SSSDScientific and Social Sustainable Development3066-89643066-8980Art and Design10.61369/SSSD.2025100043Article基于GRU 模型的人工智能分析法在滑坡预测领域 的研究及应用https://artdesignp.com/journal/SSSD/1/10/10.61369/SSSD.2025100043郭国林,王琪,唐珺,张添彬20251102025-07-28滑坡作为山地与库区地区最为频繁且危害严重的地质灾害之一，其演化受地质条件、降雨入渗、库水位消涨及人为扰动等多种因素耦合控制。传统的统计回归与静态机器学习方法难以有效捕捉台阶式、非平稳的位移特征。近年来，深度循环神经网络在滑坡预测中展现了良好的建模能力，其中门控循环单元（GRU）以参数少、训练快、收敛稳定等优势，在工程部署中具有应用潜力。本文提出“趋势— 周期分解 + GR U 预测”的方法框架，并结合福建泉州梧垵溪与重庆二道河两处典型滑坡案例开展应用验证。结果表明，该方法在趋势项拟合上R² ≥ 0.995，周期项预测的RMSE约为1.7–3.4 mm，MAPE 控制在8%–12%，优于SVM 与部分集成模型。研究表明，基于GRU 的预测框架能够显著提升地质灾害预警的稳定性与可解释性，为库区与河道重点工程的安全运行提供了支撑。滑坡位移预测,GRU 模型,时间序列分解,GNSS 监测,地质灾害预警

[1]Zhang, W., Li, H., Tang, L., et al. Displacement prediction of Jiuxianping landslide using gated recurrent unit (GRU) networks. Acta Geotechnica, 2022, 17: 1367–1382. https://doi.org/10.1007/s11440-022-01495-8[2]Wang, F.; Zhou, G.; Hu, H.; Wang, Y.; Fu, B.; Li, S.; Xie, J. Reconstruction of LoD-2 Building Models Guided by Façade Structures from Oblique Photogrammetric Point Cloud. Remote Sens. 2023, 15, 400. https://doi.org/10.3390/rs15020400[3]Huang, G., Du, S., Wang, D. GNSS techniques for real-time monitoring of landslides: a review. Satellite Navigation, 2023, 4: 5. https://doi.org/10.1186/s43020-023-00095-5[4]Wang, J., Zhang, Z. Landslide Deformation Prediction Based on GNSS Time Series and RNN. Remote Sensing, 2021, 13(6): 1055. https://doi.org/10.3390/rs13061055[5]Zhou, C., Yin, K., Cao, Y., et al. Displacement prediction of step-like landslides via kernel extreme learning machine. Landslides, 2018, 15(11): 2211–2225. https://doi. org/10.1007/s10346-018-1022-0.[6]Miao, F., Wu, Y., Xie, Y. et al. Prediction of landslide displacement with step-like behavior based on multialgorithm optimization and a support vector regression model. Landslides 15, 475–488 2018. https://doi.org/10.1007/s10346-017-0883-y.[7]Xu, S., Niu, R. Displacement prediction of Baijiabao landslide based on EMD and LSTM neural networks. Computers & Geosciences, 2018, 111: 87–96. https://doi. org/10.1016/j.cageo.2017.10.013.[8]Wang, H., Chen, Z., Du, H., et al. A dynamic prediction model of landslide displacement based on VMD–SSO–LSTM approach. Scientific Reports, 2024, 14: 59717. https://doi.org/10.1038/s41598-024-59517-2.[9]Ma, W., Dong, J.，Wei, Z., et al. Landslide Displacement Prediction With Gated Recurrent Unit and Spatial-Temporal Correlation. Frontiers in Earth Science, 2022, 10: 950723. https://doi.org/10.3389/feart.2022.950723.[10]Bai, D.; Lu, G.; Zhu, Z.; Zhu, X.; Tao, C.; Fang, J.; Li, Y. Prediction Interval Estimation of Landslide Displacement Using Bootstrap, Variational Mode Decomposition, and Long and Short-Term Time-Series Network. Remote Sens. 2022, 14, 5808. https://doi.org/10.3390/rs14225808.