摘要
针对碳排放数据的随机性及波动性因素所导致预测精度不高等问题,提出基于序列重构的VMD-SSA-LSSVM(variational mode decomposition-sparrow search algorithm-least square support vector machine)组合模型进行短期碳排放预测.首先将区域的碳排放数据序列经过VMD进行分解得到4个不同中心频率的子序列和一个残差序列,降低数据不规律性对碳排放预测带来的干扰;接着对分解后的各个分量进行序列重构,提高对突变点的预测精度;然后根据不同分量各自的特点,使用SSA优化核函数中相关的参数,对重构后得到的各个序列建立SSA-LSSVM预测模型;最后将所有序列的预测值融合得到预测结果.算例结果表明基于序列重构的组合模型能够有效提高短期碳排放预测的精度.
Due to the randomness and volatility of carbon emission series,the prediction accuracy was not high.A combined VMD-SSA-LSSVM(variational mode decomposition-sparrow search algorithm-least square support vector machine)model based on sequence reconstruction was proposed for short-term carbon emission prediction.Initially,VMD decomposed the daily carbon emission data series into four sub-sequences with distinct center frequencies,along with one residual sequence to mitigate irregular data interference.Subsequently,sequence reconstruction was applied to each decomposed sequence to enhance prediction accuracy,particularly at mutation points.Utilizing the relevant parameters of SSA optimization kernel function,an SSA-LSSVM prediction model was established for each sequence post-reconstruction,taking into account the characteristics of different components.At last,the predicted values were fused to obtain the predicted results.The results showed that the combined model based on sequence reconstruction could effectively improve the accuracy of short-term carbon emission prediction.
作者
徐正林
程志友
张帅
杨猛
XU Zhenglin;CHENG Zhiyou;ZHANG Shuai;YANG Meng(School of Internet,Anhui University,Hefei 230039,China;Power Quality Engineering Research Center of the Ministry of Education,Anhui University,Hefei 230601,China;School of Electronic and Information Engineering,Anhui University,Hefei 230601,China)
出处
《安徽大学学报(自然科学版)》
北大核心
2025年第4期28-37,共10页
Journal of Anhui University(Natural Science Edition)
基金
国家自然科学基金资助项目(61672032)
安徽省自然科学基金资助项目(2108085QE237)。
关键词
短期碳排放预测
序列重构
变分模态处理
最小二乘支持向量机
short-term carbon emission prediction
sequence reconstruction
variational mode decomposition
least square support vector machine
