首页|基于ARIMA模型的甘肃省碳排放预测分析

基于ARIMA模型的甘肃省碳排放预测分析

扫码查看
在"双碳"目标视域下,为了甘肃省如期实现"双碳"目标,给碳减排方案和措施的制定提供理论基础,基于2009-2022年年甘肃省碳排放数据,运用ARIMA模型对2023-2033年的甘肃省碳排放量、碳排放强度和人均碳排放量进行预测,通过数据学习训练,得出碳排放量预测的最佳模型为ARIMA(2,1,0),平均绝对百分比误差MAPE为 0。016 2,碳排放强度预测的最佳模型为(0,1,0),平均绝对百分比误差MAPE为0。032 8,人均碳排放量的最佳预测模型为(1,1,0),平均绝对百分比误差MAPE为0。018 6,预测值均处于95%置信区间,预测的精度较高。研究结果表明:甘肃省未来的碳排放量和人均碳排放量均依然呈逐年上升趋势,碳排放强度逐年下降。根据预测的数据和趋势可知,甘肃省要想如期实现"双碳"目标,必须制定强有力的碳减排方案和措施,并以相关法律法规条文强制执行和落实。
Carbon Emission Prediction and Analysis in Gansu Province Based on ARIMA Model
In the perspective of the"dual carbon"target,in order to achieve the"dual carbon"target as scheduled in Gansu Province and provide a theoretical basis for the formulation of carbon reduction plans and measures,this article is based on the car-bon emissions data of Gansu Province from 2009 to 2021,and uses the ARIMA model to predict the carbon emissions,carbon inten-sity,and per capita carbon emissions of Gansu Province from 2023 to 2033.Through data learning training,the best model for pre-dicting carbon emissions is ARIMA(2,1,0),Mean Absolute Percentage Error(MAPE)is 0.016 2,the best model for predicting car-bon emission intensity is(0,1,0),MAPE is 0.0328,the best model for predicting per capita carbon emissions is(1,1,0),MAPE is 0.018 6,and the predicted values are all within the 95%confidence interval,indicating high prediction accuracy.The research results indicate that the future carbon emissions and per capita carbon emissions in Gansu Province are still on the rise year by year,while the carbon emission intensity is decreasing year by year.According to the predicted data and trends,it can be seen that in order for Gansu Province to achieve the"dual carbon"goals on schedule,it is necessary to develop strong carbon reduction plans and mea-sures,and enforce and implement them through relevant laws and regulations.

carbon emissionscarbon emission intensityARIMA model

张巨峰

展开 >

陇东学院新能源学院,甘肃庆阳 745000

碳排放量 碳排放强度 ARIMA模型

甘肃省哲学社会科学规划项目

2023YB048

2024

山西大同大学学报(自然科学版)
山西大同大学

山西大同大学学报(自然科学版)

影响因子:0.271
ISSN:1674-0874
年,卷(期):2024.40(4)