Labs researchers capture R&D 100 Awards

October 18, 2023 8:00 am Published by

Among the R&D 100 awards and nominations Sandia took part in this year, two awards featured innovations with implications for energy and manufacturing.

R&D World magazine honors inventors by identifying the 100 most technologically significant products and advancements each year and recognizing the winning innovators and their organizations. Winners are chosen from an international pool of submissions from universities, private corporations, and government labs.

Learn more about the winning technologies below.

PowerModelsONM: Optimizing Operations of Networked Microgrids for Resilience

Software/Services – Winner

Utilities can use PowerModelsONM to plan for networked microgrids to support rapid recovery during extreme-event-induced grid outages model. 

PowerModelsONM software optimizes networked microgrids for power restoration during blackouts and other extreme events. It is the only physics-based optimization software package featuring networked microgrids for modeling restoration of electric power distribution feeders.

Utilities can use PowerModelsONM to plan for networked microgrids to support rapid recovery during extreme-event-induced grid outages. Superior validation is achieved using utility data sets for software simulation and hardware-in-the-loop experiments. Sandia is a partner on PowerModelsONM with Los Alamos National Laboratory, National Renewable Energy Laboratory and National Rural Electric Cooperative Association.

Materials Data-Driven Design

Special Recognition: Market Disruptor / Services – Silver medal

This project enables manufacturers to account for the internal structure of materials when shaping and forming a part for the first time by leveraging a proprietary deep learning model.

MAD3 is an innovative software that leverages the power of machine learning to modernize the forming and stamping processes of sheet metals. It predicts the parameters that characterize the directional mechanical behavior of a metal alloy 1,000 times faster than existing solutions. As a result, the software significantly reduces expensive and time-consuming forming and stamping trials.

More explicitly, metal alloys such as aluminum or steel used in various manufacturing processes like stamping and forming exhibit directional strength and formability that cause the metal to distort. The reaction, called plastic anisotropy, determines whether the material is capable of being shaped to the desired component fit and finish, and whether it will withstand the applied performance load. As a result, accurate predictions of the metal’s plastic anisotropy are crucial in major manufacturing and supported by automotive and aerospace metal manufacturers as well as suppliers.

However, the cost of characterizing plastic anisotropy has skyrocketed because characterization requires specialized equipment and significant technical expertise. The software uses state-of-the-art data-driven and machine-learning techniques to first extract a unique fingerprint descriptor of the metal alloy’s internal structure, then subsequently uses these descriptors to predict the plastic anisotropy parameters in an accurate and efficient manner.

These anisotropy parameters can be used to perform forming and stamping simulations with unprecedented accuracy since they incorporate the effect of the polycrystalline grain structure.

See past years’ R&D 100 award-winning technologies on our website.