Formability behavior of AH-32 shipbuilding steel strengthened by friction stir process


SEKBAN D. M., UZUN YAYLACI E., Özdemir M. E., ÖZTÜRK Ş., YAYLACI M., Panda S. K.

Theoretical and Applied Fracture Mechanics, vol.132, 2024 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 132
  • Publication Date: 2024
  • Doi Number: 10.1016/j.tafmec.2024.104485
  • Journal Name: Theoretical and Applied Fracture Mechanics
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Aquatic Science & Fisheries Abstracts (ASFA), Communication Abstracts, Compendex, INSPEC, Metadex, DIALNET, Civil Engineering Abstracts
  • Keywords: AH-32 steel, Artificial neural network, Finite element analysis approach, Friction stir process
  • Recep Tayyip Erdoğan University Affiliated: Yes

Abstract

Ships are built by bringing materials into various forms and then joining them. Although materials such as wood, composite materials, polyethylene, and aluminum alloys are used in shipbuilding, it is known that commercial ships are generally produced from steel materials. Relatively strong steels used in shipbuilding couse problems such as reduced formability and weldability due to their chemical content. In this context, increasing the strength of such steels without changing their chemical composition is extremely important. Although many methods are used to increase the mechanical properties of steels without changing their chemical composition, the friction stir process (FSP) comes to the fore in terms of the increased rate in strength, reasonable decrease in elongation values, and its application to plate-type materials. In this study, FSP was applied to AH-32 steel used in shipbuilding, and the strength and formability values of the steel after the process were examined comparatively with mechanical tests, finite element method, and artificial neural network model. The results determined that steel's strength improved significantly after the FSP, while the formability behavior decreased very limitedly. The results also showed that the mechanical test results, the model results created with finite elements, and the artificial neural networks are highly consistent.