amirhossein4096 / phononicbgrpredictdesign Goto Github PK

This repository contains a MATLAB application for predicting the bandgap ratio and performing inverse design of phononic crystals using a trained neural network. The tool leverages advanced machine learning techniques to optimize and tailor phononic crystal structures for specific bandgap properties. Key features include:

• Bandgap Ratio Prediction: Accurately predicts the bandgap ratio of phononic crystals using a trained neural network model.
• Inverse Design: Facilitates the design of phononic crystals with desired bandgap properties through an inverse design approach.
• User-Friendly Interface: MATLAB-based graphical interface for ease of use and accessibility.

1. Prediction Tab:

   • Adjust values of e/a, r/a, and h/a using sliders.
   • Instantly view the predicted bandgap ratio in a text box.
   • Automatically update the 3D view of the phononic crystal unit cell within the app.

2. Inverse Design Tab:

   • Set boundary ranges for e/a, d/a, and h/a using range sliders.
   • Input the desired bandgap ratio.
   • Click "Compute" to perform inverse design with minimal optimization using a trained neural network and particle swarm optimization.

1. Clone the Repository:

   git clone https://github.com/Amirhossein4096/PhononicBGRPredictDesign.git

2. Install Dependencies:

   • Ensure you have MATLAB installed.
   • Additional required toolboxes: Neural Network Toolbox, etc.

3. Run the Application:

   • Open MATLAB.
   • Navigate to the cloned repository directory.
   • Run the main script: app_main.m.

• Prediction Tab:

  • Use the sliders to set the values of e/a, r/a, and h/a.
  • View the predicted bandgap ratio and 3D unit cell visualization.

• Inverse Design Tab:

  • Set boundary ranges for e/a, d/a, and h/a.
  • Input the desired bandgap ratio.
  • Click "Compute" to start the inverse design process.

This project is licensed under the MIT License - see the LICENSE file for details.

For any questions or suggestions, please contact [email protected].

A MATLAB app for predicting bandgap ratios and inverse design of phononic crystals using a trained neural network.

Phononic Bandgap Predictor and Inverse Design App using Neural Networks