pyControl

Abstract

The primary aim is to develop a program to allow the user design the experiment and the run the process independently. The program works in following fashion:

1. Setting laser parameters for patterning,
2. Setup surrogate model, initiate moving the XYZ stage,
3. Auto adjust z axis of motion controller for each spot,
4. Start patterning,
5. Acquiring Raman Data,
6. Raman Data Analysis and export,
7. Update surrogate model, predict new candidates,
8. Repeat process 2-7 until all the spots are patterned,
9. Save the Optimizer model for post-analysis.

Instruments:

1. Newport ESP300 XYZ controller
2. Thorlabs MFF101 Flip Mirror
3. Lighthouse Photonics Sprout G12W Laser
4. Princeton Instruments Isoplane SCT320 Raman

Current Challenge

1. Update notebooks to python files.

2. Velocity control for line patterning.

HOW-TO

1. Git clone this repo to your local machine.
2. Create a new virtual environment, activate it and install requirements.
3. Run notebook pyControl/Updated Codes/main program.ipynb. (https://github.com/hududed/pyControl/blob/master/codes/main%20program.ipynb) -->Inside current directory, a new folder will be created as "Campaign " + "Current Date" . In this folder experimental data will be recorded. --> Parameters will be written at directory/campaign_current_date/dataset.csv. (TO-DO: fixed paths)
4. Imports pressure, IsoPlane Spectrometer, Motion Controller, Flipper Mirror and Laser files.
5. Run main program.
6. Set start (x1,y1), end coordinates (x2,y2) and intervals (dx,dy).
7. Following outputs are written to results: (TO-DO: fixed paths)
   • Background1D and Background2D csv files are created. The center wavelengths are 1500 and 2700.
   • Two Raw Raman spectrum files foreground1D.csv and foreground2D.csv for each pattern.
   • One updated file for MBO dataset.csv with each pattern as newline,
   • the optimizer model opt.state.rds are saved at `directory/campaign_current_date/'
   • All the csv files for foreground and background 1D, 2D are saved at `directory/campaign_current_date/'.

Line-Patterning

1. Define threads: (a) motion controller thread (b) mirror switching thread
2. Prepare initial data:
   • dataset.csv contains post patterning spots raman spectra data
   • dataset-pre.csv contains pre patterning spots raman spectra data
   • data.csv contains only one data for one line and this file interacts with BO.
3. Assign number of lines to be patterned, starting position in x axis, starting position in y axis, vertical gap between the lines.