https://doi.org/10.6110/KJACR.2025.37.8.363

Seung Jae Lee ; Sang Woo Kim ; Cheong Ha Hwang ; Hyun Jung Lee ; Chengxi Yao ; Kwang Chul Noh ; Tae Sung Kim

This study evaluated the performance of air purification devices (ventilation devices and air purifiers) in real-world classroom environments to establish operational standards. Air purifiers demonstrated superior performance in reducing particulate matter (PM), while ventilation systems were more effective in removing carbon dioxide (CO2). The performance of these systems was influenced by outdoor air quality, and an optimal ventilation rate was identified. Increasing the number of devices or airflow rates did not proportionally enhance air quality control. For installation, placing air purifiers along the classroom walls (CADR 35.3 CMM) reduced direct airflow interference and energy loss compared to diagonal placements (CADR 25.3 CMM), resulting in better PM removal. However, results varied based on airflow direction, highlighting the need for optimized placement in combined operations for each condition and environment. The ventilation device achieved CO2 reduction close to healthy standards, outperforming air purifiers. Additionally, the optimal arrangement (diagonal direction) for simultaneously operating two types of air purifiers to remove fine dust and CO2 was established.

https://doi.org/10.6110/KJACR.2025.37.8.370

Kwangwon Choi ; Jaewan Joe ; Sangwoo Ha ; Dongyun Lee ; Jungsoo Mun

This study presents a parameter simplification method for a grey-box model based on a radiant floor heating system in residential buildings. The grey-box model requires numerous parameters to accurately reflect the physical and thermal characteristics of buildings, resulting in complex modeling and increased computational costs. To address this challenge, the study develops a method to simplify the model structure, thereby improving computational efficiency while maintaining physical interpretation. The existing grey-box model consists of 24 thermal resistances and 17 thermal capacities. In contrast, the proposed simplified models, 24R-8C and 16R-6C, reduce the number of parameters from 24R-17C to 24R-8C and 16R-6C, respectively, enhancing computational efficiency. This simplification minimizes costs while delivering satisfactory prediction performance. The average root mean square error (RMSE) for indoor air temperature and floor surface temperature in the base model (24R-17C) was approximately 0.51℃. The two simplified models, 24R-8C and 16R-6C, demonstrated RMSE values of approximately 0.51℃ and 0.63℃, respectively. The outcomes of this study are expected to contribute to practical energy-saving effects when the proposed models are applied to real-time heating control via model predictive control (MPC).

https://doi.org/10.6110/KJACR.2025.37.8.384

Deokgeun Kim ; Ajin Jo ; Jaeman Song ; Hiki Hong

Research on predicting solar thermal systems has primarily utilized artificial neural networks (ANNs), which heavily rely on numerical weather prediction (NWP) data. However, ANN models struggle to capture temporal dependencies in time series data, and discrepancies between NWP data and actual observations can reduce prediction accuracy. This study proposes a long short-term memory (LSTM)-based prediction model that integrates NWP, numerical weather observation (NWO), and numerical system observation (NSO) data. The model predicts three output variables and was developed into three configurations: LSTM 1, LSTM 2, and LSTM 3, each based on different input setups. LSTM 3 outperformed LSTM 1 in predicting supplied energy, reducing the mean absolute error (MAE) and root mean square error (RMSE) by 28.2% and 25.3%, respectively, while improving the R² value by 34.5%. For predicting the top storage tank temperature, LSTM 3 also achieved reductions in MAE and RMSE of 27.3% and 23.5%, and improved R² by 12.2%. In terms of the prediction of acquired energy, solar irradiation was the most significant variable, with an importance score of 0.068, followed closely by supplied energy at 0.067 and top storage tank temperature at 0.057. These results indicate that the combined influence of environmental and system variables is crucial for enhancing prediction accuracy.

https://doi.org/10.6110/KJACR.2025.37.8.395

Sangheon Jeong ; Soowon Chae ; Jinhwan Oh ; Hobyung Chae ; Yujin Nam

This study examines the use of an AI-based optimal flow control model to improve the energy efficiency of heating and cooling systems in modern buildings. Traditional flow control methods, such as On/Off control, are unable to adapt dynamically to changing environmental conditions, resulting in excessive energy consumption and inconsistent performance. To overcome this limitation, we developed and tested an AI-driven control system on a PVT-Integrated Air Source Heat Pump (PVT-ASHP) hybrid system in Busan, South Korea. Our real-scale experiments compared the performance of the AI-based control with conventional methods, focusing on key performance indicators such as cooling coefficient of performance (COP), energy savings, and system stability. The results showed that the AI-based model effectively optimized flow rates based on real-time environmental data, leading to a 12% reduction in energy consumption and a 14% improvement in cooling COP compared to traditional methods. These findings underscore the potential of AI-driven control strategies to enhance energy efficiency in buildings, contributing to the goals of carbon neutrality and zero-energy structures. Future research will aim to expand the adaptability of the AI model to various climate conditions.

https://doi.org/10.6110/KJACR.2025.37.8.403

Joon Ahn ; Hyouck Ju Kim

This study utilizes polyethylene (PE) pellets, derived from waste plastics, as fuel in an industrial boiler. A novel burner system was developed that incorporates pyrolysis in a pre-combustion chamber, followed by combustion in the main chamber. This approach enhances combustion control and efficiency compared to traditional grate-based systems. The burner was applied to an industrial boiler that consumes a significant portion of industrial energy in South Korea. Combustion characteristics, including flame images, smoke scale measurements, and concentrations of carbon monoxide (CO) and nitrogen oxides (NOx), were experimentally assessed. The results suggest that using PE pellet fuel combined with pre-combustion provides a sustainable and efficient solution for converting waste plastic into energy while reducing harmful emissions.