A Novel Framework of Developing a Predictive Model for Powder Bed Fusion Process.

Area of Science:

• Materials Science and Engineering
• Additive Manufacturing
• Computational Modeling

Background:

• Powder bed fusion (PBF) is a versatile metal additive manufacturing technique.
• Current PBF research often lacks a systematic approach for modeling numerous parameters simultaneously.
• Optimizing PBF requires quantitative methods to link process parameters to material properties and final part quality.

Purpose of the Study:

• To develop a predictive modeling framework for PBF of 316L stainless steel.
• To establish correlations between laser specifications, volumetric energy density, and mechanical properties.
• To enable intelligent selection of process parameters for desired part performance.

Main Methods:

• A two-phase framework for studying PBF process parameters.
• Development of predictive models using support vector regression, random forest regression, and neural networks.
• Analysis of correlations between laser parameters, volumetric energy density, and material properties.

Main Results:

• Achieved high part density up to 99.31% with optimized parameters.
• Demonstrated predictive models with approximately 10% error.
• Identified an optimal volumetric energy density range for enhanced mechanical properties.

Conclusions:

• The proposed framework and predictive models facilitate efficient PBF process optimization.
• Intelligent parameter selection leads to improved density, microstructure uniformity, hardness, and impact strength.
• This approach enables the production of high-quality 316L stainless steel parts meeting specific performance requirements.