Analysis of Wall Jet Scouring Parameters Using Multiple Regression and Soft Computing Techniques

Title: Advanced Analysis Techniques for Wall Jet Scouring Study

In a recent study on wall jet scouring, researchers have utilized multiple linear regression analysis (MLRA) and non-linear multiple regression analysis (MNLRA) to analyze the impact of various parameters on maximum equilibrium scour depth, distance to scour depth from the end of rigid apron, height of dune, and distance to maximum height of dune crest. The study considered factors such as apron length, tail water level, densimetric Froude number, and median sediment size as influencing variables.

The results of the analysis revealed rising trends in all dependent parameters with apron length, indicating the dissipation of energy of the jet as it travels over the apron. Additionally, the study developed regression equations for each dependent parameter based on the input parameters.

Further analysis involved the application of machine learning approaches such as Artificial Neural Network-Particle Swarm Optimization (ANN-PSO) and Gene Expression Programming (GEP) to model the parameters. The ANN-PSO model showed promising results in predicting the dependent parameters accurately compared to other regression and soft computing techniques.

The study also included uncertainty and reliability analysis, which highlighted the predictive capabilities and robustness of the models. The ANN-PSO model demonstrated lower confidence intervals and higher reliability indices, indicating its consistency and reliability in predicting the parameters.

Overall, the advanced analysis techniques used in this study provide valuable insights into wall jet scouring behavior and offer a more accurate and reliable approach to predicting the parameters involved in the process.