DeepSeek在储能研究中的应用前景展望

doi:10.19799/j.cnki.2095-4239.2025.0189

储能科学与技术 ›› 2025, Vol. 14 ›› Issue (2): 467-478.doi: 10.19799/j.cnki.2095-4239.2025.0189

DeepSeek在储能研究中的应用前景展望

高宇辰(), 李蔚林(), 陈翔(), 袁誉杭, 牛艺琳, 张强()

清华大学化学工程系，复合固态电池北京市重点实验室，绿电化工研究中心，北京 100084

收稿日期:2025-02-28 修回日期:2025-03-03 出版日期:2025-02-28 发布日期:2025-03-18
通讯作者: 陈翔,张强 E-mail:gyc22@mails.tsinghua.edu.cn;liwl22@mails.tsinghua.edu.cn;xiangchen@mail.tsinghua.edu.cn;zhang-qiang@mails.tsinghua.edu.cn
作者简介:高宇辰（2000—），男，博士研究生，研究方向为锂电池材料人工智能研究，E-mail：gyc22@mails.tsinghua.edu.cn
李蔚林（2004—），男，本科生，研究方向为锂键化学与人工智能，E-mail：liwl22@mails.tsinghua.edu.cn；
基金资助:
国家自然科学基金项目(T2322015);国家重点研发计划项目(2021YFB2500300);科学探索奖

A perspective on DeepSeek application in energy storage research

Yuchen GAO(), Weilin LI(), Xiang CHEN(), Yuhang YUAN, Yilin NIU, Qiang ZHANG()

Beijing Key Laboratory of Complex Solid State Batteries and Tsinghua Center for Green Chemical Engineering Electrification (CCEE), Department of Chemical Engineering, Tsinghua University, Beijing 100084, China

Received:2025-02-28 Revised:2025-03-03 Online:2025-02-28 Published:2025-03-18
Contact: Xiang CHEN, Qiang ZHANG E-mail:gyc22@mails.tsinghua.edu.cn;liwl22@mails.tsinghua.edu.cn;xiangchen@mail.tsinghua.edu.cn;zhang-qiang@mails.tsinghua.edu.cn

摘要/Abstract

摘要：

在现代能源体系中，化石能源正逐步向可再生能源转型，能源存储将成为新型电力系统的关键调节单元，但这一进程面临研发低效、系统优化复杂、安全管控滞后以及市场机制不完善等多重挑战。深度求索(DeepSeek)大模型凭借其低能耗、高能效以及卓越的推理能力，为突破储能领域关键瓶颈开辟了新路径。DeepSeek通过采用多头潜在注意力、混合专家模型及多词元预测等核心技术，显著降低了模型训练与推理的能耗成本，展现出在储能研究领域的广泛应用前景，有望推动材料研发从“经验试错”到“智能设计”的范式跃迁，在系统优化中构建多尺度耦合的数字孪生底座，在安全管控中推动被动响应向主动预警的模式转型，在政策分析中建立数据驱动的市场动态评估体系。本文提出“系统共生、能效共进”的发展模式，为人工智能与清洁能源技术的深度融合构建了技术基座，有望加速零碳算力基础设施的构建，引领储能技术迈向智能化新纪元。

关键词: 深度求索大模型, 大语言模型, 人工智能, 储能技术

Abstract:

During the global energy system's transition to renewable energy, energy storage technology has emerged as the core regulatory unit of new power systems, yet it faces multifaceted challenges, including inefficient material development, complex system optimization, lagging safety management, and imperfect market mechanisms. The DeepSeek large language model, with its low energy consumption, high efficiency, and exceptional reasoning capabilities, proffers an innovative pathway to address critical bottlenecks in energy storage. Through core technologies such as multi-head latent attention, DeepSeek mixture-of-experts models, and multi-token prediction, DeepSeek significantly reduces energy costs in both model training and inference. Its broad application prospects in energy storage research are expected to drive a paradigm shift from “trial-and-error” to “intelligent design” in materials development, establish multi-scale coupled digital twin frameworks for system optimization, transform safety management from passive response to proactive early warning, and create data-driven dynamic market evaluation systems for policy analysis. The “system symbiosis and energy-efficiency co-evolution” development paradigm provides a technological foundation for the deep integration of artificial intelligence with clean energy technologies, potentially accelerating the construction of carbon-neutral computing infrastructure and ushering energy storage technology into an intelligent new era.

Key words: DeepSeek, large language model, artificial intelligence, energy storage technology

中图分类号:

TP 18

高宇辰, 李蔚林, 陈翔, 袁誉杭, 牛艺琳, 张强. DeepSeek在储能研究中的应用前景展望[J]. 储能科学与技术, 2025, 14(2): 467-478.

Yuchen GAO, Weilin LI, Xiang CHEN, Yuhang YUAN, Yilin NIU, Qiang ZHANG. A perspective on DeepSeek application in energy storage research[J]. Energy Storage Science and Technology, 2025, 14(2): 467-478.

图/表 8

图1

图2

图3

图4

图5

表1

图6

表2

DeepSeek与GPT-4o提出600 Wh/kg全固态金属锂电池设计方案对比"

设计角度		DeepSeek-R1		GPT-4o
核心材料	正极材料	富锂锰基氧化物 $≥$ 300 mAh/g		富锂镍钴铝氧化物镍钴锰氧化物钴酸锂
	负极材料	锂金属负极 3860 mAh/g		锂金属负极
	电解质	硫化物电解质(Li₁₀GeP₂S₁₂) $≥$ 300 mS/cm		硫化物电解质氧化物电解质聚合物电解质

界面优化		正极	三维互穿 Li₃PO₄-Li₂S梯度涂层	界面涂层优化固态电解质-电极的界面
界面优化		负极	锂合金缓冲层	界面涂层优化固态电解质-电极的界面

电池结构		双极堆叠		薄膜电池设计提高电导率
		超薄集流体
		质量分配优化正极活性物质占比 $≥$ 70%

理论能量密度计算		610 Wh/kg		无

表2

参考文献 53

1	AGENCY I E. Net Zero by 2050: A Roadmap for the Global Energy Sector[M]. OECD, 2021. DOI:10.1787/c8328405-en.
2	国家能源局. 2024年可再生能源并网运行情况 [EB/OL]. (2025-01-27). [2025-02-13]. https://video.cpnn.com.cn/news/nytt/202501/t20250127_1769671.html
3	代宇涵, 刘春, 周朋, 等. 双碳背景下电力系统储能技术的应用与研究进展[J]. 储能科学与技术, 2024, 13(8): 2772-2774. DOI: 10.19799/j.cnki.2095-4239.2024.0695.
	DAI Y H, LIU C, ZHOU P, et al. Application and research progress of energy storage technology in power systems under the dual carbon background[J]. Energy Storage Science and Technology, 2024, 13(8): 2772-2774. DOI: 10.19799/j.cnki.2095-4239.2024.0695.
4	中关村储能产业技术联盟. 2024年度cnesa储能数据重磅发布 [EB/OL]. (2025-01-15). [2025-02-13]. https://mp.weixin.qq.com/s/ytgUGjJcdBq3pHI9vby51A
5	张继阳, 郑秀, 赵斌, 等. 电网级大规模储能的电池技术进展[J]. 电池, 2024, 54(5): DOI: 10.19535/j.1001-1579.2024.05.029
	ZHANG J Y, ZHENG X, ZHAO B, et al. Progress in battery technology for large-scale grid-level energy storage[J]. Battery Bimonthly, 2024, 54(5): DOI: 10.19535/j.1001-1579.2024.05.029
6	李政, 李伟起, 张忠伟, 等. "双碳" 目标下我国电力系统灵活性资源发展策略研究[J]. 中国工程科学, 2024, 26(4): 108-120. DOI: 10.15302/J-SSCAE-2024.04.018.
	LI Z, LI W Q, ZHANG Z W, et al. Development strategy of flexible resources in China's power system under the carbon peaking and carbon neutrality goals[J]. Strategic Study of CAE, 2024, 26(4): 108-120. DOI: 10.15302/J-SSCAE-2024.04.018.
7	BENAYAD A, DIDDENS D, HEUER A, et al. High-throughput experimentation and computational freeway lanes for accelerated battery electrolyte and interface development research[J]. Advanced Energy Materials, 2022, 12(17): 2102678. DOI:10. 1002/aenm.202102678.
8	LOMBARDO T, DUQUESNOY M, EL-BOUYSIDY H, et al. Artificial intelligence applied to battery research: Hype or reality?[J]. Chemical Reviews, 2022, 122(12): 10899-10969. DOI:10.1021/acs.chemrev.1c00108.
9	LI Z, HOU L P, YAO N, et al. Correlating polysulfide solvation structure with electrode kinetics towards long-cycling lithium-sulfur batteries[J]. Angewandte Chemie (International Ed), 2023, 62(43): e202309968. DOI:10.1002/anie.202309968.
10	LI Z, YU L G, BI C X, et al. A three-way electrolyte with ternary solvents for high-energy-density and long-cycling lithium-sulfur pouch cells[J]. SusMat, 2024, 4(2): e200. DOI:10.1002/sus2.200.
11	黄海泉, 黄晓巍, 姜望, 等. 新型配电网分布式储能系统方案及配置研究综述[J]. 南方能源建设, 2024, 11(4): 42-53. DOI: 10.16516/j.ceec.2024.4.05.
	HUANG H Q, HUANG X W, JIANG W, et al. A review of distributed energy storage system solutions and configurations for new distribution grids[J]. Southern Energy Construction, 2024, 11(4): 42-53. DOI: 10.16516/j.ceec.2024.4.05.
12	FAN X L, JI X, CHEN L, et al. All-temperature batteries enabled by fluorinated electrolytes with non-polar solvents[J]. Nature Energy, 2019, 4: 882-890. DOI:10.1038/s41560-019-0474-3.
13	沈馨, 张睿, 赵辰孜, 等. 金属锂电池中力-电化学机制研究进展[J]. 储能科学与技术, 2022, 11(9): 2781-2797. DOI: 10.19799/j.cnki.2095-4239.2022.0326.
	SHEN X, ZHANG R, ZHAO C Z, et al. Recent advances in mechano-electrochemistry in lithium metal batteries[J]. Energy Storage Science and Technology, 2022, 11(9): 2781-2797. DOI: 10.19799/j.cnki.2095-4239.2022.0326.
14	黄家辉, 邝祝芳. 人工智能与储能技术融合的前沿发展[J]. 储能科学与技术, 2024, 13(9): 3161-3181. DOI: 10.19799/j.cnki.2095-4239.2024.0575.
	HUANG J H, KUANG Z F. The forefront of the integration of artificial intelligence and energy storage technologies[J]. Energy Storage Science and Technology, 2024, 13(9): 3161-3181. DOI: 10.19799/j.cnki.2095-4239.2024.0575.
15	袁誉杭, 高宇辰, 张俊东, 等. 大语言模型在储能研究中的应用[J]. 储能科学与技术, 2024, 13(9): 2907-2919. DOI: 10.19799/j.cnki.2095-4239.2024.0176.
	YUAN Y H, GAO Y C, ZHANG J D, et al. The application of large language models in energy storage research[J]. Energy Storage Science and Technology, 2024, 13(9): 2907-2919. DOI: 10.19799/j.cnki.2095-4239.2024.0176.
16	CHEN X, LIU X Y, SHEN X, et al. Applying machine learning to rechargeable batteries: From the microscale to the macroscale[J]. Angewandte Chemie, 2021, 133(46): 24558-24570. DOI:10.1002/ange.202107369.
17	YAO N, CHEN X, FU Z H, et al. Applying classical, Ab initio, and machine-learning molecular dynamics simulations to the liquid electrolyte for rechargeable batteries[J]. Chemical Reviews, 2022, 122(12): 10970-11021. DOI:10.1021/acs.chemrev.1c00904.
18	GAO Y C, YAO N, CHEN X, et al. Data-driven insight into the reductive stability of ion-solvent complexes in lithium battery electrolytes[J]. Journal of the American Chemical Society, 2023, 145(43): 23764-23770. DOI:10.1021/jacs.3c08346.
19	GAO Y C, YUAN Y H, HUANG S Z, et al. A knowledge-data dual-driven framework for predicting the molecular properties of rechargeable battery electrolytes[J]. Angewandte Chemie (International Ed), 2025, 64(4): e202416506. DOI:10.1002/anie.202416506.
20	清华四川能源互联网研究院. 清华四川院助力江苏首座AI智慧调控光储充换一体化站建设[EB/OL]. (2025-01-06). [2025-02-13]. https://mp.weixin.qq.com/s/hVgxfkTv6bAlRpCAjddWjA
21	WANG Y, FENG X N, GUO D X, et al. Temperature excavation to boost machine learning battery thermochemical predictions[J]. Joule, 2024, 8(9): 2639-2651. DOI:10.1016/j.joule.2024.07.002.
22	JAIN M, SUN X Q, DATTA S, et al. A machine learning framework to deconstruct the primary drivers for electricity market price events[C]//2023 IEEE Power & Energy Society General Meeting (PESGM). July 16-20, 2023, Orlando, FL, USA. IEEE, 2023: 1-5. DOI:10.1109/PESGM52003.2023.10252752.
23	PATTERSON D, GONZALEZ J, LE Q, et al. Carbon emissions and large neural network training[EB/OL]. 2021: 2104.10350. https://arxiv.org/abs/2104.10350v3.
24	WILSON L. Average household electricity consumption-2025[EB/OL]. (2025). [2025-02-13]. https://shrinkthatfootprint.com/average-household-electricity-consumption/
25	ARGERICH M F, PATIÑO-MARTÍNEZ M. Measuring and improving the energy efficiency of large language models inference[J]. IEEE Access, 2024, 12: 80194-80207. DOI:10.1109/ACCESS.2024.3409745.
26	OPENAI. AI and compute[EB/OL]. (2018-05-16). [2025-02-13]. https://openai.com/index/ai-and-compute/
27	AGENCY I E. Electricity 2024[R]. Paris: International Energy Agency, 2024.
28	COMPANY M. Global energy perspective 2024[R]. McKinsey & Company, 2024.
29	DEEPSEEK-AI. DeepSeek-v3 technical report[EB/OL]. (2025-02-18). [2025-02-25]. https://arxiv.org/abs/2412.19437
30	RAY PERRAULT, CLARK J. Artificial intelligence index report 2024[R]. USA: Stanford University Human-Cenrered Artificial Intelligence, 2024.
31	LIU A, FENG B, XUE B, et al. DeepSeek-v3 technical report[J]. arXiv preprint arXiv:, 2024.
32	DeepSeek-AI, GUO D Y, YANG D J, et al. DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning[EB/OL]. 2025: 2501.12948. https://arxiv.org/abs/2501.12948v1.
33	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [J]. Advances in neural information processing systems, 2017, 30.
34	RAMIJ J, NOWLAN S, HINTON G. Adaptive mixture of local expert[J]. Neural Comput, 1991, 3: 79-87.
35	SHAZEER N, MIRHOSEINI A, MAZIARZ K, et al. Outrageously large neural networks: The sparsely-gated mixture-of-experts layer[EB/OL]. 2017: 1701.06538. https://arxiv.org/abs/1701. 06538v1.
36	SHEN S, HOU L, ZHOU Y Q, et al. Mixture-of-experts meets instruction tuning: A winning combination for large language models[C]//International Conference on Learning Representations, 2023.
37	WANG L A, GAO H Z, ZHAO C G, et al. Auxiliary-loss-free load balancing strategy for mixture-of-experts[R/OL]. 2024. https://ui.adsabs.harvard.edu/abs/2024arXiv240815664W/abstract.
38	GLOECKLE F, IDRISSI B Y, ROZIèRE B, et al. Better & faster large language models via multi-token prediction[EB/OL]. (2024-04-30). [2025-02-27]. https://arxiv.org/abs/2404.19737.
39	LI N, CHEN Z P, REN W C, et al. Flexible graphene-based lithium ion batteries with ultrafast charge and discharge rates[J]. Proceedings of the National Academy of Sciences of the United States of America, 2012, 109(43): 17360-17365. DOI:10.1073/pnas.1210072109.
40	WU X S, LIU T, LEE Y G, et al. Glycerol triacetate-based flame retardant high-temperature electrolyte for the lithium-ion battery[J]. ACS Applied Materials & Interfaces, 2024, 16(19): 24590-24600. DOI:10.1021/acsami.4c02323.
41	LI G X, KOVERGA V, NGUYEN A, et al. Enhancing lithium-metal battery longevity through minimized coordinating diluent[J]. Nature Energy, 2024, 9: 817-827. DOI:10.1038/s41560-024-01519-5.
42	CHOI I R, CHEN Y L, SHAH A, et al. Asymmetric ether solvents for high-rate lithium metal batteries[J]. Nature Energy, 2025. DOI:10.1038/s41560-025-01716-w.
43	ZOU C F, YANG L, LUO K L, et al. Performance improvement of Li₆PS₅Cl solid electrolyte modified by poly(ethylene oxide)-based composite polymer electrolyte with ZSM-5 molecular sieves[J]. ACS Applied Energy Materials, 2022, 5(2): 2356-2365. DOI:10.1021/acsaem.1c03819.
44	CUI L F, ZHANG S, JU J W, et al. A cathode homogenization strategy for enabling long-cycle-life all-solid-state lithium batteries[J]. Nature Energy, 2024, 9: 1084-1094. DOI:10.1038/s41560-024-01596-6.
45	KRAFT M A, OHNO S, ZINKEVICH T, et al. Inducing high ionic conductivity in the lithium superionic argyrodites Li₆₊ xP_1– xGexS₅I for all-solid-state batteries[J]. Journal of the American Chemical Society, 2018, 140(47): 16330-16339. DOI:10.1021/jacs.8b10282.
46	ZHOU W D, LI Y T, XIN S, et al. Rechargeable sodium all-solid-state battery[J]. ACS Central Science, 2017, 3(1): 52-57. DOI:10.1021/acscentsci.6b00321.
47	GAO X, YU Z A, WANG J Y, et al. Electrolytes with moderate lithium polysulfide solubility for high-performance long-calendar-life lithium-sulfur batteries[J]. Proceedings of the National Academy of Sciences of the United States of America, 2023, 120(31): e2301260120. DOI:10.1073/pnas.2301260120.
48	LUO C, HU E Y, GASKELL K J, et al. A chemically stabilized sulfur cathode for lean electrolyte lithium sulfur batteries[J]. Proceedings of the National Academy of Sciences of the United States of America, 2020, 117(26): 14712-14720. DOI:10.1073/pnas.2006301117.
49	KAMAYA N, HOMMA K, YAMAKAWA Y, et al. A lithium superionic conductor[J]. Nature Materials, 2011, 10(9): 682-686. DOI:10.1038/nmat3066.
50	ZHU L, WANG Y W, CHEN J C, et al. Enhancing ionic conductivity in solid electrolyte by relocating diffusion ions to under-coordination sites[J]. Science Advances, 2022, 8(11): eabj7698. DOI:10.1126/sciadv.abj7698.
51	JANEK J, ZEIER W G. Challenges in speeding up solid-state battery development[J]. Nature Energy, 2023, 8: 230-240. DOI:10.1038/s41560-023-01208-9.
52	LIU L L, XU J R, WANG S, et al. Practical evaluation of energy densities for sulfide solid-state batteries[J]. eTransportation, 2019, 1: 100010. DOI:10.1016/j.etran.2019.100010.
53	KATO Y, HORI S, KANNO R. Li₁₀GeP₂S₁₂-type superionic conductors: Synthesis, structure, and ionic transportation[J]. Advanced Energy Materials, 2020, 10(42): 2002153. DOI:10.1002/aenm.202002153.

Question	DeepSeek-R1	DeepSeek-V3	GPT-o1	GPT-4o	Qwen-plus
total	89	81	83	71	76
Q1	9	8	8	9	8
Q2	9	7	8	7	8
Q3	9	8	7	7	7
Q4	9	9	8	8	9
Q5	9	9	9	7	8
Q6	9	9	8	8	8
Q7	8	8	6	6	8
Q8	8	7	7	7	7
Q9	9	8	8	7	8
Q10	10	8	9	5	5

[1]	梁振飞, 王兴兴, 胡皓晨, 李艳红, 欧阳博学, 孙晓云, 高瑞茂, 叶骏, 王德仁. 锌溴液流电池电解液与隔膜技术研究进展[J]. 储能科学与技术, 2025, 14(2): 583-600.
[2]	卢鸣. 基于数据挖掘的热能储存异常数据检测方法[J]. 储能科学与技术, 2025, 14(2): 831-833.
[3]	许堉坤, 俞钧, 蒋超, 王婧骅, 赵婉茹. 成本角度分析储能技术与电力系统的优化设计[J]. 储能科学与技术, 2025, 14(2): 876-878.
[4]	孙爱民, 洪建军, 郑建锋. 新能源并网中主要储能技术类型[J]. 储能科学与技术, 2025, 14(2): 879-882.
[5]	田野, 王珊珊, 姚旭, 刘佳鑫, 韩晓冬. 配电网通信与分布式储能技术协同应用的意义与发展[J]. 储能科学与技术, 2025, 14(1): 190-192.
[6]	闻娜, 刘成伟, 张晓阳, 高健, 麻丽明. 智能电网中自动化控制与储能技术的协同应用研究[J]. 储能科学与技术, 2025, 14(1): 219-221.
[7]	袁誉杭, 高宇辰, 张俊东, 高岩斌, 王超珑, 陈翔, 张强. 大语言模型在储能研究中的应用[J]. 储能科学与技术, 2024, 13(9): 2907-2919.
[8]	黄家辉, 邝祝芳. 人工智能与储能技术融合的前沿发展[J]. 储能科学与技术, 2024, 13(9): 3161-3181.
[9]	焦君宇, 张全權, 陈宁波, 王冀钰, 芦秋迪, 丁浩浩, 彭鹏, 宋孝河, 张帆, 郑家新. 电池大数据智能分析平台的研发与应用[J]. 储能科学与技术, 2024, 13(9): 3198-3213.
[10]	许晶, 王宇琦, 符晓, 杨其凡, 连景臣, 王力奇, 肖睿娟. 基于大数据的电池新材料设计[J]. 储能科学与技术, 2024, 13(9): 2920-2932.
[11]	钟逸, 冷彦, 陈思慧, 李培义, 邹智, 刘洋, 万佳雨. 基于大语言模型RAG架构的电池加速研究：现状与展望[J]. 储能科学与技术, 2024, 13(9): 3214-3225.
[12]	刘子玉, 姜泽坤, 邱伟, 徐泉, 牛迎春, 徐春明, 周天航. 人工智能在长时液流电池储能中的应用：性能优化和大模型[J]. 储能科学与技术, 2024, 13(9): 2871-2883.
[13]	朱文韬, 周杨, 徐艺敏, 施涛. 电池储能技术在新能源发电系统中的应用与优化[J]. 储能科学与技术, 2024, 13(8): 2737-2739.
[14]	张楚, 陈栋才, 陈湘萍, 蔡永翔. 多应用场景下储能最优配置经济性效益分析[J]. 储能科学与技术, 2024, 13(6): 2078-2088.
[15]	孙健, 陶建龙, 胡芸蓉, 蔡潇龙, 杨勇平. 基于热泵型储电技术国内外研究综述[J]. 储能科学与技术, 2024, 13(6): 1963-1976.

DeepSeek在储能研究中的应用前景展望

A perspective on DeepSeek application in energy storage research

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 53

相关文章 15

编辑推荐

Metrics

本文评价