Prediction of Surface Subsidence of Deep Foundation Pit Based on Wavelet Analysis

Predicting surface settlement in deep foundation pit engineering plays a central role the safety of construction. Recently, static or dynamic methods are usually applied to predict ground projects. In this work, we propose model combining wavelet noise reduction and radial basis neural network (XW-RBF) reduce interference monitoring data. The results show that XW-RBF predicts an average relative error 0.77 root square 0.13. prediction performance is better than original data with structure has higher accuracy. caused by construction surrounding environment can be removed via method, discreteness reducing 30%. More importantly, our reflect law change future trend high credibility. findings study indicate could optimize for accuracy during prediction, which inspires potential application engineering.