Commit d4ca50f7 authored by Ibrahim Ahmed's avatar Ibrahim Ahmed
Browse files

made preprocessing/ioops scripts + ambient temp calculation

parent 64e604a8

.gitignore

+1 −0
Changes for .gitignore: 1 added line, 0 removed lines.
Original line number Diff line number Diff line
@@ -5,3 +5,4 @@ desktop.ini 
.vscode/

*.pyc

__pycache__

*.ipynb_checkpoints
 
 No newline at end of file

SystemInfo/ESB Chiller 1.csv

+12798 −12799

File changed.

Preview size limit exceeded, changes collapsed.

SystemInfo/ESB Chiller 2.csv

+12954 −12954

File changed.

Preview size limit exceeded, changes collapsed.

SystemInfo/Tower Fan Speed %.csv

+12830 −12830

File changed.

Preview size limit exceeded, changes collapsed.

docs/5-preprocessing.md

+11 −1
Changes for docs/5-preprocessing.md: 11 added lines, 1 removed line.
Original line number Diff line number Diff line 
# Pre-processing



## Data type



Ideally, all cells in the dataset are numerical. Some preprocessing is required to remove errant `???` symbols and text.



The raw data contain dates with timezones. Preprocessing removes timezone information.



These operations are done in `ioops.py` when excel files are converted to csv.



## Missing temperature values



The towers timeseries occasionally have missing values for `Outside Air Temperature`, `Outside Air Humidity`, and `Ambient Wet-Bulb` temperature. However, given two values, the third can be calculated.
@@ -10,7 +18,9 @@ The following empirical and physical relationships are used: 


2. [Magnus formula][2], enhanced with Arden Buck fitting constants, relates dew point temperature to ambient temperature and relative humidity.



3. [Wet bulb temperature equation][3] by Roland Stull that relates ambient temperature and relative humidity.

3. [Wet bulb temperature equation][3] by Roland Stull that relates ambient temperature and relative humidity. This relation and its inverse are used to calculate ambient or wet bulb temperatures when one is absent.



These operations are defined in `thermo.py` and called in `preprocess.py` to clean up csv files.



[1]: https://en.wikipedia.org/wiki/Arden_Buck_equation

[2]: https://en.wikipedia.org/wiki/Dew_point#Calculating_the_dew_point