Predicting China’s carbon emission driven by population factors
-
摘要: 基于组合神经网络(LSTM-IPSO-BP)模型,研究中国碳排放强度的影响因素以及在未来人口数量和结构变动条件下中国碳排放强度的变化趋势.结果表明:1)中国未来人口数量的持续下降将会导致碳排放强度的增加,人口数量下降的速率与碳排放强度正相关;2)城镇化水平提高虽然会降低碳排放强度,但中国未来城镇化进程放缓会增加碳减排的压力;3)在现有各影响因素发展趋势下,特别是人口数量和结构变动的条件下,中国将难以在2030年前实现碳达峰,这也表明未来10年中国政府需要加大碳排放政策的调控力度.Abstract: Combined neural network model LSTM-IPSO-BP was applied to study factors influencing the intensity and change trend in China’s carbon emission driven by future population and structure changes. It is found that continuous decline in China’s population in the future will lead to increased carbon emission intensity, the rate of population decline is positively correlated with carbon emission intensity. Although improved urbanization level will reduce carbon emission intensity, the slowdown of China’s future urbanization process will increase pressure on carbon emission reduction. Under current policies and development trend of various influencing factors, especially population and structural changes, it is difficult for China to achieve a carbon peak before 2030, therefore the Chinese government needs to strengthen the control of carbon emission policies in the following decade.
-
Key words:
- carbon emission /
- path prediction /
- negative population growth /
- urbanization /
- neural network model
-
表 1 我国碳排放量影响因素选择
编号 变量名 描述 A1 经济水平 国内生产总值GDP(亿元) A2 产业结构 第二产业占GDP比例(%) A3 人口规模 人口总量(万人) A4 城镇化水平 城镇人口占比(%) A5 能源消费总量 年末能源消费总量(万t标准煤) A6 能源结构 煤炭消费占能源消费总量比例(%) A7 电力消费 电力消费总量(亿kw·h) A8 交通发展水平 私人汽车拥有量(万辆) A9 森林覆盖率 森林面积占土地总面积的比例(%) -
[1] 张军莉,刘丽萍. 国内区域碳排放预测模型应用综述[J]. 环境科学导刊,2019,38(4):15 [2] GRUBB M,SHA F,SPENCER T,et al. A review of Chinese CO2 emission projections to 2030:the role of economic structure and policy[J]. Climate Policy,2015,15(sup1):S7 doi: 10.1080/14693062.2015.1101307 [3] GREEN F,STERN N. China’s changing economy:implications for its carbon dioxide emissions[J]. Climate Policy,2017,17(4):423 doi: 10.1080/14693062.2016.1156515 [4] 张炎治,冯颖,张磊. 中国碳排放增长的多层递进动因:基于SDA和SPD的实证研究[J]. 资源科学,2021,43(6):1153 [5] 吉兴全,赵国航,于一潇,等. 基于4E平衡的碳排放因素分解与峰值预测方法[J]. 高电压技术,2022,48(7):2483 [6] ZHANG S,YUAN R,LI N. Persistent mitigation efforts and implications for China’s emissions peak using statistical projections[J]. Science of the Total Environment,2022,826:154127 doi: 10.1016/j.scitotenv.2022.154127 [7] 渠慎宁,郭朝先. 基于 STIRPAT 模型的中国碳排放峰值预测研究[J]. 中国人口·资源与环境,2010,20(12):10 [8] 郭筱祎. 双碳目标背景下陕西省碳排放预测及减排路径研究[J]. 煤炭经济研究,2021,41(9):15 [9] LIU P,LIN B R,ZHOU H,et al. CO2 emissions from urban buildings at the city scale:system dynamic projections and potential mitigation policies[J]. Applied Energy,2020,277:115546 doi: 10.1016/j.apenergy.2020.115546 [10] 王苗,王鹏雅,陈鑫. 基于灰色神经网络模型的美国碳排放预测[J]. 科技资讯,2019,17(14):202 [11] 张迪,王彤彤,支金虎. 基于IPSO-BP神经网络模型的山东省碳排放预测及生态经济分析[J]. 生态科学,2022,41(1):149 [12] HOCHREITER S,SCHMIDHUBER J. Long short-term memory[J]. Neural Computation,1997,9(8):1735 doi: 10.1162/neco.1997.9.8.1735 [13] LI H D,ZHOU T M,JIA C. The influence of the universal two-child policy on China’s future population and ageing[J]. Journal of Population Research,2019,36(3):183 doi: 10.1007/s12546-019-09228-7 -