PhaseFlux

Try it Now!

Visit www.phaseflux.com to use PhaseFlux directly in your browser - no installation required!

Description

PhaseFlux is a Streamlit-based web application that predicts flow regimes in multiphase flow systems under adiabatic conditions. It utilizes a LightGBM model to provide predictions based on various input parameters related to fluid properties and flow conditions. The app also offers feature importance analysis using SHAP (SHapley Additive exPlanations) values.

Important Note

This app is for demonstration purposes only and should not be used for real-world applications. The predictions should be validated against experimental data or established correlations for critical applications.

Features

• Instant flow regime predictions based on user-input parameters
• Advanced feature contribution analysis using SHAP values

Online Usage

1. Navigate to www.phaseflux.com
2. Enter your flow parameters in the input fields
3. Click "Predict Flow Regime" to get instant results
4. Explore the SHAP analysis to understand prediction factors

Local Installation

If you prefer to run PhaseFlux locally, you have two options:

Option 1: Using Docker (Recommended)

Prerequisites

• Docker
• Docker Compose

Steps

1. Clone the repository:

   git clone https://github.com/benettia/phaseflux.git
   cd phaseflux
   

2. Build and run the Docker container:

   docker-compose up --build
   

3. Access the application at:

   http://localhost:8501
   

Option 2: Using Virtual Environment

Prerequisites

• Python 3.11+
• pip

Steps

1. Clone the repository:

   git clone https://github.com/benettia/phaseflux.git
   cd phaseflux
   

2. Create and activate a virtual environment:

   python -m venv .venv
   source .venv/bin/activate  # On Windows, use `.venv\Scripts\activate`
   

3. Install the required packages:

   pip install -r requirements.txt
   

4. Run the Streamlit app:

   streamlit run src/app.py
   

5. Access the application at:

   http://localhost:8501
   

Contributing

We welcome contributions to PhaseFlux! Whether it's bug fixes, feature additions, or documentation improvements, please feel free to:

1. Fork the repository
2. Create a feature branch
3. Submit a Pull Request

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

How to Cite

The model used for this app is from:

@misc{alma9963263908776,
  author = {Benetti, Alessandro},
  title = {Flow regime estimation in two-phase flows employing machine learning: Investigation of the most promising dimensionless feature set},
  year = {2021},
  keywords = {flow regime estimation, two-phase flow, machine learning, feature selection, algorithm optimization},
  language = {English}
}