This paper is published in Volume-7, Issue-4, 2021
Area
Electronics and Computer
Author
Vinayak Pragada, Romil Keniya, Abhijeet Chauhan, Beena Ballal
Org/Univ
Vidyalankar Institute of Technology, Mumbai, Maharashtra, India
Pub. Date
08 July, 2021
Paper ID
V7I4-1199
Publisher
Keywords
Machine Learning, Arduino, Time Series Forecasting, Gas Sensors

Citationsacebook

IEEE
Vinayak Pragada, Romil Keniya, Abhijeet Chauhan, Beena Ballal. Automated air pollution monitoring and forecasting system using Machine Learning, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Vinayak Pragada, Romil Keniya, Abhijeet Chauhan, Beena Ballal (2021). Automated air pollution monitoring and forecasting system using Machine Learning. International Journal of Advance Research, Ideas and Innovations in Technology, 7(4) www.IJARIIT.com.

MLA
Vinayak Pragada, Romil Keniya, Abhijeet Chauhan, Beena Ballal. "Automated air pollution monitoring and forecasting system using Machine Learning." International Journal of Advance Research, Ideas and Innovations in Technology 7.4 (2021). www.IJARIIT.com.

Abstract

Various Air Quality Monitoring systems have been recently developed and installed in some of the cities such as Mumbai, Bangalore, Delhi, Hyderabad, etc. But the actual results do not satisfy enough. Several issues in the existing system were the number of gases detected were very few. There was a need for such a system that monitors more amounts of gases in less amount of time. Time was also an issue regarding the monitoring of the gases. Some gases consumed an enormous amount of time, so it became crucial to decrease them and make a more efficient system. The proposed system is automated air pollution monitoring and forecasting system using machine learning which uses an Arduino Mega, MQ gas sensors module for its development along with some other basic components like16x2LCD, Buzzer, Potentiometer. MQ135 gas sensor detects gases like NH3, NOx, alcohol, benzene, smoke, CO2, SO2, etc. which are the main reason for degrading air quality. Wi-Fi module is used to connect our system to the Cloud so that all the readings and the data can be transferred to the desired server for forecasting purposes. A trained model is expected to use the ARIMA algorithm for the best prediction purposes. This model is then applied to the upcoming data and forecasting is done. The various existing system was learned which are developed till date, it was found that the air quality numerical model such as WRF-Chem, community Multi-scale air quality model (CMAQ), CAMx, NAQPMS was used. The drawbacks in these models were found like, Source list was not updated in time for WRF-Chem, Detailed information about the source of the pollutants, and other variables are generally not known. After knowing all the facts from the study, this system is expected to use a Scikit machine learning tool for forecasting. The proposed model monitors the upcoming data in no time and if the value goes beyond the threshold value it activates the buzzer and the server is notified.