Most useful parts of electrical engineering?

[X89codered89X]

I'm currently enrolled in a PhD program in control systems engineering. I'm wondering what tends tend to be the most useful subsets of this field, and consequently the most important to be well-versed in. I've read Kalman filters are very important in real world applications. Optimal Control and Convex optimization also seem to be important.

My goal is ultimately to work in industry, possibly including starting my own company. I'm debating actually, if a job like this would require a PhD or just a masters. Especially if, in true startup fashion, I would have to be able to do everything including the applied digital/analog stuff and spend time learning more applied things (i'm not well versed in digital circuits at all! and this would take considerable time/effort).

P.S. *I change my mind about whether I want to get a PhD or not every day even thought I'm very passionate about nonlinear dynamics, control systems, and these sorts of things. It's a complete crap-shoot, despite my interest in the field. The PhD degree, at this point (I've just started) seems to be like a master's plus teaching experience. I have no interest in teaching/becoming a professor. Research would be fine/ good depending on what I get to work on, but I am tired of the TAing stuff and just want to learn this stuff for myself. Especially when I feel mathematically literate/mature enough to just buy things like a Dover optimal control book, or a Dover adaptive control book which are like ~$9, and just learn it by reading and teaching myself.

 

[JakeBrodskyPE]

I am a practicing control systems engineer. I am employed at a large water and sewer utility. The processes we have here are stupidly simple. They have to be.

Once effluent has left the plant, there is no recalling it. This has GOT to work.

Complex control systems are interesting from a theoretical perspective, but you need to remember that they're going to be used by Homer Simpson. Yes, there is more truth to the character of Homer Simpson than any of us would like to admit. Secondly, you need to remember that there are no geniuses in the cockpit or on the plant floor at 2 AM. Add a bit of stress to that from something going wrong, and you are going to have someone who is just barely able to function. Believe me, I've been there. I have been that idiot.

So whatever you do, it has to be as solid as a brick, whether it is going on a spacecraft , airliner, car, substation, turbine controls, boiler, catalytic cracker, distillery, or whatever.

I strongly recommend that whatever course you choose, whether the Ph.D or the Masters, that you take the EIT and aim for the PE BEFORE going into business for yourself. Learn how the real world works. Talk to the end users of the control systems you design.

Also realize that even the PhD and the PE don't mean a damned thing. You need experience. Engineering is a field of experience. You learn by doing. Where I work, we wouldn't let a recent graduate go out on their own for at least a year, and we won't give them significant projects to work on for at least three years.

Yes, we're parochial and stuck in our ways. When you lay pipe in the ground and expect it to last for over a century, you tend to be very conservative. When we build a pumping station, it is expected to last for decades. It better damned well be right because people are issuing bonds on this stuff with maturity not expected for 20 years.

You may find some places that are more willing to try some new ideas. But remember that if you expect traction, you're going to have to do more than show your certificates. You're going to need a track record.