Simulation Data for Proton Transport in Nafion Membranes with AI Predictions

Xingyu Zhang's dataset, published on figshare in April 2026, provides simulation data for understanding chemical degradation in Nafion membranes used in fuel cells. The 2.6 KB CSV file contains data used to train interpretable machine learning and deep learning models for predicting proton transport properties. The models leverage a high-fidelity, multiscale simulation dataset to link nanostructure, hydration, temperature, and degradation.

Use Cases

• Training interpretable ML/DL models to predict proton conduction properties based on nanostructure data.
• Applying gradient-weighted class activation mapping to identify influential spatial regions within membrane nanostructures.
• Quantifying the impact of nanostructural and environmental factors using Shapley-value analysis from random forest models.
• Constructing mechanism maps to link hydration, temperature, and degradation with distinct conduction regimes.
• Enabling degradation-aware optimization and inverse design of proton exchange membrane fuel cells.

Strengths

• Data is derived from a described high-fidelity, multiscale simulation framework.
• The dataset is licensed under CC-BY-NC-4.0, permitting non-commercial reuse with attribution.
• The work employs multiple AI frameworks (3D CNN, random forest) for multimodal analysis.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 2.6 KB, indicating limited scope or highly summarized data.

Provenance

Source

Xingyu Zhang via figshare

Collection Method

Generated from high-fidelity, multiscale simulations of membrane nanostructure.

Time Range

The simulation data likely represents a snapshot of modeled conditions; no specific temporal range is stated.

Freshness

Last updated 2026-04-23 15:35:16; freshness should be verified.

Geography

Not applicable; the data is from computational materials science simulations.