Papers
I'm reading a paper which says that there is no way to simulate a system using a particular value of x, so we have to model it in such a way that it's qualitatively wrong but quantitatively right.
The problem is that you want to model two values, one of which is much smaller than the first...by like a dozen orders of magnitude. I know that I'm going to have read more about this (they solve the problem using an approximation for the smaller value and the ratio of the two)...but I'm really having a problem with the notion of a "qualitative" model.
Admittedly, coming from a background in antennas, "qualitative" simply doesn't cut it. You're designing a device to work within a pretty narrow set of parameters, and if you have any issues like the one mentioned above, the best approximation is to chuck the really small value.
This, I guess, is new territory for me. I'm pretty sure this is one of the issues we'll be looking at with my thesis project, so it is obviously something to be improved upon. It's just bizarre to move from an area where we can and do figure things out fairly precisely into one where the actual numbers aren't as important as how the system as a whole behaves.
The problem is that you want to model two values, one of which is much smaller than the first...by like a dozen orders of magnitude. I know that I'm going to have read more about this (they solve the problem using an approximation for the smaller value and the ratio of the two)...but I'm really having a problem with the notion of a "qualitative" model.
Admittedly, coming from a background in antennas, "qualitative" simply doesn't cut it. You're designing a device to work within a pretty narrow set of parameters, and if you have any issues like the one mentioned above, the best approximation is to chuck the really small value.
This, I guess, is new territory for me. I'm pretty sure this is one of the issues we'll be looking at with my thesis project, so it is obviously something to be improved upon. It's just bizarre to move from an area where we can and do figure things out fairly precisely into one where the actual numbers aren't as important as how the system as a whole behaves.