Jul 2026· Proceedings of the Institution of Civil Engineers : Water Management· Vol 179, pp. 96-119· 0 citations· 105 references
Abstract
Four machine learning (ML) techniques (artificial neural network with particle swarm optimisation (ANN–PSO), adaptive neuro-fuzzy inference system (ANFIS), multivariate adaptive regression splines (MARS) and M5 Tree) were used to model scour depth around bridge piers under clear-water scour (CWS) conditions. A total of 912 datasets collected under experimental and field conditions were compiled from published literature. The most influential input parameter combinations were identified using the gamma test. Among 37 combinations involving the pier width to flow depth ratio (b/y), approach velocity to critical velocity ratio (V/Vc), critical Froude number (Frc), pier width to median sediment size ratio (b/d50) and sediment gradation (σg), the optimal combination was selected based on minimum gamma and Vratio values. The results indicate that the MARS model achieved superior performance, yielding a coefficient of determination (R2) of 0.942 and a mean absolute percentage error (MAPE) of 0.086, outperforming the ANN–PSO, ANFIS and M5 Tree models. The trained models were further validated using independent field datasets from Chemung, Honey Creek, Tanana and Clarks Fork Rivers and compared with ten existing scour prediction equations. The MARS and ANFIS models consistently showed higher accuracy, with R2 > 0.90 and MAPE < 33%. The findings demonstrate the superiority of ML approaches over empirical models and identify key parameters governing CWS, supporting reliable bridge scour assessment.
This study proposes advanced stacking ensemble machine learning approaches to predict soil Liquidity Index and Undrained Shear Strength and highlights the robust potential of ensemble modeling and tailored optimization in the field of geotechnical engineering.
Giovanni Spagnoli, Mohammadreza Mahmoudi, S. Shimobe et al.· E3S Web of Conferences· 0 citations
Accurate prediction of groundwater levels is a key concern for environmental monitoring and sustainable water resources management. Inspired by a Random Forest (RF) model trained with three popular hyperparameter optimizations—Particle Swarm Optimization (PSO), Simulated Annealing (SA), and Genetic Algorithm (GA)—this research proposes a new method for flattening groundwater level forecasting. The study dataset includes groundwater table (GWT) information and two independent variables, latitude and longitude, collected as a single-time measurement from 335 observation wells in Bangladesh’s Bogura district. The performance of models was measured using some metric form, such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R2, and Mean Absolute Error (MAE). The PSO-based optimization (PSO-RF) was more effective than GA and SA, observing the prediction ability, as reported in the study. The PSO-RF model demonstrated better generalization on the test data by achieving a Test MSE of 0.42561, a Test RMSE of 0.65239, a Test R² of 0.84093, and a Test MAE of 0.49997, which were lower than those from all used models. This paper demonstrates the significance of hyperparameter optimization in enhancing the performance of machine learning models for environmental predictions. 2D and 3D spatial GWT distribution maps were carried out with the derived data set, consequently established by ArcGIS software. The proposed PSO-RF model is promising for the prediction of groundwater levels in Bangladesh. It can be further extended to other regions for similar applications in environmental problem prediction and water resource management
This study evaluates the prediction of flowing bottom-hole pressure (FBHP) in dry gas wells using machine learning techniques, specifically Random Forest and Artificial Neural Network (ANN) models. Unlike earlier work based on PROSPER-generated synthetic data, this study utilizes a real field dataset of 206 samples obtained from the ProBHP repository, originally compiled by Govier and Fogarasi (1975) and Asheim (1986). The dataset comprises 10 input variables, including production rates, well depth, tubing size, temperatures, and wellhead pressure, with measured bottom-hole pressure (MBHP) as the target. Feature importance analysis identified well depth, oil rate, water rate, and wellhead pressure as the most influential parameters. The data were split into 80% training and 20% testing sets, with Z-score-based outlier removal reducing the training data slightly. The Random Forest model showed strong predictive performance, achieving a test R2 of 0.81, MAE of 93.73 psig, and RMSE of 123.39 psig, with a cross-validation R2 of 0.72 ± 0.11. In contrast, the ANN model performed poorly, with a test R2 of 0.05 and MAE of 209.55 psig. Overall, the results highlight the reliability of Random Forest for FBHP prediction using real field data, while also showing the limitations of a simple ANN model on small, noisy datasets. The identified key parameters provide useful insights for well performance monitoring and production optimization in dry gas systems.
Fred Akpososo, V. Aimikhe, D. Kalu et al.· SPE Nigeria Annual Internati...· 0 citations
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.
Shiwei Hu, Rong Hu, Hong Zhang et al.· SAE technical paper series· 0 citations
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
The proposed AFELA–machine learning framework provides a computationally efficient, reliable, and interpretable approach for tunnel stability assessment in sloping rock mass conditions.
A. Kumar, V. Chauhan, Aayush Kumar et al.· Transportation Infrastructur...· 1 citation