Content area
Full text
Key Words:
loT MQ-7 MQ-131 ThingSpeak Machine learning Neural network
ABSTRACT
Air quality is a vital concern globally, and Sri Lanka, according to WHO statistics, faces challenges in achieving optimal air quality levels. To address this, we introduced an innovative loT-based Air Pollution Monitoring (ARM) Box. This solution incorporates readily available Commercial Off-The-Shelf (COTS) sensors, specifically MQ-7 and MQ-131, for measuring concentrations of Carbon Monoxide (CO) and Ozone (03) ,Arduino and "ThingSpeak" platform. Yet, those COTS sensors are not factory-calibrated. Therefore, we implemented machine learning algorithms, including linear regression and deep neural network models, to enhance the accuracy of CO and 03 concentration measurements from these non-calibrated sensors. Our findings indicate promising correlations when dealing with MQ-7 and MQ-131 measurements after removing outliers.
(ProQuest: ... denotes formulae omitted.)
INTRODUCTION
Over time, the Earth's atmosphere has undergone changes, influenced by both natural events and human activities. Unfortunately, these alterations have led to an increase in air pollution, impacting humans and plant life negatively. The concerning shift is gradually making the Earth's atmosphere less conducive to the well-being of both humans and other living organisms (Choudhary & Garg, 2013). Addressing these challenges is crucial for a more optimistic environmental future.
As air pollution is a common problem that affects almost all the countries in the world, continuously measuring air pollutants keeps track of the well-being of the public, animals and plants, etc. Usually, economically wellestablished countries are concerned with measuring air quality to obtain sustainable goals with highly accurate real-time or conventional air quality monitoring systems. In Sri Lanka, air quality is mostly monitored by the Central Environmental Authority (CEA) and the National Building Research Organization (NBRO) using conventional chemical methods and the Mobile Ambient Air Quality Monitoring Lab (MAAQML).
It is possible to reduce air pollution by studying the changes in the composition of different types of gasses in the air and taking appropriate measurements using conventional air quality measuring equipment (Yi et al. 2015). However, due to their high cost, low and middle-income countries tend to use cost-effective sensors to measure air pollution and implement loT devices (Yi et al. 2015). The air quality of a particular area can be monitored using sensors (gaseous and meteorological) and Arduino/Raspberry Pi (Malleswari & Mohana 2022). In...