Assessment of Machine Learning Models for Predicting behavior of Laterally Confined Buoyant Jets

Buoyant jets are employed to discharge saline or hot wastewater into the ambient environment by various industrial plants which cloud pose a threat to marine environments and coastal ecosystems. Therefore, understanding the behavior of these jets is crucial for managing their adverse environmental impacts. To characterize the behavior of a confined buoyant jet, a three-dimensional model was employed to solve the transport of mass and momentum using the Computational Fluid Dynamics (CFD) tool OpenFOAM. Various flow conditions were defined by changing the Reynolds number (Re) and densiometric Froude number (Fr) to examine the flow characteristics of vertical buoyant jets subjected to lateral confinement.

The generated datasets were then utilized to train and test different machine-learning algorithms. The machine learning models were developed to predict the flow characteristics based on flow key parameters, including the geometrical parameters. Various machine learning algorithms, including support vector machines (SVM), Extreme Learning Machine (ELM), and Adaptive Neuro-Fuzzy Inference System (ANFIS), were trained and tested using these datasets. The performance of these models was assessed using multiple statistical metrics, and the results were benchmarked against those obtained from a multigene genetic-programming (MGGP) model and an existing regression-based empirical equation.

This study demonstrates the potential of machine learning algorithms to accurately predict the behavior of laterally confined vertical buoyant jets. The findings suggest that machine learning can serve as a reliable and efficient tool in environmental engineering and impact assessments, providing a fast and viable alternative to traditional CFD methods. These models offer a powerful solution for accurately predicting initial dilution properties, supporting the design and assessment of environmental engineering systems.