Water Quality Diagnosis using IOT and KNN Classifier Integrated with Think Speak Cloud

Main Article Content

Dr.S.Selvi, Dr G Revathy, Dr.N.S.Kavitha & Dr.S.Russia

Abstract

The textile and dyeing industries pave a foremost providing responsibility in our Indian Economy and creates a non discrete monetary progress in nation. However, Water pollution causes a major downfall simultaneously. The discharge of chemical wastages from the industries into the river has a direct impact on the biological organisms. To face this challenge and to prevent the water resources from getting defected, this project is being proposed with a novel solution. This paper provides a multi technological solution, by integrating the Machine Learning techniques with a Cloud-Based Wireless Sensor Network (WSN). This would lead to the dismissal of Waste Water Chemical effluent discharge into the river in an efficient manner. The waste water is monitored using Internet of Things(IoT) sensors for various attributes that include pH, conductivity, dissolved Oxygen-Biological Oxygen Demand(BOD),Chemical Oxygen Demand(COD),Total Dissolved Solid (TDS) Content,etc . In order to monitor pH, conductivity, and dissolved oxygen characteristics from wastewater released into water sources, this suggested paper uses an integrated cloud-based environment. The system uses AT commands in conjunction with the HTTP GET technique to gather and upload sensor data to the ThingSpeak cloud via GPRS internet connectivity in order to deliver an Internet of Things (IOT) based real-time online monitoring. Additionally, the system is supported with an SMS gateway service provided by the Telerivet mobile messaging platform to send message alerts to the responsible organisation.The experimentation has been simulated using Jupyter Notebook for different values of K and the data acquired through IOT sensors are being visualized in live using ThinkSpeak Cloud. The comparison of various classification models were made and its found that KNN provides a betters classification when compared to the other relevant models.

Article Details

Section
Articles