yliu2-19 / surface_roughness_analysis

This code takes the roughness of a surface, finds its power spectral density, and fits a power law to its PSD data.

surface_roughness_analysis

This code takes the roughness of a surface, finds its power spectral density, and fits a power law to its PSD data.

This README file will be used as a set of steps.

For a 1D roughness analysis, do the following.

1. After measuring the roughness of a surface, save the data as an matrix in IDL.
2. Take a slice of the data. (Ex: slice = XY[0:8250,1000])
3. Take the PSD of the data. (Ex: PSD = fft_powerspectrum(slice))
4. Convert that slice and PSD to a .txt file. They should then be in your home directory. (Ex. IDL> restore, 'si_pds' IDL> file = 'slice.txt' IDL> get_lun, unit IDL> openw, unit, file IDL> printf, unit, slice IDL> close, unit IDL> free_lun, unit)
5. (Since I was using windows, this is how I downloaded the files locally) Download the files in windows by opening a cmd window and inputting the following with you credentials.

   pscp.exe [email protected]:slice.txt C:\Put\Directory\You\Want\It\In

6. Make sure all the .m files and data are in the same directory.
7. Open roughness_analysis.m and change the surface slice data and PSD data with your own.
8. Change the initial prediction of k and n to your liking.
9. Make sure the x and y axes are consistent. (recall that the last point in the FFT and PSD should be half of the sampling frequency)
10. You might have to change the starting point of the fit to get rid of influence from outliers.
11. Replace any titles and labels with your own
12. Assuming your surface is isotropic, review your k2d and n2d.

---

For a 2D roughness analysis, do the following.

1. As described above, take a matrix of the roughness data and download it onto your computer.
2. Make sure all the .m files and data are in the same directory.
3. Replace the roughness data in two_d_roughness_analysis.m with your own.
4. When you call psd_2D_calculation in two_d_roughness_analysis.m, replace with your own parameters as described in psd_2D_calculation.m.
5. Change the initial prediciton of k and n to your liking.
6. Change the initial prediction of k2d and n2d to your liking.
7. Make sure the x and y axes are consistent. (recall that the last point in the FFT and PSD should be half of the sampling frequency)
8. You might have to change the starting point of the fit to get rid of influence from outliers.
9. Replace any titles and labels with your own.
10. Review your k2d and n2d.