Abstract
Structural health monitoring (SHM) plays a vital role in engineering by enhancing the safety and longevity of structures through the use of various sensors. This study presents a flexible polyvinylidene fluoride/carbon black (PVDF/CB) nanofiber sensor for SHM applications. The sensor was fabricated via electrospinning and demonstrated strong piezoelectric performance. Incorporation of CB effectively promoted the transformation of $\alpha $ -phase to the piezoelectric $\beta $ -phase by 6.1% compared to pure PVDF nanofibers. In addition, the sensitivity of the piezoelectric nanofibers was increased from 0.75 to 2.05 V/N due to the addition of CB. To further enhance SHM accuracy, a multilayer perceptron (MLP) neural network optimized with the adaptive moment estimation (Adam) algorithm was trained using simulated model frequency data. The proposed SHM system integrates the PVDF/CB nanofiber sensors, trained MPL model, and signal processing module for modal frequency extraction. This system enables real-time damage detection while the structure under investigation is subjected to vibration. The system achieved a minimum detection error of 2.15 mm, and a stable detection rate of 95% (relative standard deviation (RSD) = 3.6%). These findings suggest that this flexible PVDF/CB sensor offers promising potential for real-time SHM applications.
Funding Information
  • National Key Research and Development Program of China (2022YFE0103800)
  • National Natural Science Foundation of China (42174221)

This publication has 36 references indexed in Scilit: