交通运输系统工程与信息

交通运输系统工程与信息 ›› 2026, Vol. 26 ›› Issue (4): 1-16.DOI: 10.16097/j.cnki.1009-6744.2026.04.001

• 综合交通运输体系 •    下一篇

考虑非线性时滞特征的交通运输业碳排放预测

王庆荣*a，张金鹏a，朱昌锋b   

  1. 兰州交通大学，a. 电子与信息工程学院；b. 交通运输学院，兰州 730070
  • 收稿日期:2026-03-06 修回日期:2026-04-13 接受日期:2026-04-20 出版日期:2026-08-25 发布日期:2026-08-22
  • 作者简介:王庆荣（1977— ），女，宁夏固原人，教授
  • 基金资助:
    国家自然科学基金 (72161024)；甘肃省教育厅“双一流”重大研究项目 (GSSYLXM-04)

Forecasting Carbon Emissions in Transportation Sector Considering Nonlinear Time-lag Effects

WANG Qingrong*a, ZHANG Jinpenga, ZHU Changfengb   

  1. a. School of Electronic and Information Engineering; b. School of Transportation, Lanzhou Jiaotong University, Lanzhou 730070, China
  • Received:2026-03-06 Revised:2026-04-13 Accepted:2026-04-20 Online:2026-08-25 Published:2026-08-22
  • Supported by:
    National Natural Science Foundation of China (72161024); Major Research Project of the "Double First-Class" Initiative of the Education Department of Gansu Province (GSSYLXM-04)

PDF

摘要： 针对交通运输业碳排放序列高度非线性、非平稳性及驱动因素间存在复杂耦合与信息冗余导致预测精度受限的问题，本文提出一种结合两阶段特征筛选（LMIC-LASSO）、逐次变分模态分解（SVMD）、改进长颖燕麦优化算法（IAOO）及双向长短期记忆网络（BiLSTM）的混合预测模型。本文构建基于时滞最大互信息系数（TLMIC）与LASSO回归的两阶段特征筛选策略，精准识别关键驱动因素的时滞效应并剔除冗余变量。引入SVMD将原始碳排放序列自适应分解为多个平稳模态分量，降低数据的非平稳性；利用混合混沌扰动、适应t分布变异和动态反向学习策略改进长颖燕麦优化算法（AOO），并利用IAOO对BiLSTM网络的关键超参数自适应寻优，避免模型陷入局部最优。通过各模态分量分别构建基于IAOO-BiLSTM的预测模型，并对预测结果集成重构。以中国交通运输业1990—2023年的碳排放数据验证模型，结果表明：所提模型较最优对比模型的均方根误差（RMSE）、平均绝对误差（MAE）和平均绝对百分比误差（MAPE）分别降低21.88%、23.33%和24.32%，显著提高了交通运输业碳排放量的预测准确性。

关键词: 综合交通运输, 碳排放预测, 混合预测模型, 交通运输业, 非线性时滞, 改进长颖燕麦优化算法

Abstract: To address the limitations of prediction accuracy caused by high nonlinearity, non-stationarity, and complex driver coupling with information redundancy in transportation carbon emission series, a hybrid prediction model which combines a two-stage feature selection (TLMIC-LASSO), Successive Variational Mode Decomposition (SVMD), Improved Animated Oat Optimization algorithm (IAOO), and Bidirectional Long Short-Term Memory network (BiLSTM) is proposed. The two-stage feature selection strategy based on Time-Lagged Maximal Information Coefficient (TLMIC) and LASSO regression is constructed to identify the time-lag effects of key driver accurately and eliminate redundant variables. The SVMD is introduced to decompose the original series of carbon emission adaptively into multiple stationary modal components, thereby reducing the non-stationarity of data. The animated oat optimization algorithm(AOO) algorithm is improved via hybrid chaotic perturbation, adaptive tdistribution mutation, and dynamic opposition-based learning strategies, and the IAOO algorithm is utilized to optimize the key hyperparameters of BiLSTM network adaptively to prevent the model from falling into the local optima. Finally, an IAOO-BiLSTM prediction model is built for each modal component, and the prediction results are integrated. The model is validated with the data of transportation carbon emission from 1990 to 2023 in China. Results indicate that compared with the optimal contrast model, the proposed model reduces the RMSE, MAE, and MAPE by 21.88% , 23.33% , and 24.32% respectively, which significantly improves the prediction accuracy of transportation carbon emission. 

Key words: integrated transportation, carbon emission prediction, hybrid prediction model, transportation sector, nonlinear time-lag, improved animated oat optimization algorithm

中图分类号: