Influence of processing parameters on the resin pocket geometry of smart composites embedded with Fiber Bragg Grating sensor and its fast prediction

Zizhao Zhao, Kaifu Zhang, Chuang Liu, Hui Cheng, Peng Zou, Mengfei Feng, Junjie Xiao, Biao Liang

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

Smart composites embedded with Fiber Bragg Grating (FBG) sensors are getting increasingly popular in aircraft structures due to their self-sensing and monitoring capability. However, a resin pocket around the FBG sensor would be formed in the manufacturing process, which strongly impacts the mechanical properties of the composite host. Therefore, the influence of primary processing parameters (FBG orientation angle, preforming temperature, and preforming pressure) on the formation of resin pockets was investigated in this work. It is found that there is a sinusoidal relation between the resin pocket geometry and the FBG orientation angle. A negative linear relation is noted between the resin pocket geometry and the preforming temperature and pressure. A prediction model based on the backpropagation neural network (BPNN) was established, which was able to quickly and accurately predict resin pocket geometry subjected to given processing parameters. This work provides useful guidance for the processing control of smart composites embedded with FBG sensors. Highlights: Resin pocket geometry was formed primarily in the preforming stage. Effect of processing parameters on resin pocket geometry was characterized. A BPNN model was established for the fast prediction of resin pocket geometry.

Original languageEnglish
Pages (from-to)6214-6225
Number of pages12
JournalPolymer Composites
Volume45
Issue number7
DOIs
StatePublished - 10 May 2024

Keywords

  • BP neural network
  • FBG sensor
  • resin pockets
  • smart composites

Fingerprint

Dive into the research topics of 'Influence of processing parameters on the resin pocket geometry of smart composites embedded with Fiber Bragg Grating sensor and its fast prediction'. Together they form a unique fingerprint.

Cite this