Scholarlink

<p>This article provides an in-depth study of the PVDF/h-BN composite films that were fabricated using the Taguchi L9 design to find the best processing conditions for piezoelectric energy harvesting. The influence of three parameters h-BN loading, hot-press temperature, and pressing duration on the structural, mechanical dielectric and piezoelectric properties of the composition films was systematically investigated. Fourier transform infrared spectroscopy, X-ray diffraction, and scanning electron microscopy showed that not only was there a good dispersion of h-BN filler, but also significant improvement in crystallinity and <i>β</i>-phase fraction with increasing h-BN content. The optimized sample generated an open-circuit voltage of 3.9–4.5&#xa0;V under finger tapping, demonstrating significant potential for energy harvesting applications. Taguchi ANOVA analysis identified h-BN weight percentage as the most influential factor affecting all responses, while, temperature and pressing time had secondary effects. A machine-learning model (Ridge Regression) was developed to predict the key responses including tensile strength, crystallinity, <i>β</i>-phase fraction, dielectric constant, and output voltage. Validation using leave-one-out cross-validation, permutation testing, and multicollinearity analysis confirmed the model's robustness and predictive reliability, with <i>R</i><sup>2</sup> values of 0.957.The integration of DOE, structure–property correlation, and machine learning provides comprehensive guidance on processing structure-performance relationships, establishing PVDF/h-BN as promising material for flexible energy harvesting application.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Smart Prediction of Piezoelectric Response in PVDF/h-BN Composites: Combining Materials Science and Machine Learning

  • 期刊论文

  • 发表日期：

    2026年7月8日

查看全文
  • Mahesh Pratap Gotte,
  • P. S. Rama Sreekanth

摘要

This article provides an in-depth study of the PVDF/h-BN composite films that were fabricated using the Taguchi L9 design to find the best processing conditions for piezoelectric energy harvesting. The influence of three parameters h-BN loading, hot-press temperature, and pressing duration on the structural, mechanical dielectric and piezoelectric properties of the composition films was systematically investigated. Fourier transform infrared spectroscopy, X-ray diffraction, and scanning electron microscopy showed that not only was there a good dispersion of h-BN filler, but also significant improvement in crystallinity and β-phase fraction with increasing h-BN content. The optimized sample generated an open-circuit voltage of 3.9–4.5 V under finger tapping, demonstrating significant potential for energy harvesting applications. Taguchi ANOVA analysis identified h-BN weight percentage as the most influential factor affecting all responses, while, temperature and pressing time had secondary effects. A machine-learning model (Ridge Regression) was developed to predict the key responses including tensile strength, crystallinity, β-phase fraction, dielectric constant, and output voltage. Validation using leave-one-out cross-validation, permutation testing, and multicollinearity analysis confirmed the model's robustness and predictive reliability, with R2 values of 0.957.The integration of DOE, structure–property correlation, and machine learning provides comprehensive guidance on processing structure-performance relationships, establishing PVDF/h-BN as promising material for flexible energy harvesting application.