基于多头自注意力-Bi-LSTM模型的盾构掘进引发的土体沉降预测

杨明辉,宋牧原,付大喜,郭炎伟,卢贤锥,张文聪,郑伟龙

Prediction of shield tunneling-induced soil settlement based on multi-head self-attention-Bi-LSTM model

Minghui YANG,Muyuan SONG,Daxi FU,Yanwei GUO,Xianzhui LU,Wencong ZHANG,Weilong ZHENG

表 6 不同深度学习模型的土体沉降预测性能指标对比（案例2）

Tab.6 Comparison of soil settlement prediction performance metrics across different deep learning models (case 2)

监测点 模型 MSE/mm MAE/mm MAPE/% RE/mm 225-左 1 0.122 0.295 9.72 0.38 2 0.141 0.357 15.55 0.59 225-右 1 0.115 0.276 8.25 0.39 2 0.138 0.334 14.12 0.41 365-左 1 0.167 0.378 9.28 0.37 2 0.181 0.415 14.23 0.61 365-右 1 0.157 0.285 8.98 0.38 2 0.174 0.392 15.36 0.57 均值 1 0.140 0.309 9.06 0.38 2 0.159 0.375 14.82 0.55