dhnp90/HyperSmart

1 HyperSmart

HyperSmart is a scientific software tool designed for the automated or semi-automated selection of hyperelastic models and calibration of their material parameters based on experimental data.
It also centralizes experimental datasets from various scientific publications into a single, well-structured repository, facilitating comparison, reuse and expansion by researchers.

2 📖 About HyperSmart Software

HyperSmart assists scientists and engineers working with hyperelastic materials such as rubber, silicone, soft biological tissues, and foam.

Its main objectives are:

• Model Selection: Suggest the most suitable hyperelastic model for the given material and experimental data.
• Parameter Calibration: Estimate material parameters using numerical methods, including Enumeration and Bayesian Updating with Structural Reliability Methods (BUS).
• Data Repository: Aggregate experimental mechanical testing data from the literature into a centralized, standardized format (YAML).
• Educational Tool: Serve as a platform for teaching concepts in material modeling and data fitting.

The program provides a graphical user interface (GUI) developed in Python with Tkinter, enabling easy navigation between the experimental data repository, model library, and calibration tools.

3 🏗️ Software Framework

HyperSmart is structured around three core components:

1. Experimental Data Repository

• Stores data in YAML format for four main deformation modes: Uniaxial Tension, Biaxial Tension, Simple Shear, Pure Shear
• Includes metadata such as material class, subclass, source, and citations.
• Supports visualization of data and export for external analysis.

3. Hyperelastic Model Library

• Organizes isotropic, incompressible hyperelastic models into categories: invariant-based, stretch-based and Hookean-type.
• Stores model definitions, equations, parameters, and reference sources in YAML.
• Enables equation display and parameter selection in the GUI.

4. Numerical Methods

• Enumeration Method: Brute-force search across parameter ranges.
• BUS Method: Bayesian inference approach for parameter updating and model selection.
• Both methods are integrated with the repository and model library for seamless workflow.

4 🖥️ Technical Details

• Language: Python
• GUI Framework: Tkinter
• Data Format: YAML
• Numerical Libraries: NumPy, SciPy
• Plotting: Matplotlib
• Version Control: Git / GitHub