ANN-Driven Optimization of VOC Adsorption on Activated Carbon with Thermal Breakthrough Forecasting and IoT-Based Real-Time Monitoring

Abstract

The Volatile Organic Compounds (VOCs) have become one of the drivers of environmental deterioration and occupational hazard, and the issue requires competent and clever adsorption methods of eliminating the same. This paper proposes a comprehensive experimental, computational and IoT-based experimental system that will maximize VOCs adsorption to activated carbon. A packed bed adsorption column was constructed and equipped with two MQ-138 and DHT22 sensors and could be directly tracked through a NodeMCU-ThingSpeak dashboard in real-time. During experiments the efficiency of VOC removal was lower at higher inlet concentration (92.3% to 76.1% at 100 ppm to 300 ppm respectively) and higher at optimized flowrate (74.5% - 89.8% at 3.0 to 1.5 L/min, respectively). The efficiency was lower at high relative humidity, because of competitive adsorption, and higher bed temperatures (up to 45°C) prolonging breakthrough time slightly. The model used was a 4-8-1 ANN whose training was carried out using Levenberg Marquardt algorithm, which had a high predictive accuracy (R²=0.987,RMSE=1.82) the experimental value being close to the computed values across a range of inputs. 3D surface mapping of ANN mode exhibited an ideal area of interaction between VOC concentration and flowrate. Also all IoT delays were less than 1.5 seconds and the sensor offset was less than ±5 ppm and ±0.5°C and thus confirming the readiness of the system deployment. The above outcomes confirm that it is possible to implement intelligent, responsive VOC mitigation tools that are informed by machine learning and integration with IoT in managing air quality in industry.