Abstract
Near-infrared (NIR) spectroscopy has been implemented for on-line monitoring of resin and volatile content during the manufacture of carbon/phenolic resin prepreg cloth. The calibration models were generated using partial least square (PLS) regression to relate the on-line spectral data to the components' content. The models had determination coefficients (R2) of 92.92 for the resin content and of 92.11 for the volatile content. The root mean square errors of prediction (RMSEP) were 0.861% and 0.273% for the resin and volatile content, respectively. The statistical results of the paired t-test revealed that there was no significant difference between the NIR method and the standard analysis method. The resin content and the volatile content were continuously monitored by NIR in the production of the carbon prepreg. The study indicates that NIR spectroscopy is a good alternative for monitoring the quality of carbon prepreg.