2 publications

• Published Subscription Original Research 9 Sep 2026

  Machine Learning-Based Structure–Property Quantification of Advanced Polymer Composites

  By Ganesh P. Dawange, Bhushan S. Chaudhari, P. William, Atul A. Barhate, Sharad Ninu Kolte, Amit V. Mohod, Dipesh B. Pardeshi

  Abstract: Advanced polymer composites are widely used in high-performance engineering due to their superior mechanical and multifunctional properties. Accurate structure–property quantification is essential for efficient material design and reducing experimental costs. Existing Machine Learning (ML) approaches often exhibit limited predictive generalization due to inadequate feature discrimination and suboptimal hyperparameter tuning. To address these limitations, the proposed method enhances the ability to capture the complex nonlinear interactions among composite structural descriptors. The …

  Published in Journal of Polymer & Composites · Vol. 14, Issue 4, 2026 Read article →

• Published Subscription Original Research 9 Sep 2026

  Artificial Intelligence and Constitutive Modeling Equations for Predictive Design of High-Performance Polymer Composites

  By Harshal P. Varade, P. William, Amit V. Mohod, Dipesh B. Pardeshi, Atul A. Barhate, Sharad Ninu Kolte, Ganesh P. Dawange

  Abstract: Growing polymer composite applications demand accurate mechanical prediction, yet complex interactions and conventional constitutive models limit predictive capability and require extensive calibration. To report these challenges, this research recommends a combined Artificial Intelligence (AI) and constitutive modeling approach based on an Enhanced Tasmanian Devil Optimizer-tuned Residual Neural Network with Multilayer Perceptron (ETDO-ResNet-MLP) for the predictive design of high-performance polymer composites. The study uses a publicly available Polymer Composite Property Dataset …

  Published in Journal of Polymer & Composites · Vol. 14, Issue 4, 2026 Read article →