PDF(726 KB)
人工智能大模型在电力设备运维场景中的应用探讨
陈晓红, 傅文润, 刘朝明, 刘泽洪, 李俊朋, 胡志亮, 胡东滨
中国工程科学 ›› 2025, Vol. 27 ›› Issue (1) : 180-192.
PDF(726 KB)
PDF(726 KB)
人工智能大模型在电力设备运维场景中的应用探讨
Application of Artificial Intelligence Large Language Model in Power Equipment Operation and Maintenance
电力设备运维是新型电力系统建设的重要环节，以人工智能（AI）大模型技术为代表的AI技术变革为传统电力设备运维的数智化提供了新机遇。本文探讨了多模态AI大模型对电力设备健康状态评估、电力设备运行状态预测、电力设备故障诊断、电力设备寿命预测、电力设备故障检修策略推荐等电力运维具体场景的赋能作用，辨识了数据问题制约电力AI大模型的应用成效、“算法黑箱”影响智能运维辅助决策的透明度与可靠性、环境变化导致电力AI大模型性能衰退等多模态AI大模型赋能电力设备运维的技术难点。着眼攻克相关技术难点，结合知识图谱检索增强生成、多模态对齐、微调和持续学习等大模型应用优化技术，构建了基于多模态AI大模型的电力设备运维系统架构，梳理了多模态AI大模型在电力设备运维场景应用时涉及的需求分析、模型训练、应用部署、运营管理等主要阶段的实现过程，进而提出了持续监控并优化数据质量、采用持续学习算法、建立模型性能反馈循环机制等大模型性能持续优化策略。进一步探讨了多模态AI大模型赋能电力设备运维的应用趋势和发展保障举措，以深化对电力设备智能运维领域的前沿技术认知，推动构建智能化、智慧化的新型电力系统。
The operation and maintenance of power equipment is a crucial aspect of the construction of new power systems. The artificial intelligence large language model (AI-LLM) presents significant opportunities for the digital intelligence of traditional power equipment operation and maintenance. This study aims to explore the enabling role of multimodal AI-LLM in health assessment, operational state prediction, fault diagnosis, life prediction, and maintenance strategy recommendation, among other specific scenarios of power equipment operation and maintenance. Additionally, this study analyzes the challenges faced by multimodal AI-LLM in enabling power equipment operation and maintenance, including the varying quality of multimodal data, the "black box" nature of algorithms leading to low transparency in decision-making processes, and model performance deterioration induced by environmental changes. To address these challenges, this study combines knowledge graph retrieval-augmented generation, multimodal alignment, fine-tuning and continuous learning, and other big model application optimization techniques to construct an AI-LLM power equipment maintenance system. It then sorts out the implementation process of multimodal AI-LLM in the operation and maintenance of power equipment, covering four stages: demand analysis, model training, application deployment, and operational management. Furthermore, strategies for continuously optimizing model performance are proposed, including the continuous monitoring and optimization of data quality, use of continuous learning algorithms, and establishment of a feedback loop mechanism for model performance. Finally, this study explores the future directions for multimodal AI-LLM in the field of power equipment operation and maintenance and provides a series of implementation safeguards to promote the intelligent transformation of power equipment operation and maintenance and support the construction of new power systems.
新型电力系统 / 电力设备运维 / 多模态AI大模型 / 检索增强生成 / 知识图谱
new power system / power equipment operation and maintenance / multimodal artificial intelligence large language model / retrieval-augmented generation / knowledge graph
| [1] |
陈晓红, 曹廖滢, 陈姣龙, 等. 我国算力发展的需求、电力能耗及绿色低碳转型对策 [J]. 中国科学院院刊, 2024, 39(3): 528‒539.
Chen X H, Cao L Y, Chen J L, et al. Development demand, power energy consumption and green and low-carbon transition for computing power in China [J]. Bulletin of Chinese Academy of Sciences, 2024, 39(3): 528‒539.
|
| [2] |
盛戈皞, 钱勇, 罗林根, 等. 面向新型电力系统的数字化电力设备关键技术及其发展趋势 [J]. 高电压技术, 2023, 49(5): 1765‒1778.
Sheng G H, Qian Y, Luo L G, et al. Key technologies and development trends of digital power equipment for new type power system [J]. High Voltage Engineering, 2023, 49(5): 1765‒1778.
|
| [3] |
饶宏, 韩丰, 陈政, 等. 我国电力安全供应保障策略研究 [J]. 中国工程科学, 2023, 25(2): 100‒110.
Rao H, Han F, Chen Z, et al. Strategy for guaranteeing power supply security of China [J]. Strategic Study of CAE, 2023, 25(2): 100‒110.
|
| [4] |
李旭斌, 田付强, 郭亦可. 新型电力系统中电力设备健康管理与智能运维关键技术探究 [J]. 电网技术, 2023, 47(9): 3710‒3727.
Li X B, Tian F Q, Guo Y K. Key technologies for health management and intelligent operation and maintenance of power equipment in new power systems [J]. Power System Technology, 2023, 47(9): 3710‒3727.
|
| [5] |
陈晓红. 打造AI大模型创新应用高地 [EB/OL]. (2023-12-04)[2023-12-25]. http://lw.news.cn/2023-12/04/c_1310 753536.htm.
Chen X H. Create a highland for the innovation and application of AI large language models [EB/OL]. (2023-12-04)[2023-12-25]. http://lw.news.cn/202312/04/c_13107535 36.htm.
|
| [6] |
李刚, 方鸿, 刘云鹏, 等. 新型电力系统中的大模型驱动技术: 现状、机遇与挑战 [J]. 高电压技术, 2024, 50(7): 2864‒2878.
Li G, Fang H, Liu Y P, et al. Large-model drive technology in new power system: Status, challenges and prospects [J]. High Voltage Engineering, 2024, 50(7): 2864‒2878.
|
| [7] |
赵俊华, 文福拴, 黄建伟, 等. 基于大语言模型的电力系统通用人工智能展望: 理论与应用 [J]. 电力系统自动化, 2024, 48(6): 13‒28.
Zhao J H, Wen F S, Huang J W, et al. Prospect of artificial general intelligence for power systems based on large language model: Theory and applications [J]. Automation of Electric Power Systems, 2024, 48(6): 13‒28.
|
| [8] |
王有元, 陈璧君. 基于层次分析结构的变压器健康状态与综合寿命评估模型 [J]. 电网技术, 2014, 38(10): 2845‒2850.
Wang Y Y, Chen B J. An integrated life estimation model of power transformer based on hierarchical architecture and health index [J]. Power System Technology, 2014, 38(10): 2845‒2850.
|
| [9] |
吴奕, 朱海兵, 周志成, 等. 基于熵权模糊物元和主元分析的变压器状态评价 [J]. 电力系统保护与控制, 2015, 43(17): 1‒7.
Wu Y, Zhu H B, Zhou Z C, et al. Transformer condition assessment based on entropy fuzzy matter-element and principal component analysis [J]. Power System Protection and Control, 2015, 43(17): 1‒7.
|
| [10] |
张晗, 常安, 王奇, 等. 基于多源信息的变压器状态评价方案研究与应用 [J]. 高压电器, 2016, 52(2): 19‒27.
Zhang H, Chang A, Wang Q, et al. Research and application of transformer condition evaluation model based on multi-source information [J]. High Voltage Apparatus, 2016, 52(2): 19‒27.
|
| [11] |
胡军, 尹立群, 李振, 等. 基于大数据挖掘技术的输变电设备故障诊断方法 [J]. 高电压技术, 2017, 43(11): 3690‒3697.
Hu J, Yin L Q, Li Z, et al. Fault diagnosis method of transmission and transformation equipment based on big data mining technology [J]. High Voltage Engineering, 2017, 43(11): 3690‒3697.
|
| [12] |
孙鹏, 耿苏杰, 王秀利. 基于时效评分函数和贝叶斯概率的电力变压器状态实时评估 [J]. 高电压技术, 2018, 44(4): 1069‒1077.
Sun P, Geng S J, Wang X L. Real-time condition assessment of power transformers based on time-effect score function and Bayesian probability [J]. High Voltage Engineering, 2018, 44(4): 1069‒1077.
|
| [13] |
Tao L F, Li S Y, Liu H F, et al. An outline of prognostics and health management large model: Concepts, paradigms, and challenges [EB/OL]. (2024-06-01)[2024-12-25]. https://arxiv.org/abs/2407.03374v1.
|
| [14] |
Jose S, Nguyen K T P, Medjaher K, et al. Advancing multimodal diagnostics: Integrating industrial textual data and domain knowledge with large language models [J]. Expert Systems with Applications, 2024, 255: 124603.
|
| [15] |
李莉, 时榕良, 郭旭, 等. 融合大模型与图神经网络的电力设备缺陷诊断 [J]. 计算机科学与探索, 2024, 18(10): 2643‒2655.
Li L, Shi R L, Guo X, et al. Diagnosis of power system defects by large language models and graph neural networks [J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(10): 2643‒2655.
|
| [16] |
Pang Z D, Luan Y X, Chen J H, et al. ParInfoGPT: An LLM-based two-stage framework for reliability assessment of rotating machine under partial information [J]. Reliability Engineering & System Safety, 2024, 250: 110312.
|
| [17] |
Trappey A J C, Trappey C V, Ma L, et al. Intelligent engineering asset management system for power transformer maintenance decision supports under various operating conditions [J]. Computers & Industrial Engineering, 2015, 84: 3‒11.
|
| [18] |
Qi B, Wang Y M, Zhang P, et al. A novel deep recurrent belief network model for trend prediction of transformer DGA data [J]. IEEE Access, 2019, 7: 80069‒80078.
|
| [19] |
Orille-Fernandez A L, Khalil N, Rodriguez S. Failure risk prediction using artificial neural networks for lightning surge protection of underground MV cables [J]. IEEE Transactions on Power Delivery, 2006, 21(3): 1278‒1282.
|
| [20] |
Li Y F, Wang H, Sun M X. ChatGPT-like large-scale foundation models for prognostics and health management: A survey and roadmaps [J]. Reliability Engineering & System Safety, 2024, 243: 109850.
|
| [21] |
Demirci M, Gözde H, Taplamacioglu M C. Improvement of power transformer fault diagnosis by using sequential Kalman filter sensor fusion [J]. International Journal of Electrical Power & Energy Systems, 2023, 149: 109038.
|
| [22] |
Ali M Z, Shabbir M N S K, Liang X D, et al. Machine learning-based fault diagnosis for single- and multi-faults in induction motors using measured stator currents and vibration signals [J]. IEEE Transactions on Industry Applications, 2019, 55(3): 2378‒2391.
|
| [23] |
Alsaif K M, Albeshri A A, Khemakhem M A, et al. Multimodal large language model-based fault detection and diagnosis in context of industry 4.0 [J]. Electronics, 2024, 13(24): 4912.
|
| [24] |
Zheng S W, Pan K, Liu J, et al. Empirical study on fine-tuning pre-trained large language models for fault diagnosis of complex systems [J]. Reliability Engineering & System Safety, 2024, 252: 110382.
|
| [25] |
Wang H, Li C X, Li Y F, et al. An intelligent industrial visual monitoring and maintenance framework empowered by large-scale visual and language models [J]. IEEE Transactions on Industrial Cyber-Physical Systems, 2024, 2: 166‒175.
|
| [26] |
吴细秀, 冷宇宽, 庞文龙, 等. GIS母线潜伏性故障电场劣化特性与故障概率预测研究 [J]. 电网技术, 2023, 47(7): 3018‒3028.
Wu X X, Leng Y K, Pang W L, et al. Electric field degradation characteristics and fault probability prediction of GIS busbar latent faults [J]. Power System Technology, 2023, 47(7): 3018‒3028.
|
| [27] |
Ferreira C, Gonçalves G. Remaining useful life prediction and challenges: A literature review on the use of machine learning methods [J]. Journal of Manufacturing Systems, 2022, 63: 550‒562.
|
| [28] |
Cen Z L, Hu S L, Hou Y D, et al. Remaining useful life prediction of machinery based on improved Sample Convolution and Interaction Network [J]. Engineering Applications of Artificial Intelligence, 2024, 135: 108813.
|
| [29] |
Chen Y, Liu C. Remaining useful life prediction: A study on multidimensional industrial signal processing and efficient transfer learning based on large language models [EB/OL]. (2024-10-04)[2024-12-15]. https://arxiv.org/abs/2410.03134v1.
|
| [30] |
Wang P W, Niu S Z, Cui H L, et al. GPT-based equipment remaining useful life prediction [R]. Changsha: The ACM Turing Award Celebration Conference-China 2024, 2024.
|
| [31] |
徐波, 韩学山, 李业勇, 等. 电力设备机会维修决策模型 [J]. 中国电机工程学报, 2016, 36(23): 6379‒6388, 6603.
Xu B, Han X S, Li Y Y, et al. Opportunistic maintenance decision-making model for power equipment [J]. Proceedings of the CSEE, 2016, 36(23): 6379‒6388, 6603.
|
| [32] |
魏少岩, 徐飞, 闵勇. 输电线路检修计划模型 [J]. 电力系统自动化, 2006, 30(17): 41‒44, 49.
Wei S Y, Xu F, Min Y. Modeling on maintenance scheduling of transmission lines [J]. Automation of Electric Power Systems, 2006, 30(17): 41‒44, 49.
|
| [33] |
刘文颖, 谢昶, 文晶, 等. 基于小生境多目标粒子群算法的输电网检修计划优化 [J]. 中国电机工程学报, 2013, 33(4): 141‒148.
Liu W Y, Xie C, Wen J, et al. Optimization of transmission network maintenance scheduling based on niche multi-objective particle swarm algorithm [J]. Proceedings of the CSEE, 2013, 33(4): 141‒148.
|
| [34] |
张金营, 王天堃, 么长英, 等. 基于大语言模型的电力知识库智能问答系统构建与评价 [J]. 计算机科学, 2024, 51(12): 286‒292.
Zhang J Y, Wang T K, Yao C Y, et al. Construction and evaluation of intelligent question answering system for electric power knowledge base based on large language model [J]. Computer Science, 2024, 51(12): 286‒292.
|
| [35] |
Kohl L, Eschenbacher S, Besinger P, et al. Large language model-based chatbot for improving human-centricity in maintenance planning and operations [J]. PHM Society European Conference, 2024, 8(1): 12.
|
/
| 〈 |
|
〉 |
