#TITLE_ALTERNATIVE#
Applied mathematics problems that we face in everyday life not only the direct problems but also the inverse problems, with the lack of information. For example, known some values of a function and we want the second derivative of this function. We can get it with the finite difference method and ot...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/20329 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Applied mathematics problems that we face in everyday life not only the direct problems but also the inverse problems, with the lack of information. For example, known some values of a function and we want the second derivative of this function. We can get it with the finite difference method and other discretization technique. However, because the data obtained from the observation is incomplete and contains errors, the result also contains errors which can be very large. Here we review the function of second-order derivative search using regularization techniques to reduce errors from the measurement. One technique is to transform the differential equation into a normal equation, by multiplying the differential equation with the transpose of the derivative operator. This technique will be compared with the Tikhonov regularization technique. We will review that in some cases solving the problem with direct inverse technique is not good enough to get satisfactory results. Meanwhile, for some necessity, we need accurate results. In the end the Tikhonov regularization technique is a better solution for the problem |
|---|