Journal Home > Volume 1 , Issue 4
iEnergy 2022, 1(4): 463-470 https://doi.org/10.23919/IEN.2022.0049
Article | Open Access | Issue | Published: 20 December 2022

Prediction on the relative permittivity of energy storage composite dielectrics using convolutional neural networks: A fast and accurate alternative to finite-element method

Show Author's Information Shao-Long Zhong,  Di-Fan Liu,  Lei Huang,  Yong-Xin Zhang,  Qi Dong,  Zhi-Min Dang( )
State Key Laboratory of Power System, Department of Electrical Engineering, Tsinghua University, Beijing 100084, China
Keywords:
convolutional neural networks, finite element method, Relative permittivity, nanocomposite dielectrics, prediction accuracy
Cite this article:
Zhong S-L, Liu D-F, Huang L, et al. Prediction on the relative permittivity of energy storage composite dielectrics using convolutional neural networks: A fast and accurate alternative to finite-element method. iEnergy, 2022, 1(4): 463-470. https://doi.org/10.23919/IEN.2022.0049
Download citation
EndNote(RIS)
BibTeX

558

Views

48

Downloads

Citations

2

Crossref

N/A

WoS

0

Scopus

N/A

CSCD

Abstract Full text About this article

The relative permittivity is one of the essential parameters determines the physical polarization behaviors of the nanocomposite dielectrics in many applications, particularly for capacitive energy storage. Predicting the relative permittivity of particle/polymer nanocomposites from the microstructure is of great significance. However, the classical effective medium theory and physics-based numerical calculation represented by finite element method are time-consuming and cumbersome for complex structures and nonlinear problem. The work explores a novel architecture combining the convolutional neural network (ConvNet) and finite element method (FEM) to predict the relative permittivity of nanocomposite dielectrics with incorporated barium titanite (BT) particles in polyvinylidene fluoride (PVDF) matrix. The ConvNet was trained and evaluated on big datasets with 14266 training data and 3514 testing data generated form a programmatic algorithm. Through numerical experiments, we demonstrate that the trained network can efficiently provide an accurate agreement between the ConvNet model and FEM by virtue of the significant evaluation metrics R2, which reaches as high as 0.9783 and 0.9375 on training and testing data, respectively. The strong universality of the presented method allows for an extension to fast and accurately predict other properties of the nanocomposite dielectrics.


menu
Abstract
Full text
Outline
About this article

Prediction on the relative permittivity of energy storage composite dielectrics using convolutional neural networks: A fast and accurate alternative to finite-element method

Show Author's information Shao-Long Zhong, Di-Fan Liu, Lei Huang, Yong-Xin Zhang, Qi Dong, Zhi-Min Dang( )
State Key Laboratory of Power System, Department of Electrical Engineering, Tsinghua University, Beijing 100084, China

Abstract

The relative permittivity is one of the essential parameters determines the physical polarization behaviors of the nanocomposite dielectrics in many applications, particularly for capacitive energy storage. Predicting the relative permittivity of particle/polymer nanocomposites from the microstructure is of great significance. However, the classical effective medium theory and physics-based numerical calculation represented by finite element method are time-consuming and cumbersome for complex structures and nonlinear problem. The work explores a novel architecture combining the convolutional neural network (ConvNet) and finite element method (FEM) to predict the relative permittivity of nanocomposite dielectrics with incorporated barium titanite (BT) particles in polyvinylidene fluoride (PVDF) matrix. The ConvNet was trained and evaluated on big datasets with 14266 training data and 3514 testing data generated form a programmatic algorithm. Through numerical experiments, we demonstrate that the trained network can efficiently provide an accurate agreement between the ConvNet model and FEM by virtue of the significant evaluation metrics R2, which reaches as high as 0.9783 and 0.9375 on training and testing data, respectively. The strong universality of the presented method allows for an extension to fast and accurately predict other properties of the nanocomposite dielectrics.

Keywords: convolutional neural networks, finite element method, Relative permittivity, nanocomposite dielectrics, prediction accuracy

References(39)

[1]

Feng, Q. K., Zhong, S. L., Pei, J. Y., Zhao, Y., Zhang, D. L., Liu, D. F., Zhang, Y. X., Dang, Z. M. (2022). Recent progress and future prospects on all-organic polymer dielectrics for energy storage capacitors. Chemical Reviews, 122: 3820–3878.
DOI Google Scholar
[2]

Zhong, S. L., Dang, Z. M., Zhou, W. Y., Cai, H. W. (2018). Past and future on nanodielectrics. IET Nanodielectrics, 1: 41–47.
DOI Google Scholar
[3]

Yin, L. J., Zhao, Y., Zhu, J., Yang, M. H., Zhao, H. C., Pei, J. Y., Zhong, S. L., Dang, Z. M. (2021). Soft, tough, and fast polyacrylate dielectric elastomer for non-magnetic motor. Nature Communications, 12: 4517.
DOI Google Scholar
[4]

Dong, W. Z., Zhao, Y., Yin, L. J., Dang, Z. M. (2022). Fabrication and actuation characterisation of a new UV curing acrylic dielectric elastomer. IET Nanodielectrics, 5: 104–111.
DOI Google Scholar
[5]

Kim, Y. J., Lee, J., Park, S., Park, C., Park, C., Choi, H. J. (2017). Effect of the relative permittivity of oxides on the performance of triboelectric nanogenerators. RSC Advances, 7: 49368–49373.
DOI Google Scholar
[6]

Yildirim, A., Grant, J. C., Es-haghi, S. S., Lee, W., Maruthamuthu, M. K., Verma, M., Sutherland, J. W., Cakmak, M. (2020). Roll-to-roll (R2R) production of large-area high-performance piezoelectric films based on vertically aligned nanocolumn forests. Advanced Materials Technologies, 5: 2000553.
DOI Google Scholar
[7]

Fan, Y. J., Shen, Z. H., Zhou, X. C., Dan, Z. K., Zhou, L., Ren, W. B., Tang, T. X., Bao, S. Y., Nan, C. W., Shen, Y. (2022). Highly sensitive strain sensor from topological-structure modulated dielectric elastic nanocomposites. Advanced Materials Technologies, 7: 2101190.
DOI Google Scholar
[8]

Grill, A., Gates, S. M., Ryan, T. E., Nguyen, S. V., Priyadarshini, D. (2014). Progress in the development and understanding of advanced low k and ultralow k dielectrics for very large-scale integrated interconnects—State of the art. Applied Physics Reviews, 1: 011306.
DOI Google Scholar
[9]

Cheng, R., Wang, Y. F., Men, R. J., Lei, Z. P., Song, J. C., Li, Y. Y., Guo, M. Q. (2022). High-energy-density polymer dielectrics via compositional and structural tailoring for electrical energy storage. iScience, 25: 104837.
DOI Google Scholar
[10]

Ji, M., Min, D., Wu, Q., Mi, R., Liu, W., Li, S., Qin, S., Zhu, S. (2022). Significantly improved energy storage density of polypropylene nanocomposites via macroscopic and mesoscopic structure designs. Frontiers in Electronic Materials, 2: 904405.
DOI Google Scholar
[11]

Wang, S., Li, Q. (2018). Design, synthesis and processing of PVDF-based dielectric polymers. IET Nanodielectrics, 1: 80–91.
DOI Google Scholar
[12]

Sun, W. D., Lu, X. J., Jiang, J. Y., Zhang, X., Hu, P. H., Li, M., Lin, Y. H., Nan, C. W., Shen, Y. (2017). Dielectric and energy storage performances of polyimide/BaTiO3 nanocomposites at elevated temperatures. Journal of Applied Physics, 121: 244101.
DOI Google Scholar
[13]

Hao, Y. N., Wang, X. H., Bi, K., Zhang, J. M., Huang, Y. H., Wu, L. W., Zhao, P. Y., Xu, K., Lei, M., Li, L. T. (2017). Significantly enhanced energy storage performance promoted by ultimate sized ferroelectric BaTiO3 fillers in nanocomposite films. Nano Energy, 31: 49–56.
DOI Google Scholar
[14]

Yu, K., Wang, H., Zhou, Y. C., Bai, Y. Y., Niu, Y. J. (2013). Enhanced dielectric properties of BaTiO3/poly(vinylidene fluoride) nanocomposites for energy storage applications. Journal of Applied Physics, 113: 034105.
DOI Google Scholar
[15]

Li, Q., Chen, L., Gadinski, M. R., Zhang, S. H., Zhang, G. Z., Li, H. U., Iagodkine, E., Haque, A. M., Chen, L. Q., Jackson, T. N., et al. (2015). Flexible high-temperature dielectric materials from polymer nanocomposites. Nature, 523: 576–579.
DOI Google Scholar
[16]

Zhou, Y., Hu, J., Dang, B., He, J. L. (2017). Effect of different nanoparticles on tuning electrical properties of polypropylene nanocomposites. IEEE Transactions on Dielectrics and Electrical Insulation, 24: 1380–1389.
DOI Google Scholar
[17]

Zakri, T., Laurent, J. P., Vauclin, M. (1998). Theoretical evidence for `Lichtenecker’s mixture formulae’ based on the effective medium theory. Journal of Physics D: Applied Physics, 31: 1589–1594.
DOI Google Scholar
[18]

Sihvola, A. (2000). Mixing rules with complex dielectric coefficients. Subsurface Sensing Technologies and Applications, 1: 393–415.
DOI Google Scholar
[19]
Wang, J. C., Shen, Z. H., Jiang, J. Y., Wang, J., Zhang, X., Shen, J., Shen, Y., Chen, W., Chen, L. Q., Nan, C. W. (2022) High-throughput finite-element design of dielectric composites for high-frequency copper clad laminates. Composites Science and Technology, 225: 109517.
DOI
[20]

Zhong, S. L., Dang, Z. M., Zha, J. W. (2018). Prediction on effective permittivity of 0–3 connectivity particle/polymer composites at low concentration with finite element method. IEEE Transactions on Dielectrics and Electrical Insulation, 25: 2122–2128.
DOI Google Scholar
[21]

Cai, Z. M., Wang, X. H., Luo, B. C., Hong, W., Wu, L. W., Li, L. T. (2017). Nanocomposites with enhanced dielectric permittivity and breakdown strength by microstructure design of nanofillers. Composites Science and Technology, 151: 109–114.
DOI Google Scholar
[22]

Cheng, Y. H., Chen, X. L., Wu, K., Wu, S. N., Chen, Y., Meng, Y. M. (2008). Modeling and simulation for effective permittivity of two-phase disordered composites. Journal of Applied Physics, 103: 034111.
DOI Google Scholar
[23]

Wang, Z. P., Nelson, J. K., Hillborg, H., Zhao, S., Schadler, L. S. (2013). Dielectric constant and breakdown strength of polymer composites with high aspect ratio fillers studied by finite element models. Composites Science and Technology, 76: 29–36.
DOI Google Scholar
[24]

Chen, C. T., Gu, G. X. (2019). Machine learning for composite materials. MRS Communications, 9: 556–566.
DOI Google Scholar
[25]

Mannodi-Kanakkithodi, A., Pilania, G., Huan, T. D., Lookman, T., Ramprasad, R. (2016). Machine learning strategy for accelerated design of polymer dielectrics. Scientific Reports, 6: 20952.
DOI Google Scholar
[26]

Shen, Z. H., Liu, H. X., Shen, Y., Hu, J. M., Chen, L. Q., Nan, C. W. (2022). Machine learning in energy storage materials. Interdisciplinary Materials, 1: 175–195.
DOI Google Scholar
[27]

Shen, Z. H., Bao, Z. W., Cheng, X. X., Li, B. W., Liu, H. X., Shen, Y., Chen, L. Q., Li, X. G., Nan, C. W. (2021). Designing polymer nanocomposites with high energy density using machine learning. Npj Computational Materials, 7: 110.
DOI Google Scholar
[28]

Lin, Y., Guan, Z. D. (2021). The use of machine learning for the prediction of the uniformity of the degree of cure of a composite in an autoclave. Aerospace, 8: 130.
DOI Google Scholar
[29]

Choudhary, K., DeCost, B., Chen, C., Jain, A., Tavazza, F., Cohn, R., Park, C. W., Choudhary, A., Agrawal, A., Billinge, S. J. L., et al. (2022). Recent advances and applications of deep learning methods in materials science. Npj Computational Materials, 8: 59.
DOI Google Scholar
[30]

Ejaz, F., Hwang, L. K., Son, J., Kim, J. S., Lee, D. S., Kwon, B. (2022). Convolutional neural networks for approximating electrical and thermal conductivities of Cu-CNT composites. Scientific Reports, 12: 13614.
DOI Google Scholar
[31]

Bhutada, A., Kumar, S., Gunasegaram, D., Alankar, A. (2021). Machine learning based methods for obtaining correlations between microstructures and thermal stresses. Metals, 11: 1167.
DOI Google Scholar
[32]

Abueidda, D. W., Almasri, M., Ammourah, R., Ravaioli, U., Jasiuk, I. M., Sobh, N. A. (2019). Prediction and optimization of mechanical properties of composites using convolutional neural networks. Composite Structures, 227: 111264.
DOI Google Scholar
[33]

Ye, S., Li, B., Li, Q. Y., Zhao, H. P., Feng, X. Q. (2019). Deep neural network method for predicting the mechanical properties of composites. Applied Physics Letters, 115: 161901.
DOI Google Scholar
[34]

Barber, P., Balasubramanian, S., Anguchamy, Y., Gong, S. S., Wibowo, A., Gao, H. S., Ploehn, H., Zur Loye, H. C. (2009). Polymer composite and nanocomposite dielectric materials for pulse power energy storage. Materials, 2: 1697–1733.
DOI Google Scholar
[35]

Rong, Q. Y., Wei, H., Huang, X. Y., Bao, H. (2019). Predicting the effective thermal conductivity of composites from cross sections images using deep learning methods. Composites Science and Technology, 184: 107861.
DOI Google Scholar
[36]

Li, X. L., Zhang, Y. C., Zhao, H., Burkhart, C., Brinson, L. C., Chen, W. (2018). A transfer learning approach for microstructure reconstruction and structure-property predictions. Scientific Reports, 8: 13461.
DOI Google Scholar
[37]

Xu, Y. J., Weng, H. B., Ju, X. Z., Ruan, H. S., Chen, J. J., Nan, C. Y., Guo, J., Liang, L. H. (2021). A method for predicting mechanical properties of composite microstructure with reduced dataset based on transfer learning. Composite Structures, 275: 114444.
DOI Google Scholar
[38]

He, K. M., Zhang, X. Y., Ren, S. Q., Sun, J. (2015). Deep residual learning for image recognition. arXiv preprint: 1512.03385v1.
Google Scholar
[39]

Wang, W. H., Wang, H., Zhou, J. N., Fan, H. L., Liu, X. (2021). Machine learning prediction of mechanical properties of braided-textile reinforced tubular structures. Materials & Design, 212: 110181.
DOI Google Scholar
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 10 November 2022
Revised: 16 November 2022
Accepted: 21 November 2022
Published: 20 December 2022
Issue date: December 2022

Copyright

© The author(s)

Acknowledgements

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 52107018 and 51937007) and National Key Research and Development Program of China (No. 2021YFB2401502).

Rights and permissions

Copyright: by the author(s). The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

Return