Jul 2026· SAE technical paper series· 0 citations· 4 references
Abstract
Accurate prediction of ground settlement induced by rectangular pipe jacking, a
prevalent trenchless technology in urban infrastructure development, remains a
significant challenge. This study addresses this by developing and evaluating a
robust machine learning (ML) framework. Leveraging 104 sets of field monitoring
data from the Liuye Avenue West Extension rectangular pipe jacking project in
Hunan, China, key construction parameters including jacking force, advance rate,
and grouting pressure were utilized as inputs to predict ground settlement. A
Particle Swarm Optimization (PSO) algorithm was integrated for automated
hyperparameter tuning of six distinct ML models: standalone Least Squares
Support Vector Machine (LSSVM), Backpropagation Neural Network (BPNN), Random
Forest (RF), and their respective PSO-optimized counterparts. Comprehensive
performance evaluation using Mean Squared Error (MSE), Mean Absolute Error
(MAE), and Coefficient of Determination (R^2) revealed that the
PSO-LSSVM hybrid model achieved superior predictive accuracy and generalization
capability. Specifically, on the test dataset, the PSO-LSSVM model yielded an
MSE of 0.367, MAE of 0.424, and an R^2 of 0.941. These findings
demonstrate that the proposed PSO-enhanced LSSVM model significantly outperforms
baseline models, offering a highly effective and reliable tool for predicting
ground deformation in similar complex pipe jacking projects.
Ground settlement induced by shield tunnel boring machine (TBM) excavation is a major geotechnical concern in urban tunneling because it may affect the safety of adjacent structures and underground infrastructure. In this study, machine learning models were developed to predict the maximum settlement induced by shield TBM excavation using a three-dimensional numerical analysis database comprising 320 simulation cases generated from combinations of tunnel diameter (D), ground elastic modulus (E), face pressure (FP), and backfill pressure (BP). Random forest (RF) and extreme gradient boosting (XGBoost) models were developed and compared with an existing regression-based settlement prediction equation. Predictive performance and generalization capability were evaluated using random split and GroupKFold validation techniques. Under random split validation, RF achieved the highest predictive performance, with a coefficient of determination of 0.997 and a root mean square error of 0.438 mm, followed by XGBoost. Both machine learning models outperformed the existing settlement prediction equation. However, model performance decreased substantially under GroupKFold validation, indicating limited generalization capability under unseen D–E grouped conditions. The results demonstrate that the developed machine learning models provide accurate predictions within the range of tunnel–ground conditions represented by the adopted numerical analysis database. The findings highlight the importance of evaluating both predictive performance and generalization capability, particularly when machine learning models developed from numerical analysis databases are applied beyond the conditions represented in the training database.
Ji-seok Yun, Wan-kyu Yoo, Gi-Jun Lee et al.· Applied Sciences· 0 citations
With the rapid development of inland waterway transportation, the demand for reliability and accuracy in hydrological forecasting has been increasing. Inland water levels are affected by rainfall, upstream flood discharge, and seasonal factors, exhibiting highly nonlinear and non-stationary characteristics. Traditional deterministic prediction models can hardly meet the requirements of refined scheduling. This paper proposes a hybrid water level prediction model that integrates Empirical Mode Decomposition (EMD), Particle Swarm Optimization (PSO), and Radial Basis Function Neural Network (RBFNN), and innovatively introduces an uncertainty quantification mechanism based on the statistic. First, EMD is used to decompose the original complex water level signal into multiple Intrinsic Mode Functions (IMFs) to reduce data nonstationarity. Second, for each IMF component, the PSO algorithm is adopted to globally optimize the centers and spread constants of the RBF neural network, constructing high-precision base prediction models. Finally, the uncertainty coefficient is calculated. Using actual water level data from the Port of Guigang as the experimental object, the results show that the prediction accuracy of the hybrid model reaches R2=0.9415, and the α coefficient can effectively quantify the dynamic risk of prediction results. This study not only provides a high-precision technical approach for inland water level prediction, but its uncertainty quantification results also offer a scientific basis for the reliability evaluation and risk early warning of hydrological forecasting.
Qi Xu, Xiao-Nuo Zhu, Cheng Zeng et al.· International Conference on...· 0 citations
Deep excavations are high-risk geotechnical activities, and accurate prediction of diaphragm-wall settlement is important for construction monitoring and deformation control. This study investigates cumulative vertical settlement at 23 diaphragm-wall monitoring points from the deep excavation of Tianjin Goldin Finance 117 in Tianjin, China. Seven prediction models—a naïve persistence model, autoregressive integrated moving average (ARIMA), K-nearest neighbors (KNN), multilayer perceptron (MLP), gated recurrent unit (GRU), Transformer, and XGBoost—were evaluated using a unified five-fold rolling-origin expanding-window validation scheme. GRU achieved the best overall baseline performance, with a mean R2 of 0.9172, a mean absolute error (MAE) of 0.1009 mm, a root mean square error (RMSE) of 0.1341 mm, and a mean absolute percentage error (MAPE) of 0.5875%. GRU was subsequently optimized using the crow search algorithm (CSA), the genetic algorithm (GA), and the whale optimization algorithm (WOA). GRU-WOA achieved the best numerical performance, with a mean R2 of 0.9289 and an RMSE of 0.1215 mm. Relative anomaly levels were further identified from predicted settlement-change rates to characterize temporal concentration and spatial clustering of settlement-change activity. The proposed framework can support priority inspection and targeted monitoring, although the resulting anomaly levels represent project-relative statistical deviations rather than code-based engineering risk classes.
Accurate prediction of the water-conducting fracture zone height is essential for water inrush prevention and safe production in coal mining. Based on extensive in-situ measurements collected from longwall panels in different mining districts, five key indicators-mining thickness, mining depth, coal seam dip angle, panel length along dip, and the hard-rock lithology ratio coefficient-were analysed. Using regression analysis, the empirical formula for predicting the water-conducting fracture zone height was refined and a multivariate nonlinear regression model was fitted. An optimal BP neural network model with the Levenberg-Marquardt algorithm and a 5:8:4:1 topology was validated, and subsequently an LWMA-PSO-BP neural network model was developed by jointly introducing the mutation operator from genetic algorithms and a linearly decreasing inertia weight (LDIW) strategy. Model fitting accuracy and generalisation were evaluated; the results indicate that the LWMA-PSO-BP model achieved the best overall performance, with a mean absolute error of 2.40 m and a mean absolute percentage error of 4.27%. In the Hebi mining district, a joint geophysical investigation integrating a microtremor survey, borehole coring, and drilling fluid loss measurements was conducted, and the water-conducting fracture zone heights for Panels 2301, 2302, 2303, and 2304 at Hemei No. 5 Mine were determined as 129.05 m, 134.21 m, 141.50 m, and 138.20 m, respectively. Field validation shows that the relative errors of the multivariate nonlinear regression and BP neural network models were 5.52% and 4.85%, respectively, whereas the LWMA-PSO-BP model yielded a relative error of only 2.99%. These results provide a reference for predicting the water-conducting fracture zone height under varied coal mining conditions.
Weiyu Guo, Yu Wang, Yi Tan et al.· Scientific Reports· 0 citations
Blast-induced ground vibrations, commonly measured in terms of Peak Particle Velocity (PPV), can affect adjoining communities, damage structures, and undermine the integrity of rock masses. Therefore, accurate PPV prediction is a prerequisite for making blasting sustainable and safe. This paper introduces a novel hybrid machine learning method that integrates Extreme Gradient Boosting (XGBoost) and Grey Wolf Optimization (GWO) in predicting PPV using 88 datasets obtained at the Akdaglar Quarry, Istanbul. Nine important input parameters were taken into account, including geological aspects and blasting design. In comparison to the traditional scaled distance, a new parameter modified scaled distance (MSD), was derived through a genetic algorithm, and showed better correlation with PPV (r = –0.869). Model performance was improved by optimal tuning of XGBoost hyperparameters through the utilization of the GWO method. The GWO-XGBoost model performed better than SVM and stand-alone XGBoost, with R = 0.9916, R² = 0.9748, RMSE = 1.0291 mm/s, and MAPE = 3.1105%. The monitoring distance and scaled distance are the most influential predictors, according to SHAP (Shapley Additive Explanations) analysis. The results demonstrate that evolutionary optimization algorithms substantially enhance PPV prediction accuracy, offering practical implications for pre-blast vibration assessment and sustainable quarry design.
Arnob Deb, Tahmid Zaman Raad, P. Roy et al.· Engineering· 0 citations
Accurate and rapid prediction of groundwater levels (GWL) is essential for effective groundwater management. Machine learning models are efficient tools for GWL prediction, but individual models often suffer from limited generalization due to inherent randomness. This study proposed a stacking-based GWL prediction framework suitable for arid regions in Northwest China. Feature variables affecting GWL were selected using variable importance in projection (VIP). Then, three machine learning models—artificial neural networks (ANN), random forests (RF), and Light Gradient Boosting Machine (LightGBM)—were developed, and their outputs were integrated using a support vector regression (SVR)-based stacking method to enhance the accuracy of GWL prediction. The results show that the factors of influencing GWL changes vary significantly across different regions, and selecting the most contributive feature variables is beneficial for model construction. Among the individual models, the RF model demonstrated higher accuracy and more stable performance, outperforming the ANN and LightGBM models. However, individual models exhibited poor generalization during validation. In contrast, the stacking model maintained high performance, demonstrating superior generalization. Compared to the best-performing individual model (RF) in validation period, the Nash–Sutcliffe efficiency (
NSE
) and Kling–Gupta efficiency (
KGE
) of stacking model improved by 0.11–0.66 and 0.05–0.41, the correlation coefficient (
R
2
) increased by 0.05–0.3, and root mean square error (
RMSE
) reduced by 0.01–0.1 m. In the stacking simulation, RF had the highest average contribution (80.2%), followed by ANN (13.9%) and LightGBM (5.9%). This study provides a stacking simulation framework based on machine learning methods for precise groundwater level simulation, which can serve as a reference for groundwater level simulation in other regions.
Xunzhen Cui, Xiaoxia Du, Haixia Dong et al.· Frontiers in Water· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
