International Congress on Fundamental and Applied Sciences(ICFAS216), İstanbul, Turkey, 22 - 26 August 2016, pp.132-133
In recent years, artificial neural networks (ANN) that was introduced in the many areas has become a powerful tool for solving non–linear therefore it started to be applied in many different areas of materials research. In this study, dynamic mechanical properties of polypropylene/polyethylene terephthalate (PP/PET) polymer blends were investigated and ANN modeling of experimental results were reviewed. In the results of dynamic mechanical analysis (DMA), glass transition temperature increases with increasing frequency and an increase in the intensity of the peaks was observed. Storage modulus did not show a significant change between the rate of 0–20% PET but the storage modulus exhibited a higher increase between rate of 20–40% PET. Storage modulus showed a serious decline at PP+50% PET blends. The ANN technique with a feed–forward back propagation algorithm was used to examine glass transition temperature and storage modulus values of PP/PET blends. PET rate and temperature are used as inputs and storage modulus, tan delta, glass transition temperature are used as outputs for ANN modelling. ANN results and the experimental results were compared and the results were observed with sufficient accuracy.