Integrating KPFM Characterisation, COMSOL Multiphysics Simulation and Physics-Informed cVAE for Multi-Polymer Triboelectric Nanogenerator Optimisation.
Saved in:
| Title: | Integrating KPFM Characterisation, COMSOL Multiphysics Simulation and Physics-Informed cVAE for Multi-Polymer Triboelectric Nanogenerator Optimisation. |
|---|---|
| Authors: | Rahul, T. Pavan1 (AUTHOR), Sreekanth, P. S. Rama1 (AUTHOR) happyshrikanth@gmail.com |
| Source: | Materials (1996-1944). May2026, Vol. 19 Issue 9, p1790. 26p. |
| Subjects: | Kelvin probe force microscopy, Simulation software, Dielectric materials, Surface charges, Nanogenerators, Energy harvesting, Mathematical optimization |
| Abstract: | Triboelectric nanogenerators (TENGs) offer a promising route for self-powered microscale energy harvesting, yet their design optimisation remains empirically challenging due to the complex interplay of material surface physics, device geometry and operating mode. In this work, we present an integrated framework that combines atomic force microscopy (AFM) characterisation, COMSOL Multiphysics 6.0 finite element simulation and physics-informed conditional variational autoencoder (cVAE) to predict and optimise TENG output performance. Four polymer dielectric materials, HDPE, LDPE, TPU, and PMMA, were characterised via Kelvin Probe Force microscopy (KPFM) for work function, surface potential and surface roughness. Surface charge density was calculated from measured KPFM potential using the parallel plate capacitor model and used as a boundary condition in COMSOL Multiphysics simulations for contact-separation and lateral sliding TENG mode for dielectric film thicknesses of 50 µm and 100 µm. The simulated open circuit voltage (Voc) and short circuit charge (Qsc) across gap distances up to 150 mm formed the training dataset for a cVAE model with eight physicochemical condition features. The trained model demonstrated strong reconstruction accuracy (R2 ≥ 0.94) and enables generative prediction across unseen design spaces. Results reveal that the LDPE/TPU pair at 50 µm thickness consistently achieves the highest electric outputs in both modes, and the sliding mode yields 25–30% higher voltages than the contact separation mode across all material pairs. This study provides a transferable data-efficient methodology for accelerating TENG material and geometry optimisation. [ABSTRACT FROM AUTHOR] |
| Copyright of Materials (1996-1944) is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Triboelectric nanogenerators (TENGs) offer a promising route for self-powered microscale energy harvesting, yet their design optimisation remains empirically challenging due to the complex interplay of material surface physics, device geometry and operating mode. In this work, we present an integrated framework that combines atomic force microscopy (AFM) characterisation, COMSOL Multiphysics 6.0 finite element simulation and physics-informed conditional variational autoencoder (cVAE) to predict and optimise TENG output performance. Four polymer dielectric materials, HDPE, LDPE, TPU, and PMMA, were characterised via Kelvin Probe Force microscopy (KPFM) for work function, surface potential and surface roughness. Surface charge density was calculated from measured KPFM potential using the parallel plate capacitor model and used as a boundary condition in COMSOL Multiphysics simulations for contact-separation and lateral sliding TENG mode for dielectric film thicknesses of 50 µm and 100 µm. The simulated open circuit voltage (Voc) and short circuit charge (Qsc) across gap distances up to 150 mm formed the training dataset for a cVAE model with eight physicochemical condition features. The trained model demonstrated strong reconstruction accuracy (R2 ≥ 0.94) and enables generative prediction across unseen design spaces. Results reveal that the LDPE/TPU pair at 50 µm thickness consistently achieves the highest electric outputs in both modes, and the sliding mode yields 25–30% higher voltages than the contact separation mode across all material pairs. This study provides a transferable data-efficient methodology for accelerating TENG material and geometry optimisation. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 19961944 |
| DOI: | 10.3390/ma19091790 |
