基于检索增强生成和智能体的建筑材料碳排放单位换算问答模型

doi:10.6040/j.issn.1672-3961.0.2025.190

山东大学学报 (工学版) ›› 2026, Vol. 56 ›› Issue (1): 97-104.doi: 10.6040/j.issn.1672-3961.0.2025.190

• 土木工程 • 上一篇

基于检索增强生成和智能体的建筑材料碳排放单位换算问答模型

阎俏^1,2,焦飞³,严毅^1,2*,杜向华⁴,刘鹏程¹

1.山东建筑大学信息与电气工程学院, 山东济南 250101;2.山东省智慧建筑与建筑节能重点实验室(山东建筑大学), 山东济南 250101;3.中国铁塔股份有限公司菏泽市分公司, 山东菏泽 274000;4.山东建筑大学计算机与人工智能学院, 山东济南 250101

发布日期:2026-02-03
作者简介:阎俏(1976— ),女,山东荣成人,教授,硕士生导师,博士,主要研究方向为建筑智能化与能效管理、人工智能技术应用. E-mail:yanqiao@sdjzu.edu.cn. *通信作者简介:严毅(1987— ),男,山东济南人,副教授,硕士生导师,博士,主要研究方向为新能源、储能、建筑节能. E-mail: yanyi19@sdjzu.edu.cn
基金资助:
国家自然科学基金青年科学基金资助项目(52007109)

Question-answering model for building material carbon emissions unit conversion based on retrieval-augmented generation and Agent

YAN Qiao^1,2, JIAO Fei³, YAN Yi^1,2*, DU Xianghua⁴, LIU Pengcheng¹

YAN Qiao^{1, 2}, JIAO Fei³, YAN Yi^{1, 2*}, DU Xianghua⁴, LIU Pengcheng¹(1. School of Information and Electrical Engineering, Shandong Jianzhu University, Jinan 250101, Shandong, China;
2. Shandong Key Laboratory of Smart Buildings and Energy Efficiency, Shandong Jianzhu University, Jinan 250101, Shandong, China;
3. Heze Branch of China Tower Co., Ltd., Heze 274000, Shandong, China;
4. School of Computer and Artificial Intelligence, Shandong Jianzhu University, Jinan 250101, Shandong, China

Published:2026-02-03

摘要/Abstract

摘要： 为解决建筑材料生产及运输阶段碳排放计算时建筑材料计量单位与碳排放因子单位不匹配的问题,提出一种基于检索增强生成(retrieval-augmented generation, RAG)和智能体(Agent)的建筑材料碳排放单位换算问答模型。通过解析典型材料换算步骤构建本地知识库,设计RAG模块,为换算提供步骤参考;开发可调用计算工具的Agent,执行换算过程中的数学运算;设计提示词模板并接入大语言模型,实现基于本地知识库的文本问答。试验结果表明,所提模型能够准确回答建材的单位换算问题,支持Web端与本地控制台交互,实现单位换算结果及推理步骤的可视化。

关键词: 建筑材料碳排放, 单位换算, 检索增强生成, 智能体, 问答模型

Abstract: To solve mismatching between the measurement units of building materials and the units of carbon emission factors during the calculation of carbon emissions in the phase of building materials production and transportation, a question-answering model for building material carbon emissions unit conversion based on retrieval-augmented generation(RAG)and Agent was proposed. A local knowledge database was constructed by analyzing typical material conversion processes, and a RAG module was designed to provide step-by-step references for the conversions. An Agent capable of calling the calculation tool was developed to perform the mathematical operations required in the conversion process. The prompt templates were designed and integrated with a large language model to answer the question based on local knowledge database. The experimental results showed that the proposed model could accurately answer the unit conversion questions of building materials, and realized visualization of the unit conversion results and reasoning steps displayed on the Web interface and local console.

Key words: building material carbon emissions, unit conversion, retrieval-augmented generation, Agent, question-answering model

中图分类号:

阎俏,焦飞,严毅,杜向华,刘鹏程. 基于检索增强生成和智能体的建筑材料碳排放单位换算问答模型[J]. 山东大学学报 (工学版), 2026, 56(1): 97-104.

YAN Qiao, JIAO Fei, YAN Yi, DU Xianghua, LIU Pengcheng. Question-answering model for building material carbon emissions unit conversion based on retrieval-augmented generation and Agent[J]. Journal of Shandong University(Engineering Science), 2026, 56(1): 97-104.

参考文献

[1] 中华人民共和国住房和城乡建设部. 建筑碳排放计算标准: GB/T 51366—2019[S]. 北京: 中国建筑工业出版社, 2019: 2.
[2] 中国建筑节能协会, 重庆大学城乡建设与发展研究院. 中国建筑能耗与碳排放研究报告(2023年)[J]. 建筑, 2024(2): 46-59.
[3] YAN Q, JIAO F, PENG W. Building materials classification model based on text data enhancement and semantic feature extraction[J]. Buildings, 2024, 14: 1859.
[4] 钱力, 张智雄, 伍大勇, 等. 科技文献大模型: 方法、框架与应用[J]. 中国图书馆学报, 2024, 50(6): 45-58. QIAN Li, ZHANG Zhixiong, WU Dayong, et al. The large language model for scientific literature: method, framework, and application[J]. Journal of Library Science in China, 2024, 50(6): 45-58.
[5] 汪伦, 艾斯卡尔·艾木都拉, 张华平, 等. 基于大语言模型的开源情报摘要生成研究[J]. 情报理论与实践, 2025, 48(5): 43-48. WANG Lun, HAMDULLA Askar, ZHANG Huaping, et al. Research on open source intelligence summary generation based on large language models[J]. Information Studies: Theory & Application, 2025, 48(5): 43-48.
[6] 赵阳. 从DeepSeek探讨大语言模型在建筑及能源行业的应用趋势和技术方法[J]. 建筑节能(中英文), 2025, 53(2): 1-6. ZHAO Yang. Application trends and technical methods of LLM in the construction industry and energy industry from DeepSeek[J]. Journal of Building Energy Efficiency, 2025, 53(2): 1-6.
[7] DU Y L, LI S, TORRALBA A, et al. Improving factuality and reasoning in language models through multiagent debate[C] //Proceedings of the 41st Interna-tional Conference on Machine Learning. Vienna, Austria: ACM, 2024: 11733-11763.
[8] LU J Y, YU L, LI X J, et al. LLaMA-reviewer: advancing code review automation with large language models through parameter-efficient fine-tuning[C] //2023 IEEE 34th International Symposium on Software Reliability Engineering(ISSRE). Florence, Italy: IEEE, 2023: 647-658.
[9] 刘雪颖, 云静, 李博, 等. 基于大型语言模型的检索增强生成综述[J]. 计算机工程与应用, 2025, 61(13): 1-25. LIU Xueying, YUN Jing, LI Bo, et al. Survey of retrieval-augmented generation based on large language models[J]. Computer Engineering and Applications, 2025, 61(13): 1-25.
[10] HINDI M, MOHAMMED L, MAAZ O, et al. Enhancing the precision and interpretability of retrieval-augmented generation(RAG)in legal technology: a survey[J]. IEEE Access, 2025, 13: 46171-46189.
[11] WANG L, MA C, FENG X Y, et al. A survey on large language model based autonomous agents[J]. Frontiers of Computer Science, 2024, 18(6): 186345.
[12] WEI Q, SUN H J, XU Y, et al. Exploring the application of large language models based AI Agents in leakage detection of natural gas valve chambers[J]. Energies, 2024, 17(22): 5633.
[13] 中国产品全生命周期温室气体排放系数库[DB/OL].(2025-05-12)[2025-09-04]. http://lca.cityghg.com
[14] 沈思, 冯暑阳, 吴娜, 等. 融合大语言模型的政策文本检索增强生成研究[J]. 数据分析与知识发现, 2025, 9(9): 37-48. SHEN Si, FENG Shuyang, WU Na, et al. Retrieval-augmented generation of policy texts based on large language models[J]. Data Analysis and Knowledge Discovery, 2025, 9(9): 37-48.
[15] 齐俊, 曲睿婷, 教传铭, 等. 基于知识图谱增强大语言模型双碳领域服务[J]. 计算机与现代化, 2024(9): 8-14. QI Jun, QU Ruiting, JIAO Chuanming, et al. Enhanced big language model dual carbon domain services based on knowledge graph[J]. Computer and Modernization, 2024(9): 8-14.
[16] Qwen Team. Qwen2-Math, 新一代数学模型[EB/OL].(2024-08-08)[2025-09-04]. https://qwenlm.github.io/zh/blog/qwen2-math/
[17] 张天鸿, 王晓玲, 余红玲, 等. 基于大语言模型的灌浆工程知识服务系统[J]. 水利学报, 2025, 56(1): 130-142. ZHANG Tianhong, WANG Xiaoling, YU Hongling, et al. Grouting works knowledge service system based on large language model[J]. Journal of Hydraulic Engi-neering, 2025, 56(1): 130-142.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于检索增强生成和智能体的建筑材料碳排放单位换算问答模型

Question-answering model for building material carbon emissions unit conversion based on retrieval-augmented generation and Agent

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

多维度评价

本文评价

推荐阅读 0

[1]	张波,徐悦,康乐,张贵军. 基于差分隐私机制和单点反馈的分布式在线优化算法[J]. 山东大学学报 (工学版), 2026, 56(1): 14-25.
[2]	崔阳,张柯,姜斌. 具有切换拓扑结构的多智能体系统故障估计[J]. 山东大学学报(工学版), 2017, 47(5): 263-270.
[3]	武炎明,王瑞云,王占山. 基于中间变量观测器的多智能体故障检测[J]. 山东大学学报(工学版), 2017, 47(5): 96-102.
[4]	叶丹,张天予,李奎. 全局信息未知的多智能体自适应容错包容控制[J]. 山东大学学报(工学版), 2017, 47(5): 1-6.
[5]	刘彬,张仁津. 一种采用两段粒子群优化的路径规划方法[J]. 山东大学学报(工学版), 2012, 42(1): 12-18.
[6]	王云,王俊,韩伟*. 基于进化算法的多智能体合作学习[J]. 山东大学学报(工学版), 2010, 40(6): 8-11.
[7]	程显毅1,2,朱倩2,管致紧1. 基于agent的模式表示模型AIM[J]. 山东大学学报(工学版), 2010, 40(5): 72-76.
[8]	云健1,3,江荻1,2,潘悟云1. 模因机制下人类元音系统演化的计算模型[J]. 山东大学学报(工学版), 2010, 40(4): 12-18.