A prediction model for finding the optimal laser parameters in additive manufacturing of NiTi shape memory alloy

• Published in: International journal of advanced manufacturing technology Vol. 105; no. 11; pp. 4691 - 4699
• Main Authors: Mehrpouya, Mehrshad, Gisario, Annamaria, Rahimzadeh, Atabak, Nematollahi, Mohammadreza, Baghbaderani, Keyvan Safaei, Elahinia, Mohammad
• Format: Journal Article
• Language: English
• Published: London Springer London 01-12-2019; Springer Nature B.V
• Subjects: Additive manufacturing; Alloys; Artificial neural networks; Austenite; Automotive parts; CAE) and Design; Computer-Aided Engineering (CAD; Engineering; Industrial and Production Engineering; Intermetallic compounds; Lasers; Martensite; Martensitic transformations; Mathematical models; Mechanical Engineering; Mechanical properties; Media Management; Nickel base alloys; Nickel titanides; Original Article; Parameters; Prediction models; Prototyping; Shape effects; Shape memory alloys; Strain; Superelasticity; Surgical implants; Three dimensional printing; Transformation temperature; Shape memory alloys; Additive manufacturing; Artificial neural network; NiTi; Modeling
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-019-04596-z

Shape memory alloys (SMAs) have been applied for various applications in the fields of aerospace, automotive, and medical. Nickel-titanium (NiTi) is the most well-known alloy among the others due to its outstanding functional characteristics including superelasticity (SE) and shape memory effect (SME). These particular properties are the result of the reversible martensite-to-austenite and austenite-to-martensite transformations. In recent years, additive manufacturing (AM) has provided a great opportunity for fabricating NiTi products with complex shapes. Many researchers have been investigating the AM process to set the optimal operational parameters, which can significantly affect the properties of the end-products. Indeed, the functional and mechanical behavior of printed NiTi parts can be tailored by controlling laser power, laser scan speed, and hatch spacing having them a crucial role in properties of 3D-printed parts. In particular, the effect of the input parameters can significantly alter the mechanical properties such as strain recovery rates and the transformation temperatures; therefore, using suitable parameter combination is of paramount importance. In this framework, the present study develops a prediction model based on artificial neural network (ANN) to generate a nonlinear map between inputs and outputs of the AM process. Accordingly, a prototyping tool for the AM process, also useful for dealing with the settings of the optimal operational parameters, will be built, tested, and validated.