SBIR-STTR Award

In this Phase I STTR project, pursuant to the Materials Genome Initiative (MGI) and Integrated Computational Materials Engineering (ICME) interests, the proposed collaborative effort between WSU and Advratech will represent the first AM optimization framework of its kind, constructed entirely from experimental sensor data collected in-situ. Rather than using in-process data to inform limited "physics-based" FE models or detect single defects long after a build is complete, this framework will leverage correlations between in-situ data, input process parameters, and output AM build characteristics to construct a "physics-capturing" empirical black box that can be used to quantify AM process uncertainty, analyze sensitivities of AM component outputs to both input process parameters and in-process information, and ultimately, to optimize each layer of SLM builds in real-time. In essence, this project will provide a wrap-around software package and optimization tool that combines each mode of in-process data to inform real-time process parameter selection based on one or more desired physical property outputs. It will be designed on our SLM R&D test bed, be seamlessly applicable to any SLM system (e.g., Concept Laser LaserCUSING, etc.), and more generally applicable to any AM system (e.g., NASA's EBF3) used to construct aerospace components.

In this Phase II STTR project, the proposed collaborative effort between UTC, AFIT, and ULRF represents a crucial step forward for AM.  UTC’s unique AM optimization and process control framework, constructed entirely from experimental sensor data collected in-situ, will finally transfer technology from our SLM test bed system to state-of-the-art and commercial-grade systems, including a Concept Laser M2 Cusing and EOS M270 system.  UTC’s framework, which leverages a “physics-capturing” empirical black box built on correlations between in-situ data, input process parameters, output AM build characteristics, and machine variations will be used to quantify AM process uncertainty across these systems.  This Phase II project will show how seamlessly UTC’s technology can be integrated in to any SLM system to inform real-time output prediction for open loop (closed architecture) systems, and real-time process parameter selection and optimization for closed loop (open architecture) systems.