Analysis of Breakthrough Curves for Pollutant Transport in Laboratory Flume Using Soft Computing Models

This study investigates pollutant transport in gravel river beds through laboratory experiments and numerical simulations. Sodium chloride was used as a tracer to simulate contaminant movement under varying flow conditions and initial concentrations. The results demonstrate that pollutant transport is governed by advection, dispersion, and mixing processes. Soft computing models, including Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference Systems (ANFIS), and Support Vector Regression (SVR), were employed to predict breakthrough curves. ANFIS exhibited the best performance in capturing the complex dynamics of pollutant transport. The study highlights the influence of initial concentration on dispersion coefficients and the importance of considering density-induced mixing effects. The findings provide valuable insights into the behavior of pollutants in gravel river beds, aiding in the development of effective strategies for water quality management and environmental protection.

Keywords:

Gravel river beds , Pollutant transport , Hyporheic zone , Soft computing models