Surrogate for Elastomer Membrane Behavior under Radial loading and Actuation — a surrogate framework for dielectric elastomer membrane instability.
This repository accompanies the paper On the Instability of Inflated Dielectric Elastomer Membranes: Energy-Based Analysis and a Surrogate Modeling Framework by Kumar, Mathur, and DasGupta. It provides two neural-surrogate packages that predict the equilibrium state and stability of an inflated circular dielectric elastomer membrane under combined mechanical pressure and electrical actuation.
- SEMBRA — voltage-controlled actuation. Use this when the membrane is connected to a constant voltage source.
- SEMBRA_CC — charge-controlled actuation. Use this when the membrane is initially charged and then disconnected from the power supply.
Both packages share an identical architecture and methodology, differing only in the underlying electrostatic energy formulation. Each is self-contained and can be installed and used independently.
Choose the package corresponding to your loading scenario:
- For voltage-controlled analysis, see SEMBRA's README.
- For charge-controlled analysis, see SEMBRA_CC's README.
Each package provides a predict(alpha, alpha_p, p) function and a
Jupyter notebook for interactive use, along with full inference and
validation pipelines.
- Python 3.11 or higher
See the individual package READMEs for dependency details.
If you use this software, please cite the archived release:
Kumar, N., Mathur, A., & DasGupta, A. (2026). SEMBRA Framework: Surrogate models for instability of dielectric elastomer membranes (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.20723832
The accompanying paper is in preparation; its citation will be added here upon publication.
MIT — see SEMBRA/LICENSE and SEMBRA_CC/LICENSE.