Machine learning helps build high-performance thermoelectric devices
Aerospace and mechanical engineers at the University of Notre Dame have developed a machine-learning assisted superfast new way to make high-performance, energy-saving thermoelectric devices. Yaniliang Zhang, associate professor of aerospace and mechanical engineering at the University of Notre Dame, and colleagues Alexander Dowling and Tengfei Luo have developed a high-performance, energy-saving Thermoelectric equipment.
The novel process uses intense pulsed light to sinter thermoelectric materials in less than a second (conventional sintering in a thermal oven can take hours). The team accelerated this method of turning nanoparticle inks into flexible devices, using machine learning to determine the optimal conditions for the ultrafast but complex sintering process.
Zhang said flexible thermoelectric devices offer great opportunities for direct conversion of waste heat into electricity as well as solid-state refrigeration. They have additional benefits as power sources and cooling devices – they don’t emit greenhouse gases, and they’re durable and quiet because they don’t have moving parts.
Despite their potentially wide-ranging implications in energy and environmental sustainability, thermoelectric devices have not gained widespread application due to the lack of a method for rapid and cost-effective automated manufacturing. Machine-learning-assisted ultrafast flash sintering will now make it possible to produce high-performance, environmentally friendly devices much faster and at much lower cost.
Zhang offered, “The results can be applied to power everything from wearable personal devices to sensors and electronics, to the industry Internet of Things. The successful integration of photonic flash processing and machine learning has been shown to support energy and electronic materials.” Can be generalized to a wide range of highly scalable and low-cost fabrications.
Zhang is the principal investigator of the Advanced Manufacturing and Energy Lab at Notre Dame. Dowling, assistant professor of chemical and biomolecular engineering, and Luo, Dorini family professor for energy studies – both experts in machine learning – contributed to this research, along with doctoral student Murtaza Saidi-Jawash (now at California State Long Beach). Assistant Professor). , doctoral student Ke Wang and postdoctoral associate Minxiang Zeng (now an assistant professor at Texas Tech University).
Hard data made the essence! “Films also show excellent ductility with 92% retention of power factor (PF) after 10”3 Bending circles with 5 mm bending radius. Furthermore, a wearable thermoelectric generator based on flash-singled films generates a very competitive power density of 0.5 mW cm.-2 at a temperature difference of 10 K.”
This is great for a technology that hasn’t been given a ton of research money. Flexible thermoelectric generation is actually a very new field. Is this commercial? We’ll have to see if/when anyone tries it. When it is tried it is expected to be a resounding success and lead to a lot of investment and progress,
via New Energy and Fuel by Brian Westenhaus
Read more from Oilprice.com:
#Machine #learning #helps #build #highperformance #thermoelectric #devices