融合大模型和数字孪生的公共建筑智慧运维系统

doi:10.11996/JG.j.2095-302X.2024061200

图学学报 ›› 2024, Vol. 45 ›› Issue (6): 1200-1206.DOI: 10.11996/JG.j.2095-302X.2024061200

• “大模型与图学技术及应用”专题 • 上一篇下一篇

融合大模型和数字孪生的公共建筑智慧运维系统

许璟琳¹^,²(), 彭阳², 欧金武², 谈骏杰², 舒江鹏¹(), 余芳强³

1.浙江大学工程师学院，浙江杭州 310015
2.上海建工四建集团有限公司，上海 201103
3.上海建工集团股份有限公司，上海 200080

收稿日期:2024-07-16 接受日期:2024-10-22 出版日期:2024-12-31 发布日期:2024-12-24
通讯作者:舒江鹏(1987-)，男，研究员，博士。主要研究方向为智能建造与检测。E-mail：jpeshu@zju.edu.cn
第一作者:许璟琳(1989-)，女，博士研究生。主要研究方向为工程数字化、数字孪生和大模型应用。E-mail：jinglin.xu@qq.com
基金资助:
上海市国资委能级提升项目(2022008);上海市科技创新行动计划项目(22dz1207102)

An intelligent maintenance system for public buildings integrating digital twin and large language model

XU Jinglin¹^,²(), PENG Yang², OU Jinwu², TAN Junjie², SHU Jiangpeng¹(), YU Fangqiang³

1. Polytechnic Institute, Zhejiang University, Hangzhou Zhejiang 310015, China
2. Shanghai Construction No.4 (Group) Co. LTD, Shanghai 201103, China
3. Shanghai Construction Group Co. LTD, Shanghai 200080, China

Received:2024-07-16 Accepted:2024-10-22 Published:2024-12-31 Online:2024-12-24
Contact: SHU Jiangpeng (1987-), researcher, Ph.D. His main research interests cover intelligent construction and testing, etc. E-mail：jpeshu@zju.edu.cn
First author：XU Jingnlin (1989-), PhD candidate. Her main research interests cover engineering digitization, digital twins, and large model applications. E-mail：jinglin.xu@qq.com
Supported by:
Shanghai State-Owned Assets Supervision and Administration Commission Level Promotion Project(2022008);Shanghai Science and Technology Innovation Action Plan Project(22dz1207102)

摘要/Abstract

摘要：

为解决基于数字孪生的建筑智慧运维面临的系统操作复杂、海量建设文档信息难以查阅、复杂场景决策支持弱等问题，构建了融合大模型和数字孪生的建筑智慧运维系统，创新了基于检索增强生成的海量信息高效检索技术、基于大模型的建筑运维服务高效调用技术、基于群体智能的楼宇智能调适技术等，在3类典型运维场景进行了应用验证，表明融合大模型和数字孪生构建的公共建筑智慧运维系统，有助于提供运维个性化服务、提升用户体验、提供复杂决策支持，实现更便捷、更舒适、更安全、更绿色的公共建筑智慧运维管理。

关键词: 大语言模型, 数字孪生, 公共建筑, 智慧运维

Abstract:

To address the challenges encountered in smart building operations and maintenance based on digital twins, such as complex system operations, difficulties in accessing a vast amount of construction documentation, and limited decision support in complex scenarios, a smart building operations and maintenance system integrating large models and digital twins was constructed. Innovations include efficient retrieval technology for massive information based on Retrieval Augmented Generation, efficient invocation technology for building operations and maintenance services based on large models, and intelligent building adaptation technology based on swarm intelligence. The system was applied and verified in three typical operations and maintenance scenarios. The results demonstrated that integrating large models and digital twins in constructing a public building smart operations and maintenance system aided in providing personalized building operations and maintenance services, enhanced user experience, offered complex decision support, and enabled more convenient, comfortable, safe, and green smart operations and maintenance management for public buildings.

Key words: large language models, digital twin, public buildings, intelligent maintenance

中图分类号:

许璟琳, 彭阳, 欧金武, 谈骏杰, 舒江鹏, 余芳强. 融合大模型和数字孪生的公共建筑智慧运维系统[J]. 图学学报, 2024, 45(6): 1200-1206.

XU Jinglin, PENG Yang, OU Jinwu, TAN Junjie, SHU Jiangpeng, YU Fangqiang. An intelligent maintenance system for public buildings integrating digital twin and large language model[J]. Journal of Graphics, 2024, 45(6): 1200-1206.

图/表 10

图1 运维大模型采用的RAG过程

Fig. 1 RAG process for maintenance LLM

图2 文档检索服务数据流图

Fig. 2 Data flow of file search service

图3 基于大模型的运维服务调用

Fig. 3 Maintenance service invocation using LLM

图4 基于群体智能的楼宇智能调适

Fig. 4 Intelligent building adjustment based on group intelligence

图5 整体技术架构

Fig. 5 Methodology framework

图6 运维大模型查询空调关联文档

Fig. 6 Document query about air-conditioners by LLM

图7 运维大模型预约会议室

Fig. 7 Meeting room reservation by LLM

图8 运维大模型接受语音调控机电系统

Fig. 8 LLM receives voice command and adjust electric systems

图9 运维大模型实施智能调适

Fig. 9 Intelligent adjustments by maintenance LLM

表1 运维场景满意度评价

Table 1 Performance evaluation of maintenance scenarios

场景	交互次数	评价次数	好评数	满意率/%
建设过程文档检索	1043	274	240	87.6
数字孪生系统交互优化	207	207	193	93.2
楼宇智能调适	576	305	266	87.2

参考文献 14

[1]	PENG Y, ZHANG M, YU F Q, et al. Digital twin hospital buildings: an exemplary case study through continuous lifecycle integration[J]. Advances in Civil Engineering, 2020, 2020: 8846667.
[2]	WANG H, MENG X H, ZHU X Y. Improving knowledge capture and retrieval in the BIM environment: combining case-based reasoning and natural language processing[J]. Automation in Construction, 2022, 139: 104317.
[3]	SHAHINMOGHADAM M, KAHOU S E, MOTAMEDI A. Neural semantic tagging for natural language-based search in building information models: implications for practice[J]. Computers in Industry, 2024, 155: 104063.
[4]	MO Y, ZHAO D, DU J, et al. Automated staff assignment for building maintenance using natural language processing[J]. Automation in Construction, 2020, 113: 103150.
[5]	LI M H, WANG R H, ZHOU X, et al. ChatTwin: toward automated digital twin generation for data center via large language models[C]// The 10th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation. New York: ACM, 2023: 208-211.
[6]	何庆, 荆传玉, 孙华坤, 等. 基于BIM和语义网的轨道智能运维管理方法[J]. 图学学报, 2024, 45(3): 601-612. DOI
	HE Q, JING C Y, SUN H K, et al. An intelligent railway operation and maintenance management approach based on BIM and semantic web[J]. Journal of Graphics, 2024, 45(3): 601-612. (in Chinese) DOI
[7]	KIM J, CHUNG S, CHI S. Cross-lingual information retrieval from multilingual construction documents using pretrained language models[J]. Journal of Construction Engineering and Management, 2024, 150(6): 04024041.
[8]	RADFORD A, NARASIMHAN K, SALIMANS T, et al. Improving language understanding by generative pre-training[EB/OL]. [2024-05-16]. https://gwern.net/doc/www/s3-us-west-2.amazonaws.com/d73fdc5ffa8627bce44dcda2fc012da638ffb158.pdf.
[9]	蒋灿, 郑哲, 梁雄, 等. 大语言模型驱动的交互式建筑设计新范式——基于Rhino7的概念验证[J]. 图学学报, 2024, 45(3): 594-600. DOI
	JIANG C, ZHENG Z, LIANG X, et al. A new interaction paradigm for building design driven by large language model: proof of concept with Rhino7[J]. Journal of Graphics, 2024, 45(3): 594-600. (in Chinese) DOI
[10]	CHIU T K F. The impact of Generative AI (GenAI) on practices, policies and research direction in education: a case of ChatGPT and Midjourney[EB/OL]. (2023-09-04) [2024-05-16]. https://doi.org/10.1080/10494820.2023.2253861.
[11]	张华平, 李林翰, 李春锦. ChatGPT中文性能测评与风险应对[J]. 数据分析与知识发现, 2023, 7(3): 16-25. DOI
	ZHANG H P, LI L H, LI C J. ChatGPT performance evaluation on Chinese language and risk measures[J]. Data Analysis and Knowledge Discovery, 2023, 7(3): 16-25. (in Chinese) DOI
[12]	ZHANG L, CHEN Z L. Large language model-based interpretable machine learning control in building energy systems[J]. Energy and Buildings, 2024, 313: 114278.
[13]	裴丹, 张圣林, 孙永谦, 等. 大语言模型时代的智能运维[J]. 中兴通讯技术, 2024, 30(2): 56-62.
	PEI D, ZHANG S L, SUN Y Q, et al. Artificial intelligence for IT operations in era of large language model[J]. ZTE Technology Journal, 2024, 30(2): 56-62. (in Chinese)
[14]	JIANG G, MA Z H, ZHANG L, et al. EPlus-LLM: a large language model-based computing platform for automated building energy modeling[J]. Applied Energy, 2024, 367: 123431.

融合大模型和数字孪生的公共建筑智慧运维系统

An intelligent maintenance system for public buildings integrating digital twin and large language model

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 14

相关文章 10

编辑推荐

Metrics

本文评价

[1]	徐沛, 黄凯奇. 大模型引导的高效强化学习方法[J]. 图学学报, 2024, 45(6): 1165-1177.
[2]	陈晓皎, 束云峰, 汪睿涵, 周佳欢, 陈为. 大语言模型驱动的UI评估系统[J]. 图学学报, 2024, 45(6): 1178-1187.
[3]	于晗, 陈治源, 熊熙瑞, 戴原星, 蔡鸿明. 基于检索增强大语言模型的MBSE智能设计方法[J]. 图学学报, 2024, 45(6): 1188-1199.
[4]	蒋灿, 郑哲, 梁雄, 林佳瑞, 马智亮, 陆新征. 大语言模型驱动的交互式建筑设计新范式——基于Rhino7的概念验证[J]. 图学学报, 2024, 45(3): 594-600.
[5]	黄文恺, 梁智洪, 王明华, 张文丰, 王奕首. 数字孪生在航空航天结构设计、制造和运维中的应用与展望[J]. 图学学报, 2024, 45(2): 241-249.
[6]	庞博, 杨辉, 于荣荣, 张浩月, 罗强. 基于数字孪生技术的航天器机构生产线平衡研究[J]. 图学学报, 2024, 45(2): 332-338.
[7]	汤鹏, 撒国栋, 刘振宇, 谭建荣. 锻造液压机数字孪生系统设计[J]. 图学学报, 2023, 44(3): 609-615.
[8]	高铭, 张荷花, 张庭瑞, 张轩铭. 基于深度学习的公共建筑像素施工图空间识别[J]. 图学学报, 2022, 43(2): 189-196.
[9]	田凌, 刘果, 刘思超. 数字孪生与生产线仿真技术研究[J]. 图学学报, 2021, 42(3): 349-358.
[10]	文桥 1，张建平 1，向雪松 2，石涛 2，胡振中 3,1 . 基于 GIS 的大型公共建筑轻量级运维管理平台研发与应用[J]. 图学学报, 2019, 40(4): 751-760.