为解决盾构施工过程中姿态预测不精准的问题，提出一种基于最优组合赋权的EMD-GRU盾构竖向姿态预测方法。首先，对获取到的竖向姿态数据（俯仰角）进行数据预处理；其次，为提高模型的预测精度，将原始数据集基于经验模态分解（empirical mode decomposition，EMD）的方式划分为若干个较为平稳的子序列；然后，基于门控循环单元（gated recurrent unit，GRU）神经网络分别建立各个子序列的预测模型；最后，基于最优组合赋权法将各个子序列的预测结果进行融合，以得到模型的最终预测结果。为验证所建立的EMD-GRU预测模型的可靠性，依托昆明地铁5号线实际盾构施工项目加以应用，并与GRU、SVR、BP模型的预测结果进行对比。结果表明，所构建的EMD-GRU预测模型具有较高的预测精度和泛化能力，可为盾构姿态精准预测提供一种新的思路。
In order to solve the problem of inaccurate prediction
of the attitude during shield construction, a prediction method of EMDGRU
(empirical mode decompositiongated recurrent unit)
shield vertical posture based on optimal combination weighted EMDGRU is proposed. First, the acquired vertical attitude data (pitch
angle) are preprocessed; Second, in order to improve the prediction accuracy of
the model, the original data set is divided into several relatively stable subsequences based on EMD; Third, based on GRU neural network, the
prediction model for each subsequence is established
respectively; Finally, the prediction results of each subsequence are merged based on the optimal combination weighting
method to get the final prediction results of the model. In order to verify the
reliability of the established EMDGRU prediction
model, it is applied to shield construction project of Kunming metro line 5,
and the results are compared with the prediction results of GRU, SVR, and BP
models. The results show that the EMDGRU prediction
model has high prediction accuracy and generalization ability, which provides a
new idea for accurate prediction of shield attitude.
