Montana Flume Aeration Performance Evaluation with Machine Learning Models

Montana flume is derived from Parshall by eliminating diverging part and throat. The mass transfer of oxygen the atmosphere into water known as aeration. dissolved (D.O.) concentration in body determines quality. experiment was performed on six different flumes fixed a tilting prismatic rectangular channel. Experimental observations were used to develop classical machine learning models predict aeration efficiency. developed are namely multi nonlinear regression (MNLR), adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN). tested, results show that all these three perform very well. However, ANN gives better than other it has highest cc lowest rmse values. According sensitivity analysis results, Reynolds number (Re) most crucial input element determining efficiency case dimensionless datasets. discharge per unit width (q) found be relative significance dimensional