Artificial neural network modelling for asphalt concrete samples with boron waste modification


Keskin M., Karacasu M.

JOURNAL OF ENGINEERING RESEARCH, cilt.10, sa.4B, ss.26-45, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 10 Sayı: 4B
  • Basım Tarihi: 2022
  • Doi Numarası: 10.36909/jer.8124
  • Dergi Adı: JOURNAL OF ENGINEERING RESEARCH
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Arab World Research Source, Directory of Open Access Journals
  • Sayfa Sayıları: ss.26-45
  • Anahtar Kelimeler: Marshall test, Asphalt concrete, Boron waste, Artificial neural networks, Recycling, Sustainability, MIXTURES, PERFORMANCE, PREDICTION, ANN, ADSORPTION, LEACHATE
  • Eskişehir Osmangazi Üniversitesi Adresli: Evet

Özet

Civil engineering science has evolved into the 21st century with concepts of recycling and sustainability. Create sustainable habitats by evaluating waste materials in building materials is one of the most important goals of this century. This study aims to eliminate the boron waste dunes that have occurred and continue to occur in Turkey, which has the world's largest boron reserves by using in road materials. Solid boron wastes obtained from the field were crushed and added to asphalt samples in certain ratios, and the effect of Crushed Boron Waste (CBW) on asphalt samples was investigated. As a result of Marshall Design Method, it has been proved that boron wastes can be used in asphalt concrete within the specification limits. Besides, an artificial neural network (ANN) model was created for the evaluation of obtained data. As a result of Marshall Design Method, it has been proved that boron wastes can be used in asphalt concrete within the specification limits. Furthermore, regarding the examination of modelling and statistical analysis, the mechanical performance of asphalt concrete samples with and without CBW addition has been predicted in a noticeable manner. As a result of regression analysis, training and test sets r(2) values are reached 0.95-0.91 for stability and 0.91-0.87 for flow values. Finally, a simulation was prepared with the created model, and the effect of boron wastes on asphalt samples was examined in more detail.