TEXT ANALYSIS OF CHINA'S HYDROGEN ENERGY POLICY BASED ON LDA THEMATIC MODEL

doi:10.13776/j.cnki.resourcesindustries.20250901.002

Abstract

Abstract: This paper aims to study the text characteristics and evolution of China's existing hydrogen energy policy, uses 353 pieces of hydrogen energy policies from 2019 to 2023 to establish a four-dimensional framework of “temporal evolution-high frequency words analysis-thematic exploration-thematic evolution”, which is applied to study on 8 comprehensive economic zones. For temporal evolution, annual policy amounts, regional distribution and categories reveal a policy evolution and spatial differences. For high frequency words, Python's Jieba database is used to process policy texts through mapping words cloud and co-occurrence network, identifying policy focuses and regional attention. For thematic exploration, LDA mode combined with TF-IDF algorithm is used to dig the potential themes by mapping thematic words co-occurrence network and analyzing their connection between potential themes and themes. For thematic evolution, mixed proportion of each theme in the policy texts marks the thematic intensity and reveals their thematic evolutionary trend. Policy amounts show an “M-shaped” rising, peaking at 2020 and 2022, due to an issue of key policies. Eastern and northern coasts have the most policy issues, 78 and 76 respectively; planning and method categories are the most, 85 and 68 respectively; prevention category is the least with only one. Policies focus on many aspects including whole industrial chain development, safety supervision. From 2022 to 2023 policies pay more attention to hydrogen energy safety; hydrogen stations are key domain in infrastructural construction; local policies have a single application scenario, and industrial homogeneity is prominent, with fuel cells as the key demonstration application scenario; the frequency of green hydrogen-related words is higher than other high-frequency words. Hydrogen energy policy has five potential themes in enterprise support, safety management, industrial innovation, whole industrial chain development and demonstration; enterprise support is closely related to demonstration, and industrial innovation is closely related to whole industrial chain development; the internal characteristic words of safety management theme are highly aggregated. Policy thematic evolution trend is stable; enterprise support is the leading theme; safety management trend is stable after initial largely rising, lower than enterprise support. Regional policy thematic evolution trend varies a lot with economy, resources and hydrogen energy industrial development. This paper presents a “1+N” policy system regarding China's hydrogen energy industry on selecting localized development direction, increasing green hydrogen policy attention, executing safety supervision on the whole hydrogen industrial chain and enlarging multiple applied scenarios of hydrogen energy.

Key words: hydrogen energy policy, high frequency words, thematic evolution, LDA

摘要： 为了系统研究中国现有氢能产业政策的文本特征与演进规律，以2019—2023年353份中国氢能政策为研究对象，构建“时间演化-高频词分析-主题挖掘-主题演变”政策四维分析框架，结合八大综合经济区展开分析。其中，在时间演化维度，通过统计年度政策数量、区域分布及类型特征，揭示政策发展轨迹与空间差异；在高频词分析维度，基于Python的Jieba库对样本政策措施文本分词处理，绘制词云图与共现网络，识别政策关注焦点及区域注意力差异；在主题挖掘维度，采用LDA模型结合TF-IDF算法挖掘样本政策潜在主题，绘制主题词共现网络，解析政策文本的潜在主题特征与主题间联结关系；在主题演变维度，借助文档集中各主题的混合比例来表征主题强度，探究政策主题演变趋势。研究表明：1）政策数量呈“M型”波动上升，有2020年、2022年两个波峰，关键政策的发布是直接原因；东部与北部沿海是政策发布的高地（分别为78、76份）；规划类与办法类占据主导（分别为85、68份），预案类最少（1份）。2）政策对全产业链发展、安全监管等多个方面高度关注；2022—2023年对氢能安全管理的关注有提升；加氢站建设是基础设施中的重点领域；地方政策应用场景单一，产业同质化突出，均将燃料电池作为重点示范应用场景；绿氢相关词汇与其他高频词频次有明显差距。3）氢能政策有企业扶持、安全管理、产业创新、全产业链发展、示范应用5个潜在主题；企业扶持与示范应用，产业创新与全产业链发展联系较为紧密；安全管理主题内部特征词高度聚合。4）政策主题演变趋势整体平稳，企业扶持作为核心议题始终占据主导地位，安全管理在经历初期的大幅提升后趋于稳定，略低于企业扶持；区域间政策主题演变趋势差异显著，各地因经济、资源及氢能产业发展阶段不同，采取了差异化策略。最后就中国氢能产业发展提出加快构建“1+N”政策体系、因地制宜差异化选择发展方向、提升绿氢政策聚焦度、落实氢能全产业链安全防护监测预警及拓展氢能多元化应用场景的建议。

关键词: 氢能政策, 高频词分析, 主题演变, LDA

CLC Number:

F206

QIN Yu, HUANG Xiaoli, LIN Lü. TEXT ANALYSIS OF CHINA'S HYDROGEN ENERGY POLICY BASED ON LDA THEMATIC MODEL[J]. Resources & Industries, 2025, 27(5): 123-138.

秦宇, 黄晓莉, 林绿. 基于LDA主题模型的中国氢能政策文本分析[J]. 资源与产业, 2025, 27(5): 123-138.

References

毕珍珍, 2019. 日本的“氢能源基本战略”与全球气候治理[J]. 国际论坛, 21(2): 140-154, 160.〔BI Z Z, 2019. Japan's “basic hydrogen strategy” and global climate governance[J]. International Forum, 21(2): 140-154, 160.〕

柴文越, 刘小平, 梁爽, 2023. 新兴主题识别方法研究综述[J]. 现代情报, 43(12): 164-177.

〔CHAI W Y, LIU X P, LIANG S, 2023. Review of emerging topic identification methods[J]. Journal of Modern Information, 43(12): 164-177.〕

何青, 孟照鑫, 沈轶, 等, 2021. “双碳”目标下我国氢能政策分析与思考[J]. 热力发电, 50(11): 27-36.〔HE Q, MENG Z X, SHEN Y, et al., 2021. Analysis and thinking of hydrogen energy policies in China under “double carbon” target[J]. Thermal Power Generation, 50(11): 27-36.〕

黄冬玲, 刘源, 袁小帅, 等, 2024. 基于自然语言处理技术的中国氢能政策数据挖掘研究[J]. 中国科学院院刊, 39(6): 1032-1046.〔HUANG D L, LIU Y, YUAN X S, et al., 2024. Study on data mining of hydrogen energy policy in China based on natural language processing technology[J]. Bulletin of Chinese Academy of Sciences, 39(6): 1032-1046.〕

黄晓林, 胡锡晟, 黄卉, 等, 2021. 中国氢能源产业政策量化分析及区域布局研究[J]. 科技情报研究, 3(2): 83-95.〔HUANG X L, HU X S, HUANG H, et al., 2021. Research on China's hydrogen energy industry policy by quantitative analysis and its regional policy distribution[J]. Scientific Information Research, 3(2): 83-95.〕

李佳蕾, 安培浚, 肖仙桃, 2023. 学科交叉主题识别方法研究综述[J]. 数据分析与知识发现, 7(4): 1-15.〔LI J L, AN P J, XIAO X T, 2023. Review of methods for interdisciplinary topic identification[J]. Data Analysis and Knowledge Discovery, 7(4): 1-15.〕

李建林, 邵晨曦, 张则栋, 等, 2023. 氢能产业政策及商业化模式分析[J]. 发电技术, 44(3): 287-295.〔LI J L, SHAO C X, ZHANG Z D, et al., 2023. Analysis of hydrogen industry policy and commercialization model[J]. Power Generation Technology, 44(3): 287-295.〕

李湘东, 张娇, 袁满, 2014. 基于LDA模型的科技期刊主题演化研究[J]. 情报杂志, 33(7): 115-121.〔LI X D, ZHANG J, YUAN M, 2014. On topic evolution of a scientific journal based on LDA model[J]. Journal of Intelligence, 33(7): 115-121.〕

连迎洁, 孙玉玲, 罗家棠, 2025. 基于深度学习和灰色关联分析的我国氢能政策协同测度研究[J]. 世界科技研究与发展, 47(1): 111-128. 〔LIAN Y J, SUN Y L, LUO J T, 2025. Research on coordinated measurement of China's hydrogen energy policy based on deep learning and grey relational analysis[J]. World Sci-Tech R & D, 47(1): 111-128. 〕

柳扬, 李玉刚, 丁建南, 等, 2023. 我国东部沿海地区典型省份氢能产业现状及发展建议[J]. 化工设计通讯, 49(9): 161-163.〔LIU Y, LI Y G, DING J N, et al., 2023. Current situation and development suggestions of hydrogen energy industry in typical provinces of eastern coastal areas in China[J]. Chemical Engineering Design Communications, 49(9): 161-163.〕

申伟宁, 夏梓莹, 苏爽, 2022. 大气污染治理政策文本量化研究: 以京津冀城市群为例[J]. 资源与产业, 24(1): 65-72.〔SHEN W N, XIA Z Y, SU S, 2022. Quantitative research on air pollution control policy texts based on Beijing-Tianjin-Hebei (JJJ) city cluster[J]. Resources & Industries, 24(1): 65-72.〕

孙玉玲, 胡智慧, 秦阿宁, 等, 2020. 全球氢能产业发展战略与技术布局分析[J]. 世界科技研究与发展, 42(4): 455-465.〔SUN Y L, HU Z H, QIN A N, et al., 2020. Analysis of strategy and technology situation on global hydrogen industry[J]. World Sci-Tech R & D, 42(4): 455-465.〕

汤匀, 王宏杨, 岳芳, 等, 2025. 北京、上海、广州和武汉四城氢能产业政策演化比较研究[J]. 世界科技研究与发展, 47(1): 129-142.〔TANG Y, WANG H Y, YUE F, et al., 2025. A comparative study on the evolution of hydrogen energy industrial policies in four cities: Beijing, Shanghai, Guangzhou and Wuhan[J]. World Sci-Tech R & D, 47(1): 129-142.〕

仝德, 徐珩, 邱君丽, 等, 2024. 地方响应国家城市更新行动的区域差异: 基于LDA模型的政策文本量化分析[J]. 城市发展研究, 31(9): 1-6.〔TONG D, XU H, QIU J L, et al., 2024. Regional differences in local responses to national urban regeneration actions: a quantitative analysis of policy texts based on the LDA model[J]. Urban Development Studies, 31(9): 1-6.〕

王超, 孙福全, 许晔, 等, 2022. 世界主要经济体氢能发展战略剖析与启示[J]. 世界科技研究与发展, 44(5): 597-604.〔WANG C, SUN F Q, XU Y, et al., 2022. Analysis and enlightenment of hydrogen development strategy of world's major economies[J]. World Sci-Tech R & D, 44(5): 597-604.〕

王佳杰, 纪国涛, 2021. 基于PMC指数模型的氢能产业专项政策量化评价[J]. 北京经济管理职业学院学报, 36(4): 34-43.〔WANG J J, JI G T, 2021. Quantitative evaluation of special policies for hydrogen energy industry based on PMC index model[J]. Journal of Beijing Institute of Economics and Management, 36(4): 34-43.〕

王意东, 许苏予, 何太碧, 等, 2024. 中国氢能及燃料电池产业政策研究及启示[J]. 天然气工业, 44(5): 136-145.〔WANG Y D, XU S Y, HE T B, et al., 2024. Research and implication of China's hydrogen energy and fuel cell industrial policies[J]. Natural Gas Industry, 44(5): 136-145.〕

王彦哲, 欧训民, 周胜, 2022. 基于学习曲线的中国未来制氢成本趋势研究[J]. 气候变化研究进展, 18(3): 283-293.〔WANG Y Z, OU X M, ZHOU S, 2022. Future cost trend of hydrogen production in China based on learning curve[J]. Climate Change Research, 18(3): 283-293.〕

吴朝玲, 张昱晨, 杜瑞, 等, 2024. 氢储运技术的发展现状、挑战及未来方向[J]. 前瞻科技, 3(4): 22-33.〔WU C L, ZHANG Y C, DU R, et al., 2024. Current state, challenges, and development directions of hydrogen storage and transportation technologies[J]. Science and Technology Foresight, 3(4): 22-33.〕

吴小松, 张睿, 2024. 苏粤浙鲁新型工业化比较研究及对江苏启示[J]. 新型工业化理论与实践(1): 108-117.〔WU X S, ZHANG R, 2024. Comparative study on new industrialization in Jiangsu, Guangdong, Zhejiang and Shandong and its enlightenment to Jiangsu[J]. Theory and Practice of New Industrialization(1): 108-117.〕

武正弯, 2021. 德澳加日四国氢能战略比较研究[J]. 国际石油经济, 29(4): 60-66.〔WU Z W, 2021. The comparative study on hydrogen energy strategies of Germany, Australia, Canada, and Japan[J]. International Petroleum Economics, 29(4): 60-66.〕

杨琳, 2025. 中国氢能产业政策层级特征及演化趋势研究[J]. 智库理论与实践(1): 94-104.〔YANG L, 2025. Study on the characteristics and evolutionary trends of China's hydrogen energy industry policy hierarchy[J]. Think Tank: Theory& Practice(1): 94-104.〕

袁家海, 牟琪林, 许传博, 等, 2023. 基于系统动力学的中国绿氢产业发展政策仿真[J]. 中国人口·资源与环境, 33(6): 49-58.〔YUAN J H, MOU Q L, XU C B, et al., 2023. Simulation study on the development policy of China's green hydrogen industry based on system dynamics[J]. China Population, Resources and Environment, 33(6): 49-58.〕

张庆生, 黄雪松, 2023. 国内外氢能产业政策与技术经济性分析[J]. 低碳化学与化工, 48(2): 133-139.〔ZHANG Q S, HUANG X S, 2023. Analysis of domestic and foreign hydrogen energy industrial policies and technical economy[J]. Low-Carbon Chemistry and Chemical Engineering, 48(2): 133-139.〕

张荣达, 赵晓丽, 张庆斌, 等, 2024. 基于风光发电的中国各省绿氢替代灰氢经济性研究: 基于精细地理网格分析[J]. 资源与产业, 26(3): 134-146. 〔ZHANG R D, ZHAO X L, ZHANG Q B, et al., 2024. Economic feasibility of green hydrogen replacing gray hydrogen in China's provinces based on wind and photovoltaic power generation: based on detailed geographic grid analysis[J]. Resources & Industries, 26(3): 134-146.〕

赵振东, 刘国庆, 辛江慧, 等, 2023. 基于补贴政策的氢燃料电池汽车产业动力学演化研究: 以江苏省为例[J]. 资源与产业, 25(1): 27-39.〔ZHAO Z D, LIU G Q, XIN J H, et al, 2023. Dynamic evolution of hydrogen-fuel cell vehicle (HFCV) industry based on allowance policies in Jiangsu province[J]. Resources & Industries, 25(1): 27-39.〕

郑新曼, 董瑜, 2021. 政策文本量化研究的综述与展望[J]. 现代情报, 41(2): 168-177.〔ZHENG X M, DONG Y, 2021. Review on quantitative analysis of political texts[J]. Journal of Modern Information, 41(2): 168-177.〕

周海炜, 许莹, 2023. 基于内容分析法的我国能源安全政策文本量化分析[J]. 资源与产业, 25(3): 53-68.〔ZHOU H W, XU Y, 2023. Quantitative analysis of China's energy security policy texts based on content analysis[J]. Resources & Industries, 25(3): 53-68.〕

朱彤, 李雅卓, 2022. 德国能源转型进程中氢能发展经验及启示[J]. 世界石油工业, 29(5): 11-16.〔ZHU T, LI Y Z, 2022. Development and enlightenment of hydrogen energy in the German energy transition process[J]. World Petroleum Industry, 29(5): 11-16.〕

BALAJI R K, YOU F Q, 2024. Sailing towards sustainability: offshore wind's green hydrogen potential for decarbonization in coastal USA[J]. Energy & Environmental Science, 17(17): 6138-6156.

BLEI D M, NG A Y, JORDAN M I, 2003. Latent Dirichlet allocation[J]. Journal of Machine Learning Research, 3: 993-1022.

CHO A, KIM H, PARK S, 2024. Resurgence of the hydrogen energy in South Korea's government strategies from 2005 to 2019[J]. International Journal of Hydrogen Energy, 65: 844-854.

JIANG Z Z, GAO X W, 2024. Quantification and evaluation of China's hydrogen automobile industry policy[J]. International Journal of Hydrogen Energy, 56: 757-768.

ZHANG Q, CHEN W M, LING W, 2022. Policy optimization of hydrogen energy industry considering government policy preference in China[J]. Sustainable Production and Consumption, 33: 890-902.

ZHU Y P, HU Y, ZHU Y, 2024. Can China's energy policies achieve the “dual carbon” goal?: a multi-dimensional analysis based on policy text tools[J]. Environment, Development and Sustainability: 1-40.

[1]	ZHANG Guofeng, LIU Jiashi, WANG Ruixian, MA Xiaojing. IMPACTS OF TECHNICAL INNOVATION ON ECOLOGICAL INDUSTRIES BASED ON EMPIRICAL EVIDENCES FROM HEBEI'S 40 MOUNTAINOUS [J]. Resources & Industries, 2025, 27(5): 36-50.
[2]	ZHENG Guoliang, LIAO Hua. IPCC AR6'S VIEWS ON POTENTIAL IMPACTS AND HARNESS OF SOLAR RADIATION MODIFICATION ON NATURE AND SOCIETY [J]. Resources & Industries, 2024, 26(4): 1-9.
[3]	ZHU Zhiming, XU Jie, LI Hongyan, et al.. COUPLING COORDINATION BETWEEN ENERGY CONSUMPTION CARBON EMISSION EFFICIENCY AND ECOLOGICAL RESILIENCE UNDER “DUAL CARBON” [J]. Resources & Industries, 2024, 26(3): 21-35.
[4]	YAO Baoshuai, BAO Ning, LIU Xianting, et al. CORRELATION MEASUREMENT AND EFFECTS OF CHINA’S NEW ENERGY INDUSTRIES [J]. Resources & Industries, 2024, 26(2): 36-52.
[5]	MENG Xiaohong, GUO Pibin, WU Qinglong. RESEARCH STATUS AND TREND ANALYSIS OF CLEAN COAL TECHNOLOGY BASED ON PATENT INFORMATION [J]. Resources & Industries, 2024, 26(2): 164-172.
[6]	HU Xiaolan, PENG Chuansheng, SHI Jianxiang, et al. FINE MANAGEMENT PRACTICE OF COALBED METHANE GROUND EXTRACTION ON XIANDAO-1 WELL IN SHIYUAN COAL MINE [J]. Resources & Industries, 2024, 26(2): 173-180.
[7]	JU Yanping, WANG Xinhua. RE-UNDERSTANDING EXCESSIVE COAL PRODUCING CAPACITY BASED ON MEASUREMENT AND SPATIAL EVOLUTION OF LATENT CLASS RANDOM MARGINALIZATION (LCRM) [J]. Resources & Industries, 2023, 25(6): 41-52.
[8]	YANG Kaijun, WANG Wenxuan. IMPACTS OF LOCAL GOVERNMENTAL BEHAVIORS ON INDUSTRIAL ENERGY ENVIRONMENTAL EFFICIENCY ALONG“THE BELT AND ROAD” PROVINCES/CITIES [J]. Resources & Industries, 2023, 25(6): 91-104.
[9]	XIAO Yutong, CHEN Jun. EVOLUTIONARY GAMING ANALYSIS ON GREEN ENERGY INTERNATIONAL COOPERATION UNDER GLOBAL CARBON NEUTRALIZATION [J]. Resources & Industries, 2023, 25(3): 19-30.
[10]	XU Wenhao, XIE Yulei, JI Ling, et al. VARIANCE STUDY ON GUANGDONG'S ENERGY CONSUMPTION SECTOR BASED ON INPUT/OUTPUT & ECOLOGICAL NETWORK ANALYSIS [J]. Resources & Industries, 2022, 24(6): 44-53.
[11]	SUN Xiaoqi, SHI Qing. ENERGY CONSERVATION AND EMISSION REDUCTION EFFECTS OF STEEL-CAPACITY-CUTTING POLICY BASED ON INPUT/OUTPUT MODEL [J]. Resources & Industries, 2022, 24(4): 1-8.
[12]	GAO Yuan, LIU Gengyuan, CHEN Caocao, et al. BENCHMARKING-ORIENTED URBAN GREENHOUSE GAS ACCOUNTING METHODOLOGY FRAMEWORK IN CHINA [J]. Resources & Industries, 2022, 24(3): 1-20.
[13]	ZHU Xueyi. THEORETICAL FRAMEWORK CONSTRUCTION OF HIGH QUALITY DEVELOPMENT OF COAL RESOURCE CAPITALIZATION BASED ON AGGLOMERATIVE SUB-GROUPING [J]. Resources & Industries, 2021, 23(6): 31-37.
[14]	LIU Dan, HUANG Shanqing, YANG Yang, et al. ECOLOGICAL RISK TRANSMISSION MECHANISM AND SCENARIO PREDICTION OF HYDROPOWER DEVELOPMENT BASED ON BAYESIAN NETWORK AND SHARED SOCIOECONOMIC PATHWAY [J]. Resources & Industries, 2021, 23(6): 38-51.
[15]	LIU Nana, RU Ting. FEASIBILITY STUDY ON SURFACE EXTRACTION OF DUIJIANGNAN COAL FIELD CBM IN GUIZHOU PROVINCE [J]. Resources & Industries, 2021, 23(1): 63-68.