DEVELOPMENT OF A MACHINE LEARNING BASED MODEL FOR ADDITIVE MANUFACTURING TO ESTIMATE POROSITY EFFECTS ON STRESS DISTRIBUTION

Abstract

Additive manufacturing (AM) is an innovative manufacturing technique that can build complex and high value metal parts layer by layer using computerized 3D CAD models. Powder bed fusion (PBF) is one of the most used AM techniques. Aim of this project, with the help of machine leaning in additive manufacturing developed a machine learning based model for which we have to observe the effects of porosity model on the stress distribution in y-direction. Model simulation was done in ABAQUS software for which a part model was sketched in the sketch module, done all things and then created a porosity in the model in ABAQUS software, at some minimum porosity percentage a simulation was done then many other simulations was done for different porosity models and maximum porosity percentage for the model was 1% and then we made a data sheet of the required data from the result module then with the help of this data sheet produced a supervised learning model which will lead us to estimate the effect of different porosity model to the stress distribution.