Prof. Yue Cao Energy Systems Group @ OSU

 

Prof. Cao's Q&A

 
As a recent graduate, I went through a few periods of questioning about research interests, career paths, life goals, etc. I hope to use this section to answer the same questions a student may have. The list will grow longer as time goes. Note: 1) I will be as objective as possible, and 2) this is only my personal advice and does not reflect the university's official guidance.

Q: Should one pursue MS only or PhD?
A: Let's be realistic - if you are asking this question, you are probably not sure if you want to commit to a research oriented career. First of all, I want to assure you that a research career can exist both in industry and academia. I think if you really want a true power electronics career right after school, you probably want to do a PhD. Power electronics is a very comprehensive subject that requires several domains of knowledge and years of experience. An MS can get you a job but more likely in a side field, such as test engineering, basic circuit design, application engineering, etc. Most true power electronics job positions you can find are senior levels and have 5+ years experience kinds of wording, which a PhD will get you. To be honest, I tried to find a power electronics job after MS, but no luck. After I was in the PhD program, many opportunities came up, even as internships, and they possibly come to knock on your door themselves. Another point - unlike many other fields in which PhD may sound very academic, power electronics PhD life is very fun because much research work is industry relevant, and some projects are directly funded by industry. These skills are readily applicable after graduation, whether you choose industry or academia. Another major difference will be the salary and opportunity cost. Fresh MS graduates typically make $80-90k/yr in most places ($100k+ possible in high-cost cities). Fresh PhD graduates typically make $100-120k/yr in most places ($130-150k+ possible in high-cost cities), plus bonus/stock/retirement benefits (another $20-50k value), etc. PhD's do miss out 2-3 years of major income during the extra school years, but the difference should be made up pretty quickly once working full time. PhD's also tend to climb the corporate ladder faster, but MS/MBA can still do depending on the person's ability. Last is about the study length. A direct BS-to-PhD may finish in 4-5 years focusing on a defined topic, whereas a returning PhD route (work after MS first and then come back) may still require an extra 3-4 years (total 5-6 years including MS) because the research momentum needs to start over.

Q: Should one study power electronics or power systems?
A: I struggled on this question a lot back then. The widely acceptable answer is if you like software and math, choose power systems, or if you like hardware and hands on, choose power electronics. I think it is partially true because power electronics and power systems are merging, and many applications require knowledge of both. Based on my experience, there is still much software/modeling/math work in power electronics based systems, but you need to be OK with lab work. Some traditional power electronics work requires you to make circuit boards or spin motors or program controllers. Power ranges from hundreds of watts to megawatts. On the other hand, in power systems, you are basically on paper all the time - mostly modeling and coding, dealing with kV or higher systems. In terms of job placement, power electronics can land you in a variety of industry fields, including EV, aerospace, renewable energy, energy services, consumer electronics, semiconductors, oil/gas, mining/metal, modeling software, etc. For power systems, some of my friends went to ISO's, utilities, energy services, software, and financial sectors. Sample power-related job titles and companies can be found on https://engineering.oregonstate.edu/Academics/Degrees/electrical-and-computer-engineering/energy-systems In the end, I want to congratulate you on choosing a power career. It is a long-lasting experience-based career that solves humanity-wide energy issues. Although the current salary may not be as attractive as that of a hot CS software job, the job security is high, skill is always in demand, and the salary grows steadily even when you past age 60.

Q: What are some factors one may overlook when selecting a graduate school?
A: 1) Location - This is the biggest factor I didn't think about carefully when I applied to graduate schools, but it is actually vital to your career path and lifestyle. For career, ideally the school is close to where the jobs are because companies usually send multiple recruiting teams to nearby states. If you look at job placement surveys, majority of the graduates stay in-state or go to nearby states. Many of my friends (including myself) traveled across country to land jobs on the west coast - isn't it better just close by? For lifestyle, you are going to spend a significant length of 4-5 years for a PhD, why not somewhere beautiful with mild weather (not cold in winter or hot in summer) and access to mountains and beaches? Trust me you will get uneasy for your research from time to time, so a suitable place nearby to vent that frustration will be almost necessary. In my opinion, a college town within 1-2 hours from a major city is the best. A college town usually has affordable housing (~half of the cost as in big cities) and short commute (saving up to 1 hour a day!). You will have the peace of mind and all the essentials, and just go to the big city for a weekend getaway for better foods, entertainment, and big-brand shopping.
2) Not all about ranking - University-wide, college-wide and even specific major rankings do not really accurately reflect the individual research program/group's quality. There are many factors behind the ranking that do not have direct relationships with the research, such as admission rate, retention rate, graduation rate, donation dollar, etc. It is probably true that a highly-ranked school is great overall, but it doesn't mean a research program within a highly-ranked school is always good, and a research program within a relatively lower-ranked school is not good. The power and energy programs, for example, are a niche area that only a handful universities have and are good at. Some highly-ranked schools (e.g., Harvard) have a power program but that may not be the best. OSU's power program was well established since 1980's and has been running strong with a lot of investment. It is a well-reputed power program in the North American academic and industry circle. This reputation is much more important than a USnews ranking. By the way, OSU Electrical Engineering USnews ranking is about #50-55 nationally (this ranking is through a paid subscription list that's not free to public). For a "power" program ranking, we estimate OSU at top 15 nationally and top 3 on the west coast. Perhaps a more relevant ranking is by National Science Foundation's (NSF) annual research funding awards (https://dellweb.bfa.nsf.gov/Top50Inst2/default.asp), which directly measures a university's research activities and quality. OSU has been steadily ranked among the top 10 over the years, and oftentimes beats the big named schools.
3) Career objectives - If you want to get into industry or national labs after PhD, the school's ranking is not really that important; what's more important: a) your research area, skillsets, and accomplishment; b) your own or adviser's connection; c) location close to those jobs. Within industry choices, if you want to go to more of OEM's (e.g., Tesla, Apple, Amazon, etc. but search what is OEM), you may want more of a "power electronics systems" research area as the employers have a focus on system integration. If a research group's area is more on the component/circuit level, usually the job placement is more of in first-tier or second-tier suppliers (e.g., TI, Infineon, ABB, Eaton, etc.). These are general observations, but of course there are crossovers. Personally I know a friend who worked on high-efficiency high-density power electronics circuits in PhD, but he is now doing modeling and control of microgrids for an OEM. Last, if you want to get into academia (i.e., professor), then school's ranking and adviser's reputation can matter. However, this does not mean a graduate from a lower ranked school or with a young adviser does not stand a chance. It's more about the individual's capability, and there is always an opportunity to do a Postdoc in an elite university/group. For example, my previous postdoc was recommended to UIUC and later got recruited by Ohio State Univ. Also there are many lesser-known but good local or regional universities who are happy to hire a professor in power and energy directly after a PhD without a postdoc.
4) Access to adviser and facilities - It is critical to meet your adviser 1-1 often and productively. I'd say 45-60 min per week is preferred. To me an ideal research group size is 5-8 students for whom I can have enough time to give detailed attentions. Does the adviser really care about the student's career growth? Also does the group already have enough lab space for every student as well as large-scaled equipment (say kW's to 100 kW's) so that the student can immediately focus on the lab work instead of worrying about long-lead lab constructions? Does the department have high-quality faculty members in other areas (within Power and outside)?
5) Group culture and potential - After all, you want to enjoy your time inside and outside the research group. Is the research group diverse and collaborative? What is the work-life balance and vacation policy? Is there sufficient amount of independence? For the group's potential, is the adviser still active in the professional community, actively publishing, taking service roles, and gaining broader networks? Are the research topics emerging and representing the needs of the state-of-the-art?
6) Professor's potential research areas - When I meet students at conferences, often they like to ask "What's your research area, what projects do you have?" This is a hard question to be honest, because i) almost all professors cover several different research areas, so as long as the professor is not too far from your interest (e.g., magnetic design professor vs. microgrid control student), you should consider working with him/her; ii) a professor's current projects do not necessarily indicate that his/her (and your) future projects will be in the same area, so don't limit yourself to what you just assume.

Q: How important are test scores and GPA?
A: A high quantitative score is an obvious requirement as a graduate engineering student. Beyond that, personally I value great writing ability (GRE more on the reasoning and TOEFL more on the fluency) followed by speaking skills (for international students). Of course, there are also other ways to demonstrate these, such as in the statement and during a conversation. GPA is important, but I do look more closely at your EE courses' grades.

Q: What is your advising style?
A: This is a tough question, and I don't want to categorize myself to a particular type. Given my diverse background - mixed education in China and the US, several industry jobs, and under different advisers and supervisors, I'd say I would deliver a hybrid advising style that suits you. Most of my past advisers are more hands-off (or high-level guidance), and I want to stick to that type if possible. However, it will be a lie if I tell you that your life will be comfortable - one is this hurts you in the long run, and another is this is not possible for a tenure-track assistant professor. In my experience, students advised by younger professors do output more results, and it really helps the students in the long run. Described on my other pages though, I do value the importance of work-life balance and self-motivation. I meet all of my students 1-1 for 45-60 min per week and give everyone plenty of attention. We also have an attractive vacation policy.