Motivation: Composite curing is an energy-intensive and time-consuming process, with the Office of Scientific and Technical Information estimating an autoclave cure to expend 20-22 MJ/kg. Furthermore, the part temperature history must conform to specifications to minimize a variety of process-induced defects including resin decomposition, under-curing, porosity, dimensional violations, and micro-cracks. Currently, the input process parameters for composite curing are quite simple and far from optimal, since they comprise of simple step changes in temperature and pressures.

Approach: The goal of this project is to develop a novel multi-task optimization physics-informed multiple neural network (PIMNN) framework to improve the efficiency and reduction of defects in the curing process of composite aerostructures. 

Partnerships: This research is done in collaboration with the Georgia Institute of Technology (lead) and the Boeing Company.

Motivation: Friction stir processing is a promising method to recycle materials through a relatively feasible process. However, friction stir processes are fairly stochastic owing to the thermo-plastic interaction in powder metallurgy.  Hence, systematic process-material-microstructure relationships and novel solutions to improve the efficiency of friction stir processing are critical towards realizing a sustainable process.

Approach: This research aims to develop novel methods to control and refine grain structure in friction stir processing. Furthermore, this research aims to conduct rigorous systematic investigation of the underlying mechanics of the process to determine the efficiency of fabricating various metal matrix composites. 

Partnerships: This research is done in collaboration with Dr. Scott Wagner and Dr. William Emblom.

Motivation: Hybrid subtractive/additive processes are becoming more popular to leverage the precision and sustainability of machining and 3D printing processes, respectively. However, qualification methods for these complex systems are limited. Furthermore, process knowledge and transferrable frameworks that can be leveraged towards efficient subtractive/additive processes are unknown.

Approach: This research aims to develop and evaluate novel metrology methods for both hybrid processes and robotic-driven systems in industrial settings. This research also aims to investigate transferrable knowledge amongst hybrid systems for more efficient and sustainable hybrid systems.

Partners: This research is done in collaboration with Dr. Paul Sanders.

Motivation: Mechatronics is an emerging field that focuses on the development and practice of automation for industrial applications. A diverse workforce must be prepared to meet the growing job demand for Mechatronics employees for a sustainable workforce. However, Mechatronics requires highly experiential workforce development programs to provide the professional and technical skills that are readily transferrable to the industry. Such programs are lacking, and therefore the existing workforce cannot meet the current and future demand.

Approach: This project will prepare a diverse cohort of participants for the five pillars of Mechatronics: robotics, mechanics, electronics, cybersecurity, and artificial intelligence. Professional development activities will be conducted to prepare participants for their career and expand their network with industry collaborators. Furthermore, this project will foster a community among cohorts through professional mentoring and community-building activities to promote a sense of belonging.

Partnerships: This research is supported by the NSF ExLENT program with Aleksandr Sergeyev the Principal Investigator and in collaboration with Gogebic Community College and West Shore Community College.

Motivation: Electric motor design is becoming more complex to improve efficiency and reduce expended heat. However, existing methods to design electric motors are extremely complex and require multiple iterative approaches for optimization.

Approach: This research aims to investigate systematic methods to efficiently optimize the design of electric motors. Furthermore, this research also aims to incorporate manufacturability and economic considerations in the design of complex electric motors.