Modelling of the Tensile Properties of Calcium Carbonate Filled Polypropylene Composite Materials with Taguchi and Artificial Neural Networks


Creative Commons License

Şahin Y. , Şahin Ş. , İnal M. M.

IFAC-PapersOnLine, cilt.51, ss.282-286, 2018 (Diğer Kurumların Hakemli Dergileri) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 51
  • Basım Tarihi: 2018
  • Doi Numarası: 10.1016/j.ifacol.2018.11.302
  • Dergi Adı: IFAC-PapersOnLine
  • Sayfa Sayıları: ss.282-286

Özet

© 2016In this experimental study, Taguchi experiment design and Artificial Neural Networks methods were used to optimize the Tensile strength of polypropylene / calcium carbonate composite materials depending on their production parameters and filler ratios in different percentages by weight. The long-term strength (life or time) values of materials are usually obtained as a result of long-term tests also after production. For example it's known that tensile strength of polypropylene based material is decreased depending upon time after production. But changes to tensile properties, depending upon CC filler content of PPH/CC composite materials, after a long storage period which follows its production are still unclear. In this context, in this study, Tensile strength properties optimization of the calcium carbonate (CC) filled polypropylene homopolymer (PPH) composite, depending on the injection parameters is studied via Taguchi Experimental Design Method and Artificial Neural Networks (ANN). To examine the effect of injection parameters the injection pressure, the melt temperature and mold temperature impose conditions are used. CC-filled composite samples have been produced at different injection parameters by injection molding based on Taguchi experimental design method. Samples were stored for 12 months without daylight comfort conditions. Tensile tests for samples are made according to the ISO 527 standard. By this experimental study, parameter optimizations for Tensile strength properties of the PPH / CC composite materials have been revealed. At the second phase of the study, a fully connected feedforward Artificial Neural Network is utilized for modeling the Tensile strength properties of CC filled PP composite materials.