Earth, Atmosphere & Environment Sciences

Title :

Design, Development and Evaluation of Indigenous Sensors Based Air Quality Monitoring System and Data Analysis using Deep Learning

Area of research :

Earth, Atmosphere & Environment Sciences, Engineering Sciences

Focus area :

Pollution Prevention - Clean Technologies and Processes, Cleaner Production, 3Rs, Resource Efficiency, Waste Minimisation and Management, etc.

Principal Investigator :

Dr R Rani Hemamalini, St. Peters Institute of Higher Education and Research, Chennai

Timeline Start Year :


Timeline End Year :


Contact info :


Executive Summary :

Municipal Solid Waste (MSW) generation in Chennai has increased from 600 to 5000 TPD within 20 years. The per capita generation rate is 0.6 kg/day. Although about 269-acre plot in Kodungaiyur, Chennai is used for the garbage dumpsite in Chennai and only about 100 acres of the land is used for haphazard dumping of wastes. Kodungaiyur sewage water treatment plant located adjacent to the dumpsite discharges the sewage water near to the dumpsite. The waste at the site mixes with the sewage water and contaminates it further. It is proposed to Design and Implement a Environment Quality Monitoring device (EQMD) with sensors to monitor environmental quality data. The project involves the use of an Unmanned Aerial Vehicle (UAV) with low cost and high-resolution monitoring of time and space. Monitoring and tracking of the solid waste disposal sites will be done by using a sensor fitted in the UAV which will collect AAQ data. The objective is to also develop an innovative sensor in-house that would measures AAQ not just for dumpsites alone but as a reliable alternate to existing method of using CAAQMS, in industrial areas, air pollution control areas, etc. Pollution measurements of AAQ parameters such as PM 2.5, CO, CO, NO, SO, and VOC. Methane, and met data such as Wind Direction. Wind Speed, Temperature, humidity, and coordinates of locations will be collected and the same will be imported into the GIS environment using ArcGIS software. Attribute data will be assigned to spatial objects and the system becomes ready for Spatio-temporal analysis and management. The collected AAQ data would be transmitted to a server over a wireless internet connection and the server will store, and supply these data to any party who has permission to access it through android phone or website in semi-real time. Environmental quality data analysis using deep learning with EQMD data will be done based on which, the trends and variations of pollution would be analysed. Collected data would be represented in time series or hourly time steps. A prediction model will be developed using neural network. In addition, selected sensors from the market will also be tested for their data reliability accuracy, efficiency, etc and results/data compared with that of the indigenous sensor. The sensor will be tested in simulation using Pspice software, the simulated data and real-time data with in-built sensor will be compared. It is further proposed to validate the data of in-built sensor using linear regression method, and also calibrated from standard calibration centres such as NABL matching sensor data with data of Pollution Control Board. CPCB/SPCB website has a well-defined Protocol for online monitoring system. The MolFCC has given the Council of Scientific & Industrial Research (CSIR) National Physical Laboratory (NPL) with certifying air quality monitoring instruments. This is in anticipation of a rising demand by States against the backdrop of the National Clean Air Campaign for low-cost air quality monitoring instruments that can monitor levels of nitrous sales, Ducnit and particulate matter recently. The Central Government has designated CSIR-NPL as national verification agency for certifying instruments and equipments for monitoring emissions and ambient air. CSIR-NPL shall develop necessary infrastructure, management rate, testing and certification facilities conforming to international standards, according to a notification dated 22 August, 2019. Air quality data from five locations in Chennai (not covering the Dump site area) and Meteorological data for the period from 2010 to 2018 will be collected from TN Pollution Control Board and from the Regional Metrological Department respectively. Among the various monitoring stations of the pollution control board placed in Chennai, Manali is the nearest monitoring station of the study area, hence it is chosen to validate the sensor data. All the data collected by the Pollution Control Board are published daily and freely available on their website. PI has contacted the TNPCB authorities and discussed about the potential use of the research work in air pollution monitoring, particularly of solid waste management facilities. PI has obtained letter dated 27.05.2019 from TNPCB, Chennai.

Objective: i. To design a microchip made out of sensors capable of measuring environmental quality parameters. ii. To develop an efficient algorithm to determine the optimal observation period for accurate air quality prediction. iii. To compare the techno-economic cost of use of this method vis-à-vis established one of CAAQMS for monitoring air pollution.


Dr B. Shanthini, Professor & Head, Department of Information Technology, St. Peter Institute of Higher Education Technology, Avadi, Chennai - 600054

Total Budget (INR):



Output: Development of Indigenous Sensors. Comparison of and analysis of AAQ data of indigenously developed sensors with sensors available in the market and with data from CAAQMS at Manali. Integrated Air Quality Monitoring Device. Continuous monitoring of Air Quality status. Validation of Predicted Ambient air quality data through NABL. Outcome: Environmental quality data for the air pollution control area, industrial area, in and around solid waste dump yards. Pollution variation maps based on the location and time with respect to time. Extents up in which the pollution levels have effects on human settlements. Pollution contour maps for the study area. A report of the concentration of the air pollutants in the study area and comparison of data vis-à-vis data of CAAQMS. Standard output operating procedure of UAV and a Guideline Manual should be prepared for use of this technology.

Organizations involved