Speaker
Description
Different aspects of quality assessments for Additive Manufacturing (AM) parts are founded on complicated phenomena governed by intrinsic multi-physical and multi-phase interactions. To investigate the dynamic and fatigue performances of AM parts, variety of techniques including new hybrid physical-data driven scheme have been investigated. Hybrid modelling is one of the new trends in analyses and design optimization of AM parts where data modelling techniques have increasingly been combined with physical and analytical modelling for dynamic and fatigue assessments. An effective use of appropriate modelling schemes for fatigue analyses of AM parts and their agility of dealing with sophisticated effects of the manufacturing processes (e.g., defects, pores…) on service-life of products would briefly be reviewed herein. With the introduction of hybrid physical-data driven modelling techniques for the new AM process technologies and digitization drive, the need for integration of smart data-driven and physical-based models has willingly raised in the current research work. Different hybrid and data-driven schemes which include data mining, data handling, data processing and data modelling have rigorously been employed by researchers and engineers for dynamic and fatigue life simulation of AM parts at industrial scale. Furthermore, attempts have been made to setup full integrated data-bases where the available measured\mined\simulated data which are based on material characteristics and process conditions can semantically be gathered within single data depository. In the research work herein, some aspects of hybrid fatigue life assessment scheme for AM parts have been reviewed and analytical and computational benefits are scrutinized. Additionally, the employment of an efficient artificial Intelligent (AI) and Machine Learning (ML) schemes for industrial AM parts are tersely presented.