Evolutionary Modeling of Inclined Dense Jets Discharged from Multiport Diffusers

Multiport diffusers are becoming popular for discharging wastewaters into the receiving water. In this work, an approach based on multigene genetic programming (MGGP) is presented and applied to model the mixing properties of inclined dense jets discharged from multiport diffusers. Explicit mathematical models were formulated for the nondimensional terminal rise height, the nondimensional impact distance, and the nondimensional impact dilution, which are the most important parameters for describing effluent mixing properties. The developed model can be employed to predict mixing parameters as functions of the densimetric Froude number, the nondimensional port spacing, and the nondimensional current speed. The models were trained and tested with experimental data for three different ambient flow conditions: stationary, coflowing, and counterflowing. Therefore, these models are generally valid for these different scenarios. Sample Pareto-optimal MGGP models were compared, and the best ones are identified. The single-gene genetic programming (SGGP) algorithm was also utilized to develop models for the same parameters. It was found that the best MGGP models outperformed the best SGGP models and the existing empirical formulations. A confidence analysis for the best MGGP models is also reported. On the basis of a detailed evaluation of the performances of the MGGP models, the MGGP technique was proved to be a useful tool in developing models for predicting the mixing properties of inclined dense discharges from multiport diffusers.