In order to graduate from my university, I need to do a thesis for my employer (Consumer's Energy). Unfortunately, I have no idea what topic I might pick. My major is Computer Engineering, and the project is supposed to be anything that would help the employer and better me as an engineer. Kinda vague, I know, but I can elaborate if need be... Any ideas?
Finite state computer numerical modelling
For my Physics project, we built a thermal model of a spacecraft, attached a load of thermocouples and drove its temperature up and down in vacuum. We designed and built a computer model of the spacecraft and compared predicted results against those we measured.
Well, that's neat, but how does that translate to an energy company and a CE major?
Davros, this needs to be done in about 3 months time. It can be too complicated, but can't be entirely too easy either. I'm working on mine right now and it's not going so bad.
Govt, keep in mind, it's as much your company's responsibility to come up with something as it is yours.
> Govt, keep in mind, it's as much your company's responsibility to come up with something as it is yours.
Yeah, but my company suX0rZ, so I decided to take the initiative to find something myself... Or get someone here to toss a reasonable idea at me. That, and I'm bored.
hmm... interesting how you and I work for the same type of company, yet mine is 10x smaller and they overloaded me with ideas. ok... i guess overloaded wasn't a good term... but I had 5 or 6 good ideas to choose from.
hmm... I dunno... just start lookin around and seeing what you've done over the past couple years and where you could take that.
That's the problem - for the last year and a half I've been working on this lame ass program that I'd be stupid to do my thesis on. It's important enough to take up my time, but trivial enough that they don't really care when it gets done. I could get a 30 page paper out of it if I included all my code from both versions of the program. And that's about it.
> they overloaded me with ideas. ok... i guess overloaded wasn't a good term... but I had 5 or 6 good ideas to choose from.
They haven't really tried to give me any ideas yet (I only gave them the thesis manual yesterday) - like I said, I was bored, and for once, was taking initiative.
lol... ok. umm... do you know anything about handheld programming? we could do a group thesis ;)
Thesis rhymes with feces.
I'm not familiar with Consumers Energy, but if they are an energy/fuel supply company, one thing they are always interested in is cutting costs, (18 years in a gas utility is speaking here!).
One way of cutting costs is accurate prediction of demand. I wrote a whole load of demand prediction programs using simple regression statistics, Box-Jenkins type things, Bayesian statistics, and eventually an artificial neural network. Eventually, all these things were running side by side in the control room and the control staff guessed demand based on them, the dry piece of seaweed hanging on the wall and their own instincts.
If CE is anything like a utility, you'll always be able to interest them is some demand forecasting project.
>Well, that's neat, but how does that translate to an energy company and a CE major?
Numerical modelling can be applied to all sorts of things. How about modelling the supply and demand for energy, or modelling the efficiency of a turbine/generator. While working for an Aerospace company, I produced a computer program to model the performance of a stepper motor under various load conditions. The spacecraft example was just that, an example.
>Davros, this needs to be done in about 3 months time. It can be too complicated, but can't be entirely too easy either.
The project could be designed to be of a suitable complexity and depth. The spacecraft project I did all those years ago was done in under three months.
>Thesis rhymes with feces.
No it doesn't! Thesis is like Thee sys, Feces is like Fee Sees! And good luck Gov't! May the um, idea come to you in a, um, dream!
> If CE is anything like a utility, you'll always be able to interest them is some demand forecasting project
Hmmm... That, and Davros's ideas are interesting ones... Maybe I'll bring them up.... Dry piece of seaweed on the wall?
I'll keep you posted about what's up with it - hopefully we'll have a meeting in the next couple days to discuss possibilities.
Look at some of the ideas that they gave Ober.
I can acutally understand what Adrianxw is saying, having taken statistics, however in spite of that, it sounds complicated.
What type of system do they currently have? What is the input and output data format and how is it presented to each department. Take the data and provide a cleaner user interface by adding functionality to the user interface for each department (do some interviews and build the requirements) so that each department has to do less work making involving calcualtions or finding/sorting the information they use and don't use. You are probably using MS Windows? That is limiting for sure, but you should still be able to improve (in small ways) the existing system (especially user interfaces) but don't have time to re-engineer the existing system.
That's not a bad idea, Dean... However, we're big enough that there's no standard system across all of the dept.'s. We're spread out all over Michigan's Lower Peninsula, and so it'd be difficult to have some standard for every one of the thousands of people here.
Actually, I just sort of had an idea. We developed an in-house program (can't think of the name off hand) that allows substation designers to quickly design new substations or make major changes to old ones... Maybe I should have a look at that and see how it could be improved.