摘要
针对土质滑坡位移多步预测方法的缺乏以及在多时间步长下预测误差较大的问题,本文提出了一种基于多级注意力机制并行模型的滑坡位移多步预测方法。采用多输入多输出的预测策略,通过含有多头注意力机制的Transformer编码器网络分支以及经全局注意力机制(GAM)优化的双向门控循环单元(BiGRU)网络分支,两个网络分支并行处理滑坡历史监测数据,之后对并行网络提取到的滑坡特征信息通过交叉注意力机制(CAM)进行特征融合后输出预测的滑坡多步位移值。实验结果表明,多级注意力机制模型在滑坡位移多步预测中平均绝对误差(MAE)、均方根误差(RMSE)分别为2.17 mm、3.05 mm,决定系数(R^(2))为0.9689,相较于其他模型误差最低,决定系数结果最优,在长时间步下的预测效果更加稳定,有利于提前预知滑坡发展动向,为滑坡的预防与治理提供了重要的技术支持。
Aiming at the lack of multi-step prediction methods for soil landslide displacement and the issue of significant prediction errors over extended time horizons,this paper proposes a multi-step landslide displacement prediction method based on a parallel model with a multi-level attention mechanism.The method employs a multi-input multi-output prediction strategy.Utilizing a Transformer encoder branch incorporating a multi-head attention mechanism and a bidirectional gated recurrent unit(BiGRU)branch optimized with a global attention mechanism(GAM),the two parallel network branches process historical landslide monitoring data.The landslide feature information extracted by the parallel networks is then fused via a cross attention mechanism(CAM),subsequently outputting the predicted multi-step displacement values.Experimental results demonstrate that the multi-level attention mechanism model achieves a mean absolute error(MAE)of 2.17 mm,a root mean square error(RMSE)of 3.05 mm,and a coefficient of determination(R^(2))of 0.9689 in multi-step landslide displacement prediction.Compared to other models,it yields the lowest errors and the optimal R^(2)result.The model exhibits more stable prediction performance over long time horizons,facilitating the early anticipation of landslide development trends.This provides crucial technical support for landslide prevention and mitigation.
作者
任冯
肖慧
冯沂萱
吴雨洁
艾玉洁
Ren Feng;Xiao Hui;Feng Yixuan;Wu Yujie;Ai Yujie(School of Geophysics and Measurement-Control Technology,East China University of Technology,Nanchang 330013,China)
出处
《电子测量技术》
2026年第1期40-49,共10页
Electronic Measurement Technology
基金
江西省自然科学基金(20212BAB203004)
东华理工大学研究生创新基金(YC2023-S579)项目资助。
