Performance Analysis of an Outdoor Air-Introduced Ventilation Cleaner Applied with an AI based Indoor PM<sub>2.5</sub> and CO₂ Integrated Control Algorithm

Amid growing concerns about health risks, the demand for safer indoor air quality has significantly increased, necessitating advanced control solutions. Previous studies focused on ventilation, air purification control, and the proactive management of indoor air pollutants. However, most of these studies relied on computer simulations, with few applications in real-world environments. In response, this study developed an AI-based integrated control algorithm for indoor PMsub2.5/sub and CO₂ using empirical data. The algorithm incorporates prediction models for indoor PMsub2.5/sub and CO2 based on artificial neural networks, enabling proactive control. It was applied in a living lab to validate its effectiveness in improving indoor air quality and reducing energy consumption. The integrated control algorithm demonstrated an average reduction of 19.69% in indoor PMsub2.5/sub concentrations and 6.59% in CO₂ concentrations compared to the existing rule-based algorithm, while also decreasing energy consumption by 9.42%. Furthermore, the algorithm maintained indoor PMsub2.5/sub and CO₂ concentrations below established criteria for a longer duration, confirming its superior performance. The integrated control algorithm developed in this study effectively achieved both indoor air quality management and energy consumption reduction, contributing to the sustainable development of building environment management and control.