Real-Time Cloud Based Framework For Monitoring And Prediction Analysis Of Acetone In An Ambient Air Using Deep Learning

Authors

  • Shishir A. Bagal, Nitin K. Choudhari, A. R. Chaudhari Author

Keywords:

Volatile Organic Compounds (VOCs), Multi-Sensor Array (MSA), Internet of Things (IoT), ESP-32, Machine Learning (ML).

Abstract

Objective: Acetone is one of the hazardous compounds available in an ambient air causing impact on human health and environmental safety, making its monitoring and analysis more vital. An IoT based, real time framework presented deals with the development of portable and low-cost system for acetone monitoring and analysis via deep learning (DL). Since its a cost-effective, compact system, it can be installed in any industry as compared with the traditional monitoring systems that are highly expensive and difficult to install anywhere. Methodology: This cloud integrated System developed using an ESP-32 IoT Controller and a Multi Sensor Module (ZPHS01B) and deployed in an industry and the real time data samples of the acetone were collected. The data samples were analysed using XGBooster algorithm (XGBOOST) – A deep learning model for the system performance evaluation.

Findings: The system performance evaluated based on the XGBOOST algorithm clearly indicates that the developed system is highly capable of monitoring acetone concentration in an ambient air. The analysis revealed that VOC grade and acetone concentration (0.96) has a very strong correlation. Moreover, Strong Positive correlations Among VOCs i. e. Acetone vs Benzene: 0.95, Acetone vs Toluene: 0.94, Acetone vs Methane: 0.94. Acetone shows that the developed device is highly capable of detecting the Acetone concentration. The weak relations between Acetone Vs Temperature (-0.06) and Acetone Vs Humidity (0.11) indicates that environmental variables do not strongly influence acetone concentration directly.

Novelty: The IoT based real time device developed for monitoring hazardous exhaust compounds is cost-effective, compact and easily deployable at a fraction of the cost of traditional systems. The dataset generated by the system can be useful to the regulating / external agencies for the pollution audit, framing the future policies, ensuring increased workplace safety and regulatory compliance in industrial settings. This device can be employed in industrial settings that can ensure increased workplace safety and regulatory compliance by providing an affordable and scalable environmental monitoring solution.

Downloads

Published

2025-11-24

How to Cite

Real-Time Cloud Based Framework For Monitoring And Prediction Analysis Of Acetone In An Ambient Air Using Deep Learning. (2025). Vascular and Endovascular Review, 8(12s), 347-355. https://verjournal.com/index.php/ver/article/view/964