To address the core bottlenecks such as low lighting energy efficiency and poor scenario adaptability in integrated passenger hubs during high-load periods, while balancing the dual requirements of green operation and passenger comfort, a lighting energy efficiency improvement method in hubs driven by improved multi-objective optimization is proposed. First, taking a large integrated passenger hub in Guangzhou as the research object, 2 160 valid samples covering 12 typical scenarios from 2024 to 2025 are collected. The least squares method is employed to reveal the nonlinear correlation and threshold effect between passenger flow and lighting demand. Second, a “time-space-load-event” four-dimensional coupling model is constructed. To address the low search efficiency of the original NSGA-III under high-dimensional constraints, an improved algorithm is designed, in which a four-dimensional correction factor is introduced to dynamically adjust crossover and mutation strategies, thereby enhancing the algorithm’s adaptability to time-varying demand. Additionally, a constraint handling mechanism combining feasibility rules and penalty functions is established to improve the diversity of the Pareto front. Finally, a multi-objective fitness function is built with the objectives of minimizing energy consumption, maximizing comfort, and maximizing equipment service life, and an adaptive closed-loop control system is implemented in conjunction with the Digital Addressable Lighting Interface (DALI) protocol. The results show that lighting demand grows exponentially when passenger flow exceeds 120% of the designed capacity. Compared with the traditional timed control scheme, the proposed method reduces the average daily energy consumption from 592.1 kWh to 228.7 kWh (a reduction of 61.37%), increases the passenger lighting comfort score from 45.4 to 58.1 (an improvement of 27.97%), and shortens the system response delay from 1.2 s to 0.38 s. Furthermore, it extends the lighting equipment service life by 23.50%, and stably adapts to passenger flow fluctuations within ±30%. Compared with PSO and GA algorithms, the comprehensive optimization index of the proposed method is improved by 23.53% and 16.67%, respectively, and the average daily lamp failure rate is reduced from 3.2 times · d-1 under the traditional control scheme to 0.8 times · d-1, with a decrease of 75%. Validated by actual operation and maintenance data from the Guangzhou hub, the retrofit achieves annual electricity savings of approximately 132 600 kWh, with an overall electricity saving rate of 61.4%, corresponding to an annual electricity costs of approximately 106 100 CNY · yr-1. The static payback period for the retrofit using existing luminaire housings is about 5.7 years, and the integrated passenger lighting comfort score reaches 58.6, with a deviation of less than 1% from the simulation prediction of 58.1. The proposed method can effectively adapt to the dynamic demands of hubs, providing theoretical and technical support for the intelligent upgrading of lighting systems in similar transportation hubs.
JiangTao, LuZhou. Multi-objective optimization of flat skylights in the elevated railway station[J]. Journal of South China University of Technology (Natural Science Edition), 2022, 50(7): 13-24.
[9]
路洲.广东地区高架式铁路客运站平天窗多目标优化设计研究[D].广州:华南理工大学,2021.
[10]
LuZhou. Study on multi-objective optimization design of flat skylight of elevated railway passenger station in Guangdong[D]. Guangzhou: South China University of Technology, 2021.
[11]
杨铁树.照明负荷平衡设计与安全节能[J].灯与照明,2014,38(3):11-13.
[12]
YangTieshu. Load balancing design and safety energy-saving of lighting[J]. Light & Lighting, 2014, 38(3); 11-13.
LiJindong, ZhangSu, HanSong, et al. Researches on BAS-based energy management system (EMS) of stations on Beijing-Zhangjiakou High-Speed Railway[J]. Railway Standard Design, 2020, 64(1): 176-179.
[15]
UyanE, AtlarM, ÖlçerA I. Improving the energy efficiency of lighting systems for a marine equipment manufacturing plant through retrofitting, daylighting, and behavior change[J]. Journal of Cleaner Production, 2023, 413: 137216.
HeFeng, WuLiangliang, ShaoJun, et al. Research on energy efficiency optimization of LED light sources in terminal building lighting systems[J]. Lamps & Lighting, 2024(4): 26-28.
YingTengyong. Research on energy efficiency improvement methods and intelligent management strategies for lighting systems in large public places[J]. China Light & Lighting, 2025(9): 115-118.
SunJun. Energy saving, carbon emission reduction and environment protection in underground space/rail transit works in urban area[J]. Tunnel Construction, 2011, 31(6): 643-647.
ZhangBomin. Thoughts on the application of lighting technology and energy saving in high-speed railway stations[J]. Shanghai Railway Science & Technology, 2013(1): 99-100.
YinCuijiao, LiuJianshun. Research on indoor lighting layout and energy efficiency improvement based on optimisation algorithm[J]. China Light & Lighting, 2025(7): 172-174.
LiuHu, ZhengHongjuan, SunJize, et al. Construction of comprehensive energy efficiency evaluation method for high-speed rail stations[J]. Mechanical & Electrical Engineering Technology, 2023, 52(10): 275-280, 284.
YangTianjun. Study on energy consumption and energy efficiency level of highway passenger stations based on climatic divisions[J]. Journal of Highway and Transportation Research and Development, 2018(7): 128-135.
WangQingyuan, FengXiaoyun, ZhuJinlin, et al. Simulation study on optimal energy-efficient control of high speed train considering regenerative brake energy[J]. China Railway Science, 2015, 36(1): 96-103.
[36]
QiYing, LiQiushi, HuangZhaoyuan, et al. Dual-domain low-light image enhancement with hierarchical illumination guidance[J]. Knowledge-Based Systems, 2026, 332: 114835.
ZhangYongyang, ZhangWenbin, ChenKun, et al. Research on adaptive monitoring and control system for energy consumption in transportation hubs[J]. Intelligent City, 2023, 9(2): 66-69.
[39]
KaurL, KaurR. Energy-efficient smart architecture for fog-based WSN using NSGA-III and improved layer-wise clustering for enhanced building evacuation safety[J]. Annals of Operations Research, 2024: 1-24.
[40]
LarsenJ L S, HawtionK, SilvermanM, et al. Ambivalence and suicidal behavior at railway stations: a scoping review[J]. Archives of Suicide Research: Official Journal of the International Academy for Suicide Research, 2025: 1-20.
[41]
谢婉云.基于行车安全考虑的立交枢纽照明设计优化研究[D].重庆:重庆交通大学,2025.
[42]
XieWanyun. Study on lighting design optimization of interchange based on traffic safety[D]. Chongqing: Chongqing Jiaotong University, 2025.
ShiHaiqi. Design of automatic lighting system for underground space of subway hub station based on DALI protocol[J]. Microcomputer Applications, 2024, 40(11): 285-288.
[45]
KhanA M, TariqM A, AlamZ, et al. Optimizing energy efficiency through building orientation and building information modelling (BIM) in diverse terrains: a case study in Pakistan[J]. Energy, 2024, 311: 133307.
[46]
LiaoRuixuan, ZhangYiming, WangHao, et al. Multi-objective optimisation of surveillance camera placement for bridge-ship collision early-warning using an improved non-dominated sorting genetic algorithm[J]. Advanced Engineering Informatics, 2024, 69(Part B): 103918.
[47]
DebK, AgrawalS, PratapA, et al. A fast and elitist multiobjective genetic algorithm: NSGA-II[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2): 182-197.
[48]
DebK, JainH. An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, Part I: solving problems with box constraints[J].IEEE Transactions on Evolutionary Computation. 2014, 18(4): 577-601.