AI-VENT: AI-Based Monitoring and Control of Industrial Ventilation Systems

Abstract

Industrial ventilation systems are essential for controlling airborne dust and particulate matter, yet conventional fixed-speed systems often respond poorly to dynamic particle transport and changing exposure conditions. This study proposes AI-VENT: AI-Based Monitoring and Control of Industrial Ventilation Systems, an intelligent framework for real-time dust monitoring, particulate prediction, and adaptive airflow regulation. The proposed system integrates PM1.0, PM2.5, PM4.0, PM10, temperature, humidity, air velocity, and pressure sensors with machine learning-based prediction and control logic. In the simulation-based evaluation, AI-VENT achieved a mean absolute error of 0.86 µg/m³ and root mean square error of 1.21 µg/m³ for particulate concentration prediction. Compared with fixed-speed ventilation control, AI-VENT reduced peak dust accumulation by 34.7%, improved ventilation response efficiency by 29.4%, and lowered unnecessary fan energy consumption by 18.6%. These results indicate that AI-VENT can support proactive exposure reduction, safer industrial workplaces, and energy-efficient ventilation management.