文章详情

摘要

为提高OPGW光缆寿命预测的准确性，利用2011—2023年广州地区多因素气象数据，对光缆在温度波动、风荷载等工况下的退化规律进行了研究。基于热胀冷缩与载荷模型构建光缆长度变化序列，并应用Autoformer建立剩余寿命预测方法。数据预处理中采用拉格朗日插值修复异常值，并通过最小—最大归一化完成特征缩放。对比LSTM、Bi-LSTM与Bi-LSTM+Attention等模型，结果表明Autoformer在RMSE、MAE与MSE指标上均取得最优表现，能够更准确表征光缆余长的长期变化趋势。研究表明，该方法可提升寿命预测精度，为OPGW运行维护及更换策略制定提供数据支持。

关键词

OPGW；寿命预测；Autoformer；时间序列预测

参考文献

1. [1] BORZYCKI K. Influence of Temperature and Aging on Polarization Mode Dispersion of Tight-buffered Optical Fibers and Cables [J].Journal of Telecommunications and Information Technology，2005（3）：96-104.
2. [2] GÜNDAY A，KARLIK S E. Optical Fiber Distributed Sensing of Temperature, Thermal Strain and Thermo-mechanical Force Formations on OPGW Cables Under Wind Effects [C]//2013 8th International Conference on Electrical and Electronics Engineering (ELECO).Bursa：IEEE，2013：462-467.
3. [3] MALYSZKO O，ZENCZAK M. Current-Carrying Capacity of Overhead Power Transmission Lines in Different Weather Conditions [C]//XV International Symposium on Theoretical Engineering.Lübeck：VDE，2009：1-4.
4. [4] GRUNVALDS R，CIEKURS A，PORINS J，et al. Evaluation of Fibre Lifetime in Optical Ground Wire Transmission Lines [J].Latvian Journal of Physics and Technical Sciences，2017，54（3）：40-49.
5. [5] BURDIN V A，NIKULINA T G，PRAPORSHCHIKOV D E. Method for Predicting the Lifetime of an Optical Cable After the Maintenance Cycle [C]//Optical Technologies for Telecommunications 2021.Bellingham：SPIE，2022：246-251.
6. [6] NIZHGORODOV A O. Simple Approximate Formula for Forecast of the Lifetime of Optical Fibers of a Cable Communication Line [J].Infokommunikacionnye Tehnologii，2021，19（1）：29-34.
7. [7] HAN W X，HAO C S，KONG D J，et al. Cable Temperature Prediction Based on RF-GPR for Digital Twin Applications [J/OL].Applied Sciences，2023，13（13）：7700[2025-09-23].https://doi.org/10.3390/app13137700.
8. [8] HAN W X，YANG G，HAO C S，et al. A Data-Driven Model of Cable Insulation Defect Based on Convolutional Neural Networks [J/OL].Applied Sciences，2022，12（16）：8374[2025-09-11].https://doi.org/10.3390/app12168374.
9. [9] WU H X，XU J H，WANG J M，et al. Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting [J].Advances in Neural Information Processing Systems，2021，34：22419-22430.
10. [10] ZHOU S Y，ZHANG X J，LIU J Z，et al. Long-Term Prediction of Particulate Matter2.5 Concentration with Modal Autoformer Based on Fusion Modal Decomposition Algorithm [J/OL].Atmosphere，2023，15（1）：4[2025-08-26].https://doi.org/10.3390/atmos15010004.
11. [11] ASHFAQ A，CHEN Y，YAO K，et al. Microbending Loss Caused by Stress in Optical Fiber Composite Low Voltage Cable [C]//2018 12th International Conference on the Properties and Applications of Dielectric Materials (ICPADM).Xi'an：IEEE，2018：483-486.
12. [12] SETAYESHGAR A. Principles of Mechanical Design in Overhead Transmission Lines [J].Finland: Aalto University, 2016：1-73.
13. [13] LI Y Q，ZHAO H W，YUE Z X，et al. Real-Time Intelligent Prediction Method of Cable's Fundamental Frequency for Intelligent Maintenance of Cable-Stayed Bridges [J/OL].Sustainability，2023，15（5）：4086[2025-10-03].https://doi.org/10.3390/su15054086.
14. [14] QIN J J，GAO C，WANG D K. LCformer: Linear Convolutional Decomposed Transformer for Long-Term Series Forecasting [C]//International Conference on Neural Information Processing.Singapore：Springer Nature Singapore，2023：53-67.
15. [15] ALRASHEEDI F，ZHONG X，HUANG P C. Padding Module: Learning the Padding in Deep Neural Networks [J].IEEE Access，2023，11：7348-7357.
16. [16] HENRY M. An Ultra-Precise Fast Fourier Transform [J/OL].Measurement：Sensors，2024，32：101039[2025-08-19].https://doi.org/10.1016/j.measen.2024.101039.

作者简介

熊奕（1997—），男，汉族，湖北武汉人，助理工程师，硕士，研究方向：通信工程。