A fatigue life prediction approach for laser-directed energy deposition titanium alloys by using support vector regression based on pore-induced failures. (June 2022)

• Record Type: Journal Article
• Authors: Dang, Linwei; He, Xiaofan; Tang, Dingcheng; Li, Yuhai; Wang, Tianshuai
• Is Part Of: International journal of fatigue. Volume 159(2022)
• Publication Date: 2022-06
• Subjects: Fatigue life prediction -- Machine learning -- Pore -- Microstructure -- Additive manufacturing; Materials -- Fatigue -- Periodicals; Materials -- Fatigue; Periodicals; 620.1122
• DOI: 10.1016/j.ijfatigue.2022.106748
• Languages: English
• ISSNs: 0142-1123

Abstract

Highlights: The relation between pores and microstructure dominates fatigue life of laser directed energy deposition Ti-6.5Al-2Zr-Mo-V alloy. Proper variables were proposed to be stress intensity factor range and pore types determined by pore and microstructure. Fatigue life prediction model was developed and validated by using support vector regression (SVR) algorithm. Abstract: A support vector regression (SVR) algorithm was chosen in this study to develop a fatigue life prediction model by post-mortem fractography analysis. Models based on the SVR algorithm with different input variables were compared to identify optimized input variables according to errs and correlation coefficients. Variations were verified in a stress intensity factor range obtained by Murakami's approach and for pores types determined by the relationship between pore size and microstructure size. The results confirm the importance of considering the fatigue behavior of pores and microstructures during crack initiation in the fatigue life prediction for additive manufacturing (AM) metals.