| 41 | 0 | 79 |
| 下载次数 | 被引频次 | 阅读次数 |
在双碳目标下对煤炭需求量进行中长期预测可以为能源结构优化和政策制定提供科学依据。采用长短期记忆网络（long short-term memory networks,LSTM）模型、灰色预测模型（GM(1,1)）与支持向量机（support vector machine,SVM）模型，选择河南省生产总值（gross domestic product,GDP）、清洁能源消费占比、能源消费弹性系数、人口数量与城镇化率5个参数作为特征序列，对2025—2035年河南省煤炭需求量进行预测，并选择长远能源规划（long-range energy alternatives planning system, LEAP）模型对政策强制情景、技术突破情景、经济波动情景进行煤炭需求量预测。结果表明：LSTM模型、GM(1,1)模型与SVM模型预测2025—2035年煤炭需求量分别为12 009.12万～14 498.02万吨标准煤、11 092.74万～14 793.42万吨标准煤和11 209.26万～14 200.20万吨标准煤。2035年的煤炭需求量与国际能源署（International Energy Agency, IEA）基于低碳加速背景下的煤炭需求量较为接近，能够满足双碳目标下对煤炭需求的预期。
Abstract：Conducting medium and long-term prediction of coal demand under the dual-carbon goals can provide a scientific basis for the optimization of energy structure and policy-making. Herein, five parameters—gross domestic product(GDP) of Henan province, proportion of clean energy consumption, energy consumption elasticity coefficient, population, and urbanization rate were selected as feature sequences to predict Henan province's coal demand from 2025 to 2035, and long short-term memory networks(LSTM) model, grey prediction model(GM(1,1)), and support vector machine(SVM) model were employed for the prediction. The long-range energy alternatives planning system(LEAP) model was selected to predict coal demand under policy coercion, technological breakthrough, and economic fluctuation scenarios. The predicted coal demand ranging from 2025 to 2035 by the LSTM model, the GM(1,1) model and the SVM model were 120. 091 2-144. 980 2 Mtce, 110. 927 4-147. 934 2 Mtce and 112. 092 6-142. 002 0 Mtce, respectively. The coal demand in 2035 was relatively close to that of the International Energy Agency(IEA) under the background of accelerating low-carbon development, which was expected to meet the expected coal demand under the dual carbon goals.
[1]IPCC. Climate Change 2021:the physical science basis[EB/OL].[2025-04-07]. https：∥www. ipcc. ch/report/ar6/wg1/.
[2]黄建平，陈文，温之平，等.新中国成立70年以来的中国大气科学研究：气候与气候变化篇[J].中国科学：地球科学，2019, 49(10):1607-1640.HUANG J P, CHEN W, WEN Z P, et al. Review of Chinese atmospheric science research over the past 70years:climate and climate change[J]. Science China Earth Sciences, 2019, 49(10):1607-1640.(in Chinese)
[3]中华人民共和国国民经济和社会发展第十四个五年规划和2035年远景目标纲要[EB/OL].[2025-04-07].https：∥www. gov. cn/xinwen/2021-03/13/content_5592681. htm.Outline of the 14th Five-Year Plan(2021-2025)for National Economic and Social Development and Vision2035 of the People’s Republic of China[EB/OL].[2025-04-07]. https：∥www. gov. cn/xinwen/2021-03/13/content_5592681. htm.(in Chinese)
[4]NGUYEN T C V, LE Q H. Impact of globalization on coal consumption in Vietnam:an empirical analysis[J]. The Journal of Asian Finance, Economics and Business, 2020, 7(6):185-195.
[5]谢和平，吴立新，郑德志. 2025年中国能源消费及煤炭需求预测[J].煤炭学报，2019, 44(7):1949-1960.XIE H P, WU L X, ZHENG D Z. Prediction on the energy consumption and coal demand of China in 2025[J]. Journal of China Coal Society, 2019, 44(7):1949-1960.(in Chinese)
[6]宋昆鹏，宋亚开.基于ARIMA-LSTM模型的煤炭能源消费预测研究：以河南省为例[J].煤炭经济研究，2024, 44(9):48-54.SONG K P, SONG Y K. Coal energy consumption forecasting research based on ARIMA-LSTM model:a case study of Henan Province[J]. Coal Economic Research, 2024, 44(9):48-54.(in Chinese)
[7]匡立春，邹才能，黄维和，等.碳达峰碳中和愿景下中国能源需求预测与转型发展趋势[J].石油科技论坛，2022, 41(1):9-17.KUANG L C, ZOU C N, HUANG W H, et al.China’s energy demand projection and energy transition trends under carbon peak and carbon neutrality situation[J]. Petroleum Science and Technology Forum, 2022,41(1):9-17.(in Chinese)
[8]宋豪，张艳，高天明，等.基于神经网络的环境约束下我国煤炭需求预测[J].中国矿业，2021, 30(5):72-78.SONG H, ZHANG Y, GAO T M, et al. Prediction of coal demand in China under environmental constraints based on the neural network[J]. China Mining Magazine, 2021, 30(5):72-78.(in Chinese)
[9]王浩，成玮.基于GM(1,1)模型煤炭消费量预测研究[J].能源与节能，2023(5):51-53, 57.WANG H, CHENG W. Prediction of coal consumption based on GM(1,1)model[J]. Energy and Energy Conservation, 2023(5):51-53, 57.(in Chinese)
[10]赵凌生.双碳背景下京津冀地区煤炭需求预测研究[D].北京：华北电力大学，2024:19-26, 43-50.ZHAO L S. Research on coal demand forecast in Beijing-Tianjin-Hebei Region under the background of dual-carbon goals[D]. Beijing:North China Electric Power University, 2024:19-26, 43-50.(in Chinese)
[11]刘健，程根银.基于多元线性回归对我国煤炭消费量的分析[J].华北科技学院学报，2018, 15(3):108-112.LIU J, CHENG G Y. Analysis of coal consumption in China based on multiple linear regression[J]. Journal of North China Institute of Science and Technology,2018, 15(3):108-112.(in Chinese)
[12]董洁芳，王强，王兴民，等.山西省煤炭消费需求影响因素的实证研究[J].中国科学院大学学报，2016,33(5):641-650.DONG J F, WANG Q, WANG X M, et al. Empirical study on the influencing factors of coal consumption in Shanxi province[J]. Journal of University of Chinese Academy of Sciences, 2016, 33(5):641-650.(in Chinese)
[13]徐丽娜，张晓华，邓灿辉，等.中国西部地区能源消费量影响因素分析：基于协整模型的实证研究[J].中国资源综合利用，2024, 42(12):68-70.XU L N, ZHANG X H, DENG C H, et al. Analysis of factors influencing energy consumption in western China:empirical research based on cointegration model[J]. China Resources Comprehensive Utilization,2024, 42(12):68-70.(in Chinese)
[14]杨英明，李全生，方杰.基于最优加权组合模型的煤炭消费预测分析[J].煤炭工程，2018, 50(5):156-160.YANG Y M, LI Q S, FANG J. Coal consumption forecasting based on optimum weighted composition model[J]. Coal Engineering, 2018, 50(5):156-160.(in Chinese)
[15]程文荣，姚天祥.基于新陈代谢GM(1,1)模型的中国煤炭消费量预测分析[J].数学的实践与认识，2015,45(16):93-98.CHENG W R, YAO T X. Application of an information renewal GM(1,1)model predicting in coal consumption forecast of China[J]. Mathematics in Practice and Theory, 2015, 45(16):93-98.(in Chinese)
[16]李润宸.基于LSTM的煤炭消费预测模型[J].煤炭经济研究，2021, 41(4):20-24.LI R C. Coal consumption forecast model based on LSTM[J]. Coal Economic Research, 2021, 41(4):20-24.(in Chinese)
[17]李姗姗，马雨晴，孔维龄，等.基于LEAP模型的安徽省能源需求与碳排放预测[J].环境科学，2025, 46(8):4911-4922.LI S S, MA Y Q, KONG W L, et al. Energy demand and carbon emission prediction in Anhui province based on LEAP model[J]. Environmental Science, 2025, 46(8):4911-4922.(in Chinese)
[18]GRAVES A. Long short-term memory[M]∥Supervised Sequence Labelling with Recurrent Neural Networks. Berlin:Springer, 2012:1735-1780.
[19]刘思峰，邓聚龙. GM(1,1)模型的适用范围[J].系统工程理论与实践，2000, 20(5):121-124.LIU S F, DENG J L. The range suitable for GM(1,1)[J]. Systems Engineering-Theory&Practice, 2000,20(5):121-124.(in Chinese)
[20]丁世飞，齐丙娟，谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报，2011, 40(1):2-10.DING S F, QI B J, TAN H Y. An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China, 2011, 40(1):2-10.(in Chinese)
[21]河南省统计局，国家统计局河南调查总队.河南统计年鉴2024[M].北京：中国统计出版社，2024.Henan Province Bureau of Statistics, Survey Office of the National Bureau of Statistics in Henan. Henan statistical yearbook 2024[M]. Beijing:China Statistics Press, 2024.(in Chinese)
[22]国际能源署.世界能源展望2023[EB/OL].[2025-04-25]. https：∥www. iea. org/reports/world-energyoutlook-2023?wpappninja_v=0yrwa1dze.International Energy Agency. World Energy Outlook2023[EB/OL].[2025-04-25]. https：∥www. iea.org/reports/world-energy-outlook-2023? wpappninja_v=0yrwa1dze.(in Chinese)
[23]国家统计局.中国统计年鉴2024[M].北京：中国统计出版社，2024.National Bureau of Statistics of China. China statistical yearbook 2024[M]. Beijing:China Statistics Press,2024.(in Chinese)
基本信息:
中图分类号:F426.21
引用信息:
[1]杜春彦,郑可欣,尉雪菲.双碳目标下2025—2035年河南省煤炭需求量预测[J].中国科技论文,2026,21(01):34-44.
基金信息:
河南省部分战略性矿产共伴生、低品位资源再评价及战略性矿产出让区块调查评价项目资助（豫自然资函[2024]402号）
2026-01-15
2026-01-15