基于领域大语言模型与多智能体协作的工业应用快速构建方法

doi:10.3901/JME.260439

机械工程学报 ›› 2026, Vol. 62 ›› Issue (11): 147-161.doi: 10.3901/JME.260439

• 特邀专栏：制造互联与工业智能 • 上一篇

扫码分享

基于领域大语言模型与多智能体协作的工业应用快速构建方法

吴鹏程^1,2,3,4, 慕艳双^1,2,3, 徐鑫雨^1,2,3, 邱林琼⁵, 吕照亮⁶, 杨波⁴

1. 西南大学人工智能学院重庆 400715;
2. 智能传动与控制国家地方联合工程中心重庆 400715;
3. 西南大学类脑实验技术中心重庆 400715;
4. 重庆大学高端装备机械传动全国重点实验室重庆 400044;
5. 昆士兰科技大学布里斯班 4072 澳大利亚;
6. 联宝(合肥)电子科技有限公司合肥 231200

收稿日期:2025-06-04 修回日期:2025-12-25 发布日期:2026-07-29
作者简介:吴鹏程(通信作者),男,1994年出生,博士,副教授。主要研究方向为工业自动化、生成式人工智能、多模态大模型。E-mail:wu20230730@swu.edu.cn;杨波,男,1986年出生,博士,教授。主要研究方向为智能制造、工业大数据、制造业知识工程、工业大模型。E-mail:yangbo61@cqu.edu.cn
基金资助:
2024年重庆市教育委员会人文社会科学研究“基于新一代人工智能的重庆新质生产力发展模式研究”(24SKGH400)、国家重点研发计划(2023YFB3306800)和新重庆青年创新人才(CSTB2024NSCQ-QCXMX0028)资助项目。

Rapid Construction Method for Industrial Applications Based on Domain-specific Large Language Models and Multi-agents Collaboration

WU Pengcheng^1,2,3,4, MU Yanshuang^1,2,3, XU Xinyu^1,2,3, QIU Linqiong⁵, Lü Zhaoliang⁶, YANG Bo⁴

1. School of Artificial Intelligence, Southwest University, Chongqing 400715;
2. National Local Joint Engineering Center for Intelligent Transmission and Control, Chongqing 400715;
3. Brain-like E ent Technology Center, Southwest University, Chongqing 400715;
4. State Key Laboratory of Mechanical Transmissions, Chongqing University, Chongqing 400044;
5. Queensland University of Technology, Brisbane 4072 Australia;
6. Lenovo Legions Technology, Hefei 231200

Received:2025-06-04 Revised:2025-12-25 Published:2026-07-29

摘要/Abstract

摘要： 工业5.0的发展对制造业的信息化、数字化和智能化提出了更高的要求。针对传统工业应用构建过程专家经验依赖高，以及现有通用大模型开发过程已陷入幻觉陷阱等问题，提出了基于领域大语言模型与协作智能体的工业应用快速构建方法，该方法首先构建工业应用开发的知识图谱，进而利用知识图谱嵌入提示的方式，构建了面向工业应用快速开发的领域大模型。进一步，将构建的大模型进行多智能体拆解，建立了面向工业开发的需求分析智能体、编码智能体以及测试智能体等多智能体协作机制，从而实现了面向工业应用快速构建服务的开发与综合运用。为证明所提出方法的有效性与优越性，本文采用主客观综合评价的方式，选取了工业设备数据采集开发案例，将所提出方法与传统方法进行了综合对比。结果表明本文所提出不仅开发能力更强，且具有更为友好的人机交互性，能够为当前工业应用快速构建提供有力的方法支撑。

关键词: 领域大模型, 多智能体, 工业应用, 知识图谱

Abstract: The advancement of Industry 5.0 imposes higher demands on the informatization, digitization, and intelligence of manufacturing. Addressing challenges such as the heavy reliance on expert experience in traditional industrial application development and the hallucination traps prevalent in existing general-purpose large model frameworks, a rapid construction method for industrial applications leveraging domain-specific large language models (LLMs) and collaborative agents is proposed. The methodology begins by constructing a knowledge graph for industrial application development, which is then embedded into a prompting framework to establish a domain-specific LLM tailored for rapid industrial development. Subsequently, the LLM is decomposed into multiple collaborative agents, including requirement analysis agents, coding agents, and testing agents, enabling an integrated workflow for accelerated industrial application development. To validate the effectiveness and superiority of the proposed method, a subjective and objective comprehensive evaluation was conducted using real-world development cases, comparing the proposed approach with traditional methods. Results demonstrate that the proposed method not only exhibits enhanced development capabilities but also provides more user-friendly human-machine interaction, making it better suited for contemporary rapid industrial application construction services.

Key words: domain-specific large language models, multi-agents, industrial applications, knowledge graph

中图分类号:

TP391
TP2

吴鹏程, 慕艳双, 徐鑫雨, 邱林琼, 吕照亮, 杨波. 基于领域大语言模型与多智能体协作的工业应用快速构建方法[J]. 机械工程学报, 2026, 62(11): 147-161.

WU Pengcheng, MU Yanshuang, XU Xinyu, QIU Linqiong, Lü Zhaoliang, YANG Bo. Rapid Construction Method for Industrial Applications Based on Domain-specific Large Language Models and Multi-agents Collaboration[J]. Journal of Mechanical Engineering, 2026, 62(11): 147-161.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260439

http://www.cjmenet.com.cn/CN/Y2026/V62/I11/147

参考文献

[1] 李嘉佳，易茜，冯毅雄，等. 人本智造单元中人-智能系统协同双智能体工作机制研究[J]. 机械工程学报，2025，61(3)：105-118. LI Jiajia，YI Qian，FENG Yixiong，et al. Research on dual-agent work mechanism with human-smart system collaboration in human-centric smart manufacturing cell[J]. Journal of Mechanical Engineering，2025，61(3)：105-118.
[2] 臧冀原，刘宇飞，王柏村，等. 面向2035的智能制造技术预见和路线图研究[J]. 机械工程学报，2022，58(4)：285-304. ZANG Jiyuan，LIU Yufei，WANG Baicun，et al. Technology forecasting and road mapping of intelligent manufacturing by 2035[J]. Journal of Mechanical Engineering，2022，58(4)：285-304.
[3] 张超，周光辉，李晶晶，等. 新一代信息技术赋能的数字孪生制造单元系统关键技术及应用研究[J]. 机械工程学报，2022，58(16)：329-344. ZHANG Chao，ZHOU Guanghui，LI Jingjing，et al. Research on key technologies and application of new IT-driven digital twin manufacturing cell system[J]. Journal of Mechanical Engineering，2022，58(16)：329-344.
[4] PETRASCH R ，HENTSCHKE R. Process modeling for Industry 4.0 applications：Towards an Industry 4.0 process modeling language and method[C]// 201613th International Joint Conference on Computer Science and Software Engineering (JCSSE) (2016)：1-5.
[5] XI M，ROZYCKI P，QIAO J，et al. Nonlinear system modeling using RBF networks for industrial application[J]. IEEE Transactions on Industrial Informatics，2017，14(3)：931-940.
[6] 马南峰，姚锡凡，陈飞翔，等. 面向工业5.0的人本智造[J]. 机械工程学报，2022，58(18)：88-103. MA Nanfeng，YAO Xifan，CHEN Feixiang，et al. Human-centric Smart Manufacturing for Industry 5.0[J]. Journal of Mechanical Engineering，2022，58(18)：88-103.
[7] LATIF K，ADAM A，YUSOF Y，et al. A review of G code，STEP，STEP-NC，and open architecture control technologies based embedded CNC systems[J].The International Journal of Advanced Manufacturing Technology，2021，114：2549-2566.
[8] RIVET I，DIALAMI N，CERVERA M，et al. Mechanical analysis and optimized performance of G-Code driven material extrusion components[J]. Additive Manufacturing，2023，61：103348.
[9] BERTOLINI M，MEZZOGORI D，NERONI M，et al. Machine learning for industrial applications：A comprehensive literature review[J]. Expert Systems with Applications，2021，175(6)：114820.
[10] LARRAÑAGA P，ATIENZA D，DIAZ-ROZO J，et al. Industrial applications of machine learning [M]. Florida：CRC Press，2018.
[11] TERCAN H，MEISEN T. Machine learning and deep learning based predictive quality in manufacturing：a systematic review[J]. Journal of Intelligent Manufacturing，2022，33(7)：1879-1905.
[12] WANG J，MA Y，ZHANG L，et al. Deep learning for smart manufacturing：Methods and applications[J]. Journal of Manufacturing Systems，2018，48：144-156.
[13] ANNEPAKA Y，PAKRAY P. Large language models：A survey of their development，capabilities，and applications [J]. Knowledge and Information Systems. 2025，67：2967-302.
[14] LI X，WANG S，ZENG S，et al. A survey on LLM-based multi-agent systems：Workflow，infrastructure，and challenges[J]. Vicinagearth，2024，1(1)：1-43.
[15] YUKSEL H. Artificial intelligence form of digitalized communication：ChatGPT[M]. Ankara：ISRES Publishing，2023.
[16] REYHAN AH，MUTAF Ç，UZUN İ，et al. A performance evaluation of large language models in keratoconus：A comparative study of ChatGPT-3.5，ChatGPT-4.0，gemini，copilot，chatsonic，and perplexity[J]. Journal of Clinical Medicine，2024，13(21)：6512.
[17] ROUMELIOTIS KI，TSELIKAS ND. ChatGPT and open-ai models：A preliminary review[J]. Future Internet，2023，15(6)：192
[18] SARSA S，DENNY P，HELLAS A，et al. Automatic generation of programming exercises and code explanations using large language models[C]// Proceedings of the 2022 ACM Conference on International Computing Education Research-Volume 12022 Aug 3，27-43.
[19] BEURER-KELLNER L，FISCHER M，VECHEV M. Prompting is programming：A query language for large language models[C]// Proceedings of the ACM on Programming Languages，2023，7(PLDI)：1946-1969.
[20] QIAN C，LIU W，LIU H，et al. ChatDev：Communicative agents for software development[J]. arXiv:2307.079242023.
[21] 徐增林，盛泳潘，贺丽荣，等. 知识图谱技术综述[J]. 电子科技大学学报，2016，45(4)：589-606. XU Zenglin，SHENG Yongpan，HE Lirong，et al. Review on knowledge graph techniques[J]. Journal of Electronic Science and Technology of China，2016，45(4)：589-606.
[22] JI S，PAN S，CAMBRIA E，et al. A survey on knowledge graphs：Representation，acquisition，and applications [J]. IEEE Transactions on Neural Networks and Learning Systems，2021，33(2)：494-514.
[23] ZHONG L ，WU J ，LI Q ，et al. A comprehensive survey on automatic knowledge graph construction[J]. ACM Computing Surveys，2024，56(4)：1-62.
[24] ASSCHE D V ，DELVA T ，HAESENDONCK G ，et al. Declarative RDF graph generation from heterogeneous (semi-)structured data：A systematic literature review[J]. Journal Web Semant，2022，75：100753.
[25] MENG F ，YANG S ，WANG J ，et al.Creating Knowledge Graph of Electric Power Equipment Faults Based on BERT–BiLSTM–CRF Model[J]. Journal of Electrical Engineering & Technology：1-10
[26] JIA Y ，QI Y ，SHANG H ，et al. A practical approach to constructing a knowledge graph for cybersecurity[J]. Engineering，2018，4(1)：53-60.
[27] FENG C，ZHANG X，FEI Z. Knowledge solver：Teaching llms to search for domain knowledge from knowledge graphs [J]. arXiv preprint arXiv：2309.03118，2023.
[28] LI P，QIAN L，ZHAO X，et al. Joint knowledge graph and large language model for fault diagnosis and its application in aviation assembly [J]. IEEE Transactions on Industrial Informatics. 2024 Mar 8.
[29] WU P，TU H，MOU X，et al. An intelligent energy management method for the manufacturing systems using the knowledge graph and large language model [J]. Journal of Intelligent Manufacturing，2026，37：1125-1144.
[30] ADENOWO A ，ADENOWO B A. Software engineering methodologies：A review of the waterfall model and object-oriented approach[J]. International Journal of Scientific and Engineering Research，2020，4(7)：427-434.
[31] HAYDER W A .Highlighting DeepSeek-R1：Architecture，features and future implications[J]. International Journal of Computer Science and Mobile Computing，2025，14(2)：1-13.
[32] TOUVRON H，LAVRIL T，IZACARD G，et al. Llama：Open and efficient foundation language models[J]. arXiv preprint arXiv：2302.13971.
[33] PENG Junlong，LIU Xiangjun. Automated code compliance checking research based on BIM and knowledge graph [J]. Scientific Reports，2023，13(1)：7065.

基于领域大语言模型与多智能体协作的工业应用快速构建方法

Rapid Construction Method for Industrial Applications Based on Domain-specific Large Language Models and Multi-agents Collaboration

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张年华, 张永康, 陈继成, 李朋涛, 李焱, 张辉. 基于多智能体理论的分布式驱动车辆ESC-TVC鲁棒协同控制策略研究[J]. 机械工程学报, 2026, 62(8): 365-381.
[2]	郑小虎, 曹立俊, 刘骁佳, 杜思淇, 武文强, 张洁, 丁司懿. 基于知识图谱的航天产品总装质量追溯方法研究[J]. 机械工程学报, 2026, 62(5): 88-99.
[3]	姚锡凡, 胡坤, 李佳蔚. 面向工业5.0的多智能体云边端协同智慧制造体系架构与作业调度[J]. 机械工程学报, 2026, 62(11): 102-115.
[4]	郭鑫, 黄泽川, 王杰, 张凯, 赵武. 人机混合交互的产品设计知识图谱构建及智能推理方法[J]. 机械工程学报, 2025, 61(3): 142-153.
[5]	寇逸群, 杨晔, 刘颉, 胡友民, 李林, 俞百川, 徐家和, 胡中旭, 史铁林. 面向认知赋能的人机协作：进展、挑战和展望[J]. 机械工程学报, 2025, 61(3): 1-22.
[6]	唐小林, 甘炯鹏, 张振果. 横纵向耦合跟车场景下基于多智能体深度强化学习的混合动力车队协同能量管理研究[J]. 机械工程学报, 2025, 61(2): 236-246.
[7]	杨波, 申小玉, 王时龙, 何彦, 杜卡泽. 面向设备运维的人机物三元融合知识图谱构建方法[J]. 机械工程学报, 2025, 61(17): 215-232.
[8]	杨佳幸, 温沛涵, 胡亚萍. 核电装备价值链协同质量管控知识图谱构建研究[J]. 机械工程学报, 2025, 61(13): 174-191.
[9]	丁满, 孙鸣宇, 张新新, 冯光宇. 基于知识超图的产品色彩情感设计[J]. 机械工程学报, 2024, 60(17): 339-348.
[10]	喻惟刚, 文思捷, 李建军, 李心雨. 基于可供性推理的人机协作拆卸序列规划方法[J]. 机械工程学报, 2024, 60(17): 297-309.
[11]	吴秀丽, 马隆洲, 向东, 唐英. 面向退役机电产品全生命周期的知识图谱统一建模方法研究[J]. 机械工程学报, 2023, 59(7): 52-67.
[12]	从靖晨, 项忠霞, 李心雨, 董良, 李巨韬, 陈俊贤. 基于知识图谱的智能产品服务系统交互设计研究[J]. 机械工程学报, 2023, 59(11): 94-105.
[13]	张栋豪, 刘振宇, 郏维强, 刘惠, 谭建荣. 知识图谱在智能制造领域的研究现状及其应用前景综述[J]. 机械工程学报, 2021, 57(5): 90-113.
[14]	牛礼民, 杨洪源, 周亚洲, 吕建美. 混合动力汽车动力总成多智能体集成控制策略[J]. 机械工程学报, 2019, 55(12): 168-177,188.
[15]	徐影凤,汪海宁. 基于多智能体的微网运行决策支持系统结构研究[J]. 电气工程学报, 2016, 11(8): 30-36.