广东电力低碳高质量发展下的电煤需求影响分析  

Influence Analysis of Thermal Coal Demand Under the Low-carbon and High-quality Development of Electric Power in Guangdong Province

作　　者：贠晓可 张天任 黄玉萍 龚贤夫 彭勃 YUN Xiaoke;ZHANG Tianren;HUANG Yuping;GONG Xianfu;PENG Bo(School of Economics and Management,Shenyang University of Chemical Technology,Shenyang,Liaoning 110142,China;Guangzhou Institute of Energy Conversion,Chinese Academy of Sciences,Guangzhou,Guangdong 510640,China;School of Energy Science and Engineering,University of Science and Technology of China,Hefei,Anhui 230026,China;Grid Planning&Research Center,Guangdong Power Grid Co.,Ltd.,Guangzhou,Guangdong 510080,China)

机构地区：[1]沈阳化工大学经济与管理学院,辽宁沈阳110142 [2]中国科学院广州能源研究所,广东广州510640 [3]中国科学技术大学能源科学与技术学院,安徽合肥230026 [4]广东电网有限责任公司电网规划研究中心,广东广州510080

出　　处：《广东电力》2023年第11期1-10,共10页Guangdong Electric Power

基　　金：国家重点研发计划项目(2022YFB3304503);中国南方电网有限责任公司科技项目(037700KK52220041,GDKJXM20220904)。

摘　　要：从广东电力清洁、低碳、高质量发展的角度,采用灰色关联度分析法和对数平均迪氏指数法,阐述了人口规模、经济发展、产业结构、发电强度、新能源发电强度、用电强度、碳排放强度以及电煤消耗强度8个因素对省域电煤消费量的动态影响。分析2019—2022年各季度和2012—2022年各年度的广东省电煤消费量的影响因素,并利用神经网络极限学习机(extreme learning machines, ELM)、长短期记忆(long short-term memory, LSTM)网络、卷积双向长短期记忆网络(convolutional neural networks-bi-directional LSTM,CNN-BILSTM)3种模型对该省未来的电煤消费量开展季度预测和分析。研究发现:在季度分析中,经济发展效应是促进电煤消费量增加的首要驱动因素,产业结构效应是抑制电煤消费量增加最重要的驱动因素,其促进和抑制作用十分明显;在年度分析中,经济发展效应是电煤消费量增加最重要的驱动因素,碳排放强度效应是降低电煤消费量的首要驱动因素;在对未来电煤消费量的预测上,CNN-BILSTM模型的平均误差率保持在10%以下,可以为广东省电力低碳高质量发展路径规划提供数据和技术支撑。From the perspective of clean,low-carbon and high-quality electric power development in Guangdong province,this paper adopts grey relational analysis method and logarithmic mean Divisia index method to describe the dynamic effects of eight factors,such as population size,economic development,industrial structure,power generation intensity of each industry,new energy power generation intensity,power consumption intensity,carbon emission intensity and power coal consumption intensity on the power consumption of Guangdong province.By analyzing the influencing factors of thermal coal consumption in Guangdong province in 2019-2022 quarter and 2012-2022 year,three models of ELM,LSTM and CNN-BILSTM are used to forecast and analyze the future thermal coal consumption in Guangdong province on a quarterly basis.The results show that:in the quarterly analysis,the economic development effect is the primary driving factor to promote the increase of thermal coal consumption,and the industrial structure effect is the most important driving factor to inhibit the increase of thermal coal consumption.In the annual analysis,the economic development effect is the most important driving factor for the increase of thermal coal consumption,and the carbon emission intensity effect is the primary driving factor for the reduction of thermal coal consumption.For the prediction of future thermal coal consumption,the average error rate of CNN-BILSTM model is kept within 10%,which can provide data and technical support for the low-carbon and high-quality development path planning of Guangdong power industry.

关 键 词：电煤消费 需求预测 碳减排 影响分析 对数平均迪氏指数 深度学习 

分 类 号：F426.61[经济管理—产业经济]