Abstract

Accurate prediction of icing thickness on overhead transmission lines is crucial for ensuring the safe and stable operation of the lines during extreme cold weather. This study addresses the issue of significant non-stationary characteristics in the icing thickness data due to the coupling effects of various meteorological factors, such as wind speed and temperature. The authors proposed a prediction method based on Northern Goshawk Optimization (NGO) to optimize Variational Mode Decomposition (VMD), combined with Gated Recurrent Unit (GRU). First, NGO was used to adaptively optimize the key hyperparameters of VMD, achieving effective decomposition of the icing thickness data. Second, the optimized VMD decomposed the icing thickness data into a series of components with different central frequencies but local stationarity, reducing its non-stationarity. Finally, the GRU model independently predicted each decomposed component, and the final prediction was obtained by aggregating the components. The NGO-VMD-GRU model was compared with several traditional prediction models using an overhead transmission line in Henan Province as the case study. The experimental results show that the prediction accuracy of the NGO-VMD-GRU model achieves a Mean Absolute Percentage Error (MAPE) of 3.12%, which is 17.27% lower than the LSTM model, 21.45% lower than the BP neural network, and 12.83% lower than the non-optimized VMD-GRU model, providing a new solution for accurately predicting icing thickness on overhead transmission lines.

References

• Xiong W, Xu H, Xu L X, et al. RF-APJA-MKRVM Combined Prediction Model for Transmission Line Ice Coating Considering Time Cumulative Effects[J]. High Voltage Engineering, 2022(3): 948-957.
• Tu J L, Wang H L. IAOA-LSSVM-Based Prediction of Ice Thickness on Transmission Lines[J]. Modern Electronic Technology, 2024(21): 113-118.
• Jiang X L, Chang H, Hu Q, et al. Prediction and Experimental Study on Equivalent Ice Thickness of Comprehensive Load for Transmission Lines[J]. Proceedings of the CSEE, 2013(10): 177-183.
• Liu H W, Lu J Z, Lai X Y, et al. Short-term Multivariate Grey Prediction Model for Transmission Line Ice Thickness[J]. High Voltage Engineering, 2015(10): 3372-3377.
• Wang X T, Ding J J, Zhang F, et al. Improved BP Neural Network-Based Ice Prediction Technology for Transmission Lines[J]. Mechanical Design and Manufacturing, 2024(9): 306-310.
• Chen Y, Li P, Zhang Z J, et al. PCA-GA-LSSVM-Based Online Prediction Model for Transmission Line Ice Load[J]. Power System Protection and Control, 2019(10): 110-119.
• Ma T, Niu D. Icing Forecasting of High Voltage Transmission Line Using Weighted Least Square Support Vector Machine with Fireworks Algorithm for Feature Selection[J]. Applied Sciences, 2016, 6(12): 438.
• Yu T, Li Y N. Physics-Guided SSA-BiGRU Prediction Model for Ice Thickness on Transmission Lines[J]. Electric Power Science and Engineering, 2022(2): 28-36.
• Chen B, Xu Z M, Jia Y F, et al. VMD-SSA-LSTM-Based Prediction Model for Ice Coating on Overhead Transmission Conductors[J]. Journal of China Three Gorges University (Natural Science Edition), 2024(4): 105-112.
• Luo C, Fan L D, Zhao X W, et al. Study on VMD-IGWO-LSSVM-Based Ice Prediction Model[J]. Power System and Clean Energy, 2021(6): 9-17.
• Yu K, Zhu X, Cao W. Study on Traveling Wave Fault Localization of Transmission Line Based on NGO-VMD Algorithm[J]. Energies, 2024, 17(9).
• Wang C Q, Wu L W, Deng Z B, et al. Review of Time Cumulative Overhead Transmission Line Ice Prediction Models and Algorithms[J]. China Electric Power, 2024(6): 153-164.
• Li X Y, Zhou Q, Zhang H X, et al. Mid-Long-Term Power Load Forecasting of Building Group Based on Modified NGO[J]. Energies, 2025, 18(3): 668.
• He M, Wang H, Thwin M. A machine learning technique for optimizing load demand prediction within air conditioning systems utilizing GRU/IASO model[J]. Scientific Reports, 2025, 15(1): 3353.
• Ciechulski, T.; Osowski, S. Wind Power Short-Term Time-Series Prediction Using an Ensemble of Neural Networks. Energies 2024, 17, 264. https://doi.org/10.3390/en17010264.
• Tryhuba, I.; Tryhuba, A.; Hutsol, T.; Cieszewska, A.; Andrushkiv, O.; Glowacki, S.; Bryś, A.; Slobodian, S.; Tulej, W.; Sojak, M. Prediction of Biogas Production Volumes from Household Organic Waste Based on Machine Learning. Energies 2024, 17, 1786. https://doi.org/10.3390/en17071786.
• Liu J, Cong L M, Xia Y Y, et al. Combined Model of DBO-VMD and IWOA-BILSTM Neural Networks for Short-term Power Load Prediction[J]. Power System Protection and Control, 2024(8): 123-133.
• Yang B, Duan J H, Li M W, et al. Optimization of Offshore Hybrid Photovoltaic-Wave Energy Converter Arrays Based on Improved Bald Eagle Search Algorithm[J]. Power System Technology, 2024, 48(6): 2480-2489.
• Jiang F, Lin Z Y, Wang W Y, et al. Optimal Bagging Ensemble for Ultra-short-term Multivariate Load Prediction Considering Minimum Average Envelope Entropy Load Decomposition[J]. Proceedings of the CSEE, 2024, 44(5): 1777-1789.
• Abumohsen M, Owda A Y, Owda M. Electrical Load Forecasting Using LSTM, GRU, and RNN Algorithms[J]. Energies, 2023, 16(5): 2283.
• Liu C, He Q H, Lu Y J, et al. PSOEM-LSSVM Ice Prediction Model for Transmission Lines[J]. Journal of Electric Power Science and Technology, 2020(6): 131-137.
• Fan X, Wang R, Yang Y, et al. Transformer–BiLSTM Fusion Neural Network for Short-Term PV Output Prediction Based on NRBO Algorithm and VMD[J]. Applied Sciences, 2024, 14(24): 11991.
• Tao K, Xu M, Wang Q. ASSA-VMD-SI and Frechet method of pipe vibration for noise reduction and leakage identification[J]. Measurement, 2025, 242: 116277.
• Dou X, Quan X, Wu Z, Hu M, Yang K, Yuan J, Wang M. Hybrid Multi-Agent Control in Microgrids: Framework, Models and Implementations Based on IEC 61850. Energies. 2015; 8(1):31-58. https://doi.org/10.3390/en8010031.
• Jin X Z, Zhao S S, Chang H, et al. Modeling of Superheated Steam Temperature Characteristics Based on Golden Jackal Optimization Variational Mode Decomposition and Temporal Convolutional Network[J]. Proceedings of the CSEE. [2025-01-03].