management.
- Published
- 2021
- Full Text
- View/download PDF