Modeling diameter of jet grouting columns using Bayesian framework

Abstract

This study proposes a new model in explicit form to predict the diameter of the jet grouting column of three popular jet grouting systems (i.e., single, double, and triple). The proposed model can quantify the uncertainty associated with the prediction. Bayesian model selection was used to determine the optimal models and, the bootstrap sampling method was adopted to avoid bias in the collected database. The predictions agree well with the measured data, and the 95% credible intervals can cover almost all the measured data on the testing dataset, which indicates the robustness of the proposed model. A simple formula combined with a table was established to facilitate the engineers in applying the proposed model. The results reveal that soil type has a significant impact on diameter prediction. Uncertainty quantification can reveal possible fluctuations in column diameter and is vital to get a cost‐efficient design.

Examination of Determinants and Predictive Modeling of Artificially Frozen Soil Strength Utilizing the XGBoost Algorithm

“…To address this issue, the data were standardized. Normalization helps eliminate scale differences between features, ensuring that the model's impact on each feature is relatively balanced and facilitating faster convergence of the algorithm [40]. Standardization transformed the data into a normal distribution with a mean of 0 and a standard deviation of 1, as represented by Equation (1).…”