Predicting temporal clear water scour depth around bridge piers with XGBoost and SVM–PSO approaches

The phenomenon of scouring around a bridge pier involves the removal of sediment from the riverbed and banks due to water flow, particularly accelerated flow caused by obstruction from the bridge pier. This paper employs eXtreme Gradient Boosting (XGBoost) and support vector machine (SVM)–particle swarm optimization (PSO) machine learning (ML) approaches to model the temporal local scour depth around bridge piers under clear water scouring (CWS) conditions. CWS datasets, incorporating bridge pier geometry, flow characteristics, and sediment properties, are collected from existing literature. Five non-dimensional influencing parameters, such as the ratio of pier width to flow depth (b/y), the ratio of approach mean velocity to critical velocity (V/V_c), the ratio of mean sediment size to pier width (d₅₀/b), Froude number (F_r), and standard deviation of sediment (σ_g ), are chosen as input parameters. XGBoost and SVM–PSO models demonstrate superior predictive capabilities for clear water scour depth, achieving coefficient of determination (R²) values exceeding 0.90 and mean absolute percentage error and root mean square error values less than 17.07 and 0.0341, respectively. Comparison with four previous empirical models based on statistical indices reveals that the proposed XGBoost model outperforms SVM–PSO and empirical models in predicting CWS depth, so the XGBoost model is recommended for estimating CWS depth under varying temporal conditions within the specified dataset range.

• A wide range of CWS datasets under temporal conditions is collected to develop the model.

• The best combination of input parameters affecting the clear water scour depth is selected using the gamma test.

• A model is built to predict the CWS depth using the eXtreme Gradient Boosting (XGBoost) and support vector machine (SVM)–particle swarm optimization (PSO) machine learning (ML) approaches.

• The XGBoost ML approach is recommended based on the statical indices, achieving more than 90% accuracy.