Yutian Lei wins prize for paper at IEEE COMPEL workshop
Graduate student Yutian Lei won the award for best paper out of 128 peer-reviewed contenders at the 2014 IEEE Workshop on Control and Modeling for Power Electronics, or COMPEL, which took place at the University of Cantabria in Spain. Lei's publication, co-written with former graduate student Ryan May (now at Texas Instruments) and ITI researcher and ECE Assistant Professor Robert Pilawa-Podgurski, explored their research into switched-capacitor power converters.
Power converters can be found inside almost every electronic device, from cell phones and laptops to solar panels and TVs. They convert power from the source voltage, like a 4-volt battery, into the voltage the device needs, like a 1-volt cell phone. Usually they do this through a series of inductors or transformers, components on the circuit board that store energy and discharge it at the correct voltage.
Current research is trying to push power converters in two directions: making them both smaller and more energy efficient. Inductors and transformers are traditionally very efficient, but they're hard to shrink. Other components called capacitors, on the other hand, can store much more energy for their size inside circuits, but when energy is transferred directly from one capacitor to another, significant amounts of energy is dissipated as heat. As a result, circuit developers usually have to deal with a trade-off between either making their devices burn through less energy or take up less space.
Pilawa's team, however, working from foundational ideas established in Pilawa's PhD thesis, found a way to design and implement a power converter based on a series of capacitors that was not only smaller than current designs, but also about three to four times as efficient.
The team's design accomplishes this by cleverly tweaking the way that power flows through the circuit. Capacitors in a power converter usually alternate between directly taking in and discharging current, but at each current switch, or alternation, the capacitor loses some of the energy as heat. Pilawa's team solved this by altering the switching process to include soft charging, a buffer period between charged and discharged energy states that lets the capacitor conserve some of the energy it would otherwise lose when current runs through it.
Power in this method starts at the source, such as a battery, and travels through a series of step-down capacitors. Each capacitor shuffles the charge around and shaves off some voltage before passing it on to the next capacitor like a game of hot potato, until the power arrives at the end and is delivered to its output.
The theoretical part of the research was challenging, Lei admitted, but the hardest part for him as lead writer of the paper and in doing much of the research on the project was bringing the ideas and formal mathematics he had developed to the actual hardware. No matter how perfect the design looks, Lei said, something can always go wrong when it's implemented in hardware. Especially in the field of power electronics, in which very large amounts of energy are being channeled through circuits, things can get dramatic when the hardware malfunctions.
Often, especially since research tends to push our devices to their breaking point, when something goes wrong components can actually explode and the whole printed circuit board is ruined, Lei said. My office is a graveyard littered with dozens of these broken boards.
Almost equally challenging for Lei were the processes of setting his research down in writing in a language that others who hadn't been pursuing his line of research could understand, and explaining it to the brightest minds of the power electronics world at the COMPEL conference in Spain.
Presentation is traditionally the facet of this process that engineers have the least experience with, Lei said. However, Professor Pilawa really emphasized presentation skills by having us grad students present our research to each other every week and give each other feedback, and he ran us through practice sessions right up until the conference. By the time I was presenting my research to the conference, I'd gotten confident enough in my work to explain to them everything about my research.
After more than a year, Lei's efforts paid off. After his presentation, professors whose work Lei had read and admired actually came up to him and asked for a copy of the paper. Pilawa has since been contacted by a number of attendees from the conference who want pre-distribution copies of Lei's paper so that they can start research on his concept.
The primary reason we were at the conference was to listen to others' ideas and get a feel for what direction the power electronics field was taking, Lei said. That said, winning the award at the end was definitely a plus, as was getting the attention of all of these amazing figures in my field, he admitted, chuckling.
Pilawa had a thing or two to say about Lei's performance, as well. He is, according to Pilawa, not only intelligent enough to wrangle with difficult concepts, but hardworking enough to devote hours every day to testing and redeveloping the concepts he's come up with until they are physically feasible.
The best grad students are both hardworking and smart, and Lei is both, Pilawa said. This paper is also just the beginning of our efforts. Thanks to the long hours he and co-author Ryan May spent both on pen and paper and in the lab, we've highlighted a fundamental breakthrough in circuit design that research teams all over the world are now going to exploit.
Editor's note: media inquiries should be directed to Brad Petersen, Director of Communications, at bradpillinois [dot] edu or (217) 244-6376.