Mobile QR Code QR CODE : Korean Journal of Air-Conditioning and Refrigeration Engineering
Korean Journal of Air-Conditioning and Refrigeration Engineering

Korean Journal of Air-Conditioning and Refrigeration Engineering

ISO Journal TitleKorean J. Air-Cond. Refrig. Eng.
  • Open Access, Monthly
Open Access Monthly
  • ISSN : 1229-6422 (Print)
  • ISSN : 2465-7611 (Online)
Title Search Characteristics and Scalability of a QAOA-Inspired Exploration Framework for Time-Coupled Discrete Scheduling in Building Energy Systems
Authors Byungki Jeon ; JaeHyeok Heo ; HongJin Joo ; DeukWon Kim ; JongKyu Kim
DOI https://doi.org/10.6110/KJACR.2026.38.7.375
Page pp.375-392
ISSN 1229-6422
Keywords 건물 에너지 운영; 이산 스케줄링; 물리 시뮬레이션 기반 최적화; 양자 근사 최적화 알고리즘; 양자 영감 알고리즘 Building energy operation; Discrete scheduling; Physical simulation-based optimization; Quantum approximate optimization algorithm; Quantum-inspired algorithms
Abstract This study formulates the daily operation of building cooling systems as a time-coupled discrete scheduling problem to evaluate the applicability of the Quantum Approximate Optimization Algorithm (QAOA). The optimization objective is to minimize electricity costs while satisfying thermal comfort constraints during occupied hours, with hourly heat pump output levels defined as upper-level control variables. Initially, a baseline single-zone problem is examined, demonstrating that grid search, particle swarm optimization (PSO), and QAOA all converge to the same optimal solution. Subsequently, extended scenarios are introduced by gradually increasing the number of zones and thermal energy storage capacity. This allows for a comparison of the exploration characteristics of the algorithms under increasing physical system complexity. Simulation results indicate that, in these extended problems, PSO tends to concentrate search outcomes within limited regions of the cost?comfort space. In contrast, QAOA consistently identifies Pareto-optimal solutions across a broader cost range. These findings suggest that QAOA can be an effective alternative for discrete scheduling problems in building energy management, especially in situations with limited data dependency.