Appropriate dehumidification systems are essential for passive houses to prevent thermal discomfort under high dehumidification loads. In this study, an integrated air conditioning system (IACS) consisting of a dedicated outdoor air system and sensible cooling unit is developed to manage dehumidification loads in passive houses. Multi-objective optimization of the operating conditions of the IACS is performed considering the energy consumption and thermal comfort of a passive house. A thermal comfort model is constructed using a big-data-driven machine learning method to improve prediction accuracy compared to a conventional model. Additionally, an artificial-neural-network-based metamodel is developed to estimate the performance of the IACS. The optimal operating conditions are determined based on the Pareto solution set obtained from multi-objective optimization. The optimized performance and economic feasibility of the IACS are compared to those of a conventional air conditioning system (CACS). For the IACS in a passive house, the optimal indoor set-point temperature and relative humidity were determined to be 27.01 ℃ and 50.35%, respectively. The energy-saving rates of the IACS were 15.96% and 7.53% in passive and normal houses, respectively, compared to the CACS. The IACS is more economical than the CACS based on its energy-saving effects, particularly in passive houses.
Bibliographical noteFunding Information:
This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20208901010010 ).
© 2023 Elsevier B.V.
- Dedicated outdoor air system
- Integrated air conditioning system
- Machine learning
- Multi-objective optimization
- Passive house
- Zero-energy building
ASJC Scopus subject areas
- Civil and Structural Engineering
- Building and Construction
- Mechanical Engineering
- Electrical and Electronic Engineering