TY - JOUR
T1 - Characterization and predictive modeling of thermally aged glass fiber reinforced plastic composites
AU - Rahman, Md Mijanur
AU - Rahman, M. Muzibur
N1 - Publisher Copyright:
© 2024 MIM Research Group. All rights reserved.
PY - 2024
Y1 - 2024
N2 - Glass fiber reinforced plastics (GFRP) are exposed to thermal aging in their widespread aerospace applications. Evaluating the effect of mechanical properties due to thermal aging has remained a challenge. An experimental investigation to characterize the thermal aging effects of glass fiber epoxy composites as well as the development of a predictive modeling is presented here. Tensile test samples have been thermally aged at 50°C, 100°C, 150°C and 200°C for 30 mins, 60 mins, 90 mins and 120 mins. At higher temperatures, the samples have shown a gradually increasing brown color while emitting a burning smell. The tensile test shows that the UTS value decreases as the thermal aging temperature increases. The predictive model has been prepared by combining image processing, regression analysis and two cascaded artificial neural networks (ANNs). The model reads the photographic image of the sample and uses the color change as an identifier. Cascaded ANNs estimate the thermal aging temperature and time from the image processing program. Finally, the ANN's output is forwarded to the developed regression equation to get the estimated UTS. The predictive model's estimated UTS shows an average accuracy of 97% when compared to the experimental results.
AB - Glass fiber reinforced plastics (GFRP) are exposed to thermal aging in their widespread aerospace applications. Evaluating the effect of mechanical properties due to thermal aging has remained a challenge. An experimental investigation to characterize the thermal aging effects of glass fiber epoxy composites as well as the development of a predictive modeling is presented here. Tensile test samples have been thermally aged at 50°C, 100°C, 150°C and 200°C for 30 mins, 60 mins, 90 mins and 120 mins. At higher temperatures, the samples have shown a gradually increasing brown color while emitting a burning smell. The tensile test shows that the UTS value decreases as the thermal aging temperature increases. The predictive model has been prepared by combining image processing, regression analysis and two cascaded artificial neural networks (ANNs). The model reads the photographic image of the sample and uses the color change as an identifier. Cascaded ANNs estimate the thermal aging temperature and time from the image processing program. Finally, the ANN's output is forwarded to the developed regression equation to get the estimated UTS. The predictive model's estimated UTS shows an average accuracy of 97% when compared to the experimental results.
KW - Artificial neural network
KW - Glass fiber reinforced plastic composite
KW - Material characterization
KW - Predictive modeling
KW - Thermal aging
UR - http://www.scopus.com/inward/record.url?scp=85187021863&partnerID=8YFLogxK
U2 - 10.17515/resm2023.24me0609rs
DO - 10.17515/resm2023.24me0609rs
M3 - Article
AN - SCOPUS:85187021863
SN - 2148-9807
VL - 10
SP - 305
EP - 330
JO - Research on Engineering Structures and Materials
JF - Research on Engineering Structures and Materials
IS - 1
ER -