Domain-Specific Large Language Model for Maintenance Decision-Making on Wind Farms by Labeled-Data-Supervised Fine-Tuning

Dongming Fan , Meng Liu , Yi Shao , Linchao Yang , Yiliu Liu , Yue Zhang , Yi Ren , Zili Wang

Engineering ›› : 202512019

PDF (5587KB)
Engineering ›› :202512019 DOI: 10.1016/j.eng.2025.12.019
Research
research-article
Domain-Specific Large Language Model for Maintenance Decision-Making on Wind Farms by Labeled-Data-Supervised Fine-Tuning
Author information +
History +
PDF (5587KB)

Abstract

Wind farm operators always need a better maintenance strategy to increase resource utilization efficiency while controlling operation and maintenance costs. However, conventional maintenance decision-making approaches are time-consuming and have poor flexibility and adaptability to various scenarios. This study addressed these challenges by using a large language model (LLM) to understand, generate, and plan maintenance strategies for wind farms characterized by various failure modes and maintenance costs. A labelled-data-supervised fine-tuning LLM for maintenance, named LLM4M, is proposed. The proposed LLM4M model is trained on an extensive dataset of mathematical programs for maintenance to generate optimal strategies for wind farms. Compared with other large parameter LLMs, the fine-tuned LLM4M model demonstrates remarkable accuracy, with an error of approximately 2% from the optimal strategy. In addition, the generalization of the proposed LLM4M model has achieved remarkable results. If the LLM4M model correctly generates the maintenance strategy, the maintenance cost deviates from the optimal solution by only approximately 5%. Furthermore, phase transition behavior is observed, which provides considerable guidance for the development of domain-specific LLMs for the maintenance domain.

Keywords

Large language model / Maintenance decision-making / Wind farms / Fine-tuning

Cite this article

Download citation ▾
Dongming Fan, Meng Liu, Yi Shao, Linchao Yang, Yiliu Liu, Yue Zhang, Yi Ren, Zili Wang. Domain-Specific Large Language Model for Maintenance Decision-Making on Wind Farms by Labeled-Data-Supervised Fine-Tuning. Engineering 202512019 DOI:10.1016/j.eng.2025.12.019

登录浏览全文

4963

注册一个新账户 忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]

Bai X, Fan Y, Hou J, Sun Y, Liu Y, Liu J. Reliability evaluation of direct current gathering system in onshore wind farm based on reliability block diagram-sequential monte carlo. Sustain Energy Grids Netw 2024;40:101549.
[2]

Tao Z, Liu H, Si Y, Wang C, Zhu R. An opportunistic joint maintenance strategy for two offshore wind farms. Ocean Eng 2024;304:117890.
[3]

Elsahhar AU, Ezzat AA, Elsabbagh A, Elbanhawy AY. Analysis of failure and maintenance records in aging wind farms to inform end-of-life asset management. Wind Energy 2025; 28(6):e70024.
[4]

Tao Z, Zhu R, Hu J, Wang M, Chen Q, Wang C. A novel hierarchical failure analysis approach targeting the operation and maintenance of floating offshore wind turbines. Renew Energy 2025;241:122267.
[5]

Izquierdo J, Márquez AC, Uribetxebarria J, Erguido A. On the importance of assessing the operational context impact on maintenance management for life cycle cost of wind energy projects. Renew Energy 2020;153:1100-10.
[6]

Wang H, Li YF, Ren J. Machine learning for fault diagnosis of high-speed train traction systems: a review. Front Eng Manag 2024;11:62-78.
[7]

Pelajo JC, Brandão LET, Gomes LL, Klotzle MC. Wind farm generation forecast and optimal maintenance schedule model. Wind Energy 2019;22:1872-90.
[8]

Dui H, Wu X, Wu S, Xie M. Importance measure-based maintenance strategy optimization: fundamentals, applications and future directions in AI and IoT. Front Eng Manag 2024;11:542-67.
[9]

Niemi A, Skobiej B, Kulev N, Sill TF. Modeling offshore wind farm disturbances and maintenance service responses within the scope of resilience. Reliab Eng Syst Saf 2024;242:109719.
[10]

Ade Irawan C, Starita S, Chan HK, Eskandarpour M, Reihaneh M. Routing in offshore wind farms: a multi-period location and maintenance problem with joint use of a service operation vessel and a safe transfer boat. Eur J Oper Res 2023;307:328-50.
[11]

Fan D, Sun B, Dui H, Zhong J, Wang Z, Ren Y, et al. A modified connectivity link addition strategy to improve the resilience of multiplex networks against attacks. Reliab Eng Syst Saf 2022;221:108294.
[12]

Su C, Wu L. Opportunistic maintenance optimisation for offshore wind farm with considering random wind speed. Int J Prod Res 2024;62:1862-78.
[13]

Li M, Bijvoet B, Wu K, Jiang X, Negenborn RR. Optimal chartering decisions for vessel fleet to support offshore wind farm maintenance operations. Ocean Eng 2024;298:117202.
[14]

Wang Y, Han Y, Gong D, Li H. A review of intelligent optimization for group scheduling problems in cellular manufacturing. Front Eng Manag 2023;10:406-26.
[15]

Fan D, Ren Y, Feng Q, Liu Y, Wang Z, Lin J. Restoration of smart grids: current status, challenges, and opportunities. Renew Sustain Energy Rev 2021;143:110909.
[16]

Zhao X, Chen P, Tang LC. Condition-based maintenance via Markov decision processes: a review. Front Eng Manag 2025;12:330-42.
[17]

Qureshi TA, Warudkar V. Wind farm optimization by active yaw control strategy using machine learning approach. Int J Green Energy 2024;21:2628-39.
[18]

M’zoughi F, Lekube J, Garrido AJ, De La Sen M, Garrido I. Machine learning-based diagnosis in wave power plants for cost reduction using real measured experimental data: Mutriku Wave Power Plant. Ocean Eng 2024;293:116619.
[19]

Ahmed SF, Alam MSB, Hassan M, Rozbu MR, Ishtiak T, Rafa N, et al. Deep learning modeling techniques: current progress, applications, advantages, and challenges. Artif Intell Rev 2023;56:13521-617.
[20]

Lee N, Woo J, Kim S. A deep reinforcement learning ensemble for maintenance scheduling in offshore wind farms. Appl Energy 2025;377:124431.
[21]

Kavousi-Fard A, Dabbaghjamanesh M, Sheikh M, Jin T. A novel deep learning based digital twin model for mitigating wake effects in wind farms. Renew Energy Focus 2025;53:100686.
[22]

Wang Y, Zhao X, Li Z, Zhu W, Gui R. A novel hybrid model for multi-step-ahead forecasting of wind speed based on univariate data feature enhancement. Energy 2024;312:133515.
[23]

Wu F, Shen T, Bäck T, Chen J, Huang G, Jin Y, et al. Knowledge-empowered, collaborative, and co-evolving AI models: the post-LLM roadmap. Engineering 2025;44:87-100.
[24]

Fu T, Liu S, Li P. Intelligent smelting process, management system: efficient and intelligent management strategy by incorporating large language model. Front Eng Manag 2024;11:396-412.
[25]

Zhang J, Zhang C, Lu J, Zhao Y. Domain-specific large language models for fault diagnosis of heating, ventilation, and air conditioning systems by labeled-data-supervised fine-tuning. Appl Energy 2025;377:124378.
[26]

Liu Y, Zhou Y, Liu Y, Xu Z, He Y. Intelligent fault diagnosis for CNC through the integration of large language models and domain knowledge graphs. Engineering 2025;53:311-22.
[27]

Cui H, Guo X, Yu L. Leveraging pre-trained GPT models for equipment remaining useful life prognostics. Electronics 2025; 14(7):1265.
[28]

Ye W, Zhang Z, Li W, Qin Y, Zhang P. CRALA:a Chinese-centric risk simulation and assessment framework for LLM agents. In:Proceedings of the 2024 4th International Symposium on Artificial Intelligence and Intelligent Manufacturing (AIIM); 2024 Dec 20-22; Chengdu, China. New York City:IEEE; 2024. p. 561-6.
[29]

Addison L, Hosang A, Tuitt TA, Manohar K, Hosein P. Operations and Decisions ICTMOD; 2024 Nov 4-6; University City, Sharjah. A LLM-based platform for flood risk education and weather alerts in SIDS. Proceedings of the 2024 IEEE International Conference on Technology Management, New York City:IEEE; 2024. p. 1-6.
[30]

Hu J, Li Q, Fu N. Generative AI for materials discovery: design without understanding. Engineering 2024;39:13-7.
[31]

You Y, Tan K, Jiang Z, Zhang L. Developing a predictive platform for salmonella antimicrobial resistance based on a large language model and quantum computing. Engineering 2025;48:174-84.
[32]

Wang H, Gao C, Dantona C, Hull B, Sun J. DRG-LLaMA: tuning LLaMA model to predict diagnosis-related group for hospitalized patients. npj Digit Med 2024;7:16.
[33]

Liu Y, Li X, Luo Y, Du J, Zhang Y, Lv T, et al. Toward a large language model-driven medical knowledge retrieval and QA system: framework design and evaluation. Engineering 2025;50:270-82.
[34]

Zhang J, Li Y, Liu Y, Xie L. 2024 Dec 6-8; Cambridge UK. Research and practice of NL2SQL technology based on LLM for big data of enterprise finance. Proceedings of the 2024 4th International Conference on Advanced Enterprise Information System (AEIS); New York City:IEEE; 2024. p. 46-51.
[35]

Wei F, Keeling R, Huber-Fliflet N, Zhang J, Dabrowski A, Yang J, et al. 2023 Dec 15-18; Sorrento, Italy. Empirical study of LLM fine-tuning for text classification in legal document review. Proceedings of the 2023 IEEE International Conference on Big Data (BigData); New York City:IEEE; 2023. p. 2786-92.
[36]

Bock S, Bomsdorf S, Boysen N, Schneider M. A survey on the Traveling Salesman Problem and its variants in a warehousing context. Eur J Oper Res 2025;322:1-14.
[37]

Liu M, Feng Q, Fan D, Dui H, Sun B, Ren Y, et al. Resilience importance measure and optimization considering the stepwise recovery of system performance. IEEE Trans Reliab 2023;72:1064-77.
[38]

Fan D, Feng Q, Zhang A, Liu M, Ren Y, Wang Y. Optimization of scheduling and timetabling for multiple electric bus lines considering nonlinear energy consumption model. IEEE Trans Intell Transport Syst 2024;25:5342-55.
[39]

Sun C, Huang S, Pompili D. LLM-based multi-agent decision-making: challenges and future directions. IEEE Robot Autom Lett 2025; 10(6):5681-8.
[40]

Lin Z. How to write effective prompts for large language models. Nat Hum Behav 2024;8:611-5.
[41]

Liu P, Yuan W, Fu J, Jiang Z, Hayashi H, Neubig G. Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing. ACM Comput Surv 2023; 55(9):195:1-35.
[42]

Hu EJ, Shen Y, Wallis P, Allen-Zhu Z, Li Y, Wang S, et al. LoRA: low-rank adaptation of large language models. arXiv.2106.09685.
[43]

Yang A, Yang B, Zhang B, Hui B, Zheng B, Yu B, et al. Qwen2.5 technical report 2025. arXiv.2412.15115.
[44]

Han L, Zhang Y, Gao W, Li X. LexSage:multi-task optimization in legal large language model applications. In:Proceedings of the 2024 4th International Conference on Artificial Intelligence, Robotics, and Communication (ICAIRC); 2024 Dec 27-29; Xiamen, China. New York City:IEEE; 2024. p. 325-30.
[45]

Abacha AB, Yim W, Fu Y, Sun Z, Yetisgen M, Xia F, et al. MEDEC: a benchmark for medical error detection and correction in clinical notes. arXiv.2412.19260.
[46]

Fan D, Ren Y, Feng Q, Zhu B, Liu Y, Wang Z. A hybrid heuristic optimization of maintenance routing and scheduling for offshore wind farms. J Loss Prev Process Ind 2019;62:103949.
[47]

Zhang ZY. Scheduling and routing optimization of maintenance fleet for offshore wind farms using duo-ACO. Adv Mater Res 2014;1039:294-301.
[48]

Fan D, Zhang A, Feng Q, Cai B, Liu Y, Ren Y. Group maintenance optimization of subsea Xmas trees with stochastic dependency. Reliab Eng Syst Saf 2021;209:107450.
[49]

OpenAI.Pricing [Internet]. San Francisco: OpenAI; undated [cited 2025 Sep 9]. Available from:

[50]

Anthropic PBC. Pricing Internet. San Francisco: Anthropic PBC; undated [cited 2025 Sep 9]. Available from:

[51]

DeepSeeks.Models & pricing [Internet]. Hangzhou: DeepSeeks; undated [cited 2025 Sep 9]. Available from:

[52]

Alibaba Cloud. Alibaba cloud model studio [Internet] Hangzhou: Alibaba Cloud; undated [cited 2025 Sep 9]. Available from:

PDF (5587KB)

0

Accesses

0

Citation

Detail
Sections
Recommended

/

〈 〉