A Deep Learning Strategy for Water Quality Monitoring

Water quality is a global concern and monitoring water quality is essential to maintain a healthy ecosystem. Factors such as erosion, industrial waste, and global warming deteriorate the water quality catalysing the water scarcity. Therefore, a quintessential analysis and forecasting model for water quality is the need of the hour. The existing water quality forecasting techniques are awful and futile because of the complex and diversity of the generated data from such systems. Emerging technologies such as deep learning and Internet of Things (IoT) have the potential to explore productive solutions in this domain. A distinctive model for water quality monitoring and forecasting is proposed in this paper. The model proposed in this paper uses Long Short-Term Memory (LSTM) for quality categorization. Parameters such as Total Dissolved Solids (TDS), turbidity, and water temperature are collected. The data collected from the sensors is stored on a server and can be used for analysing the quality of water. Issues related to latency, packet loss in the IoT networks are addressed. An algorithm is developed to classify the water quality based on the samples read from the sensors.