Smart Iot System Based Water Quality Monitoring Prediction Using Sequential Learning Neural Network Strategy

  • Ms. S. Geetha, Dr. P. Venkateswari


Water is one of the essential elements for the existence of life. The safety and accessibility of drinking-water are major concerns throughout the globe. Health risks may arise from consumption of water contaminated with infectious agents, toxic chemicals etc. In this research, a system is proposed to check the water quality and warn the user before it gets contaminated. There are different parameters that can contaminate the water. These parameters are taken into account and used for predicting when to clean the water.  The system uses technologies such as IoT and Machine Learning. It consists of the physical and chemical sensor to measure pH, conductivity in the water, the turbidity in the water, water level in the tank, temperature, and humidity to check the parameters. The data obtained from the sensors are recorded in the database and further sent for analysis. The Sequential Learning Neural Network (SLNN) algorithm is used for predicting the result. Determining the various parameters associated with neural networks is not straight forward and finding the optimal configuration is a time and memory-consuming process. To reduce the time and memory, the Sequential Learning Neural Network system is used for this work. It is used to obtain a non-linear relationship for predicted output. The system sends the alert message to the user when any parameters are lower than the standard values. This helps the user to know beforehand about the contamination of water in their residential tanks. This technique can be limited up to residential tanks and can also be used in water treatment plants and industries.