Posted by Patrick Dingle at 04/24/2001 12:02 AM EST
College Student on team #639, Red B^2, from Ithaca High School and Cornell University.
In Reply to: Re: Mechanical Engineers
Posted by Chris Hibner on 04/23/2001 2:03 PM EST:
Chris,
Thanks very much for your post… sounds almost exactly what I’d like to do… In fact I found a undergrad concentration called mechanical systems which has courses such as robotics, feedback control systems, modeling / simulation of dynamic systems, mechatronics systems. What you describe sounds too good to be true – mechanical, math, computers all in one!
Thanks again
Patrick
: If you would like to meld computers AND mechanical engineering, you should check out either control systems or computational mechanics, which is what I do at TRW.
: Control system theory is a branch of dynamic systems in which the main objective is to automatically control the dynamic system. The system is controlled using feeback sensors (typically potentiometers, gyros, accelerometers, etc.) and a control algorithm that uses the feedback to determine how to actuate the system. When you study control, you mostly learn how to write the control algorithms to perform the way you want them to.
: Most modern control systems are digital, so it helps if you enjoy writing code in C. It is also increasingly important that you know how to work with MATLAB and Simulink. At work, I have MATLAB or Simulink open from the time I arrive until the time I go home. It is a great tool set and I would recommend that almost all engineers learn it.
: If you want to study control system theory, you should concentrate your courses in dynamics, dynamic systems, control systems, and a LOT of linear algebra and matrix algebra (matrix algebra is a subset of linear algebra). It’s also helpful to study some basic electronics and computer programming.
: If you really want to blend software design and mechanics, computational mechanics is where it’s at. In this field, you do a lot of mathematical modeling of mechanical systems (either structural, dynamic, or thermo/fluid) and create the solutions to the models using computers. Most of the solutions are approximate solutions using mathematical approximation techniques. A lot of pre-packaged solvers exist depending on the what you’re looking for (NASTRAN for Finite Element Analysis, ADAMS for dynamics, etc), but a lot of times you must write your own simulation package. I write quite a bit of simulation software here using C/C++, and MATLAB and Simulink.
: The only thing about computation mechanics is that it is mostly only offered in graduate level engineering courses. You can take a finite element analysis course as an undergrad, but you don’t spend a whole lot of time studying the mathematics behind the solvers. When you get to grad school, this is mostly what you do. I’ve taken a couple of courses in this area in grad school (I just completed on this past semester) and I’ve thouroghly enjoyed them.
: If you want to go into computational mechanics, you should take a LOT of math - especially take a lot of matrix algebra and linear algebra courses with your electives. You should also take a linear systems course and pay special attention to function spaces (this is the basis for approximation theory).
: Before I got to college I used to love to write software for various things. I used to do this as a hobby. It’s great that I now get to combine this with mechanical engineering.
: I hope this helps.
: -Chris