基于波动性和趋势性的民航碳价格预测方法  

Method for Predicting Carbon Prices in Civil Aviation Based on Volatility and Trend

作　　者：宋瑞涵 陈静杰 SONG Rui-han;CHEN Jing-jie(College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《计算机仿真》2024年第9期26-30,共5页Computer Simulation

基　　金：国家社会科学基金(22BJY020);中国民航局安全能力项目(SKZ49420210017);2021年度天津市教委社会科学重大项目(2021JWZD39)。

摘　　要：“双碳”目标已成为国际社会重要普适价值。市场措施是不可或缺的手段。而碳价则成为民航落实减碳责任、获取成本效益的重要因素。准确预判碳价格成为重大基础性问题。提出一种基于多频段特征工程的双向长短期记忆网络的碳价格预测模型,旨在破解碳价格受外界不确定因素影响而至的显著波动难题,助力民航“双碳”行动落实。考虑碳价格数据的复杂的非线性时间序列,采用自适应噪声的完整集成经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)法获取数据特征刻画。然后以高斯白噪声(White Gaussian noise,WNG)叠加进入样本熵简单序列分量以刻画原始序列强波动性,进而获取最终特征工程。最后通过门控循环单元(Gate Recurrent Unit,GRU)对碳价格进行预测。实证分析结果表明了预测模型的有效性,可为航空公司履约降碳责任提供一定方法支持。"Double carbon" target has become an important and universal value in the international community.Market measures are an indispensable means.The carbon price has become an important factor for the civil aviation to implement the carbon reduction responsibility and obtain the cost-benefit.Accurate prediction of carbon prices has become a major fundamental issue.This paper proposes a carbon price prediction model based on two-way short-term memory network based on multi-band characteristic engineering,aiming to solve the problem of significant fluctuations of carbon price affected by external uncertainties,and help facilitate the implementation of the "two-carbon" action of civil aviation.Considering the complex nonlinear time series of carbon price data,the white noise complete integrated empirical mode decomposition(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN) method is used to obtain the data characteristics.Then,the Gaussian white noise(White Gaussian noise,WNG) is added to enter the sample entropy simple sequence component to describe the strong volatility of the original sequence,and then to obtain the final feature engineering.Finally,the carbon price is predicted by the gating cycle unit(Gate Recurrent Unit,GRU).The empirical analysis results show the effectiveness of the prediction model,which can provide some method support for airlines to fulfill the contract.

关 键 词：“双碳”目标 时间序列预测 数据分解 门控循环单元 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]