This module performs the "Transient Scanning Technique" presented in:

• Brouwer, J., Tukker, J., & van Rijsbergen, M. (2013). Uncertainty Analysis of Finite Length Measurement Signals. 3rd International Conference on Advanced Model Measurement Technology for the EU Maritime Industry. [PDF]

• Brouwer, J., Tukker, J., & van Rijsbergen, M. (2015). Uncertainty Analysis and Stationarity Test of Finite Length Time Series Signals. 4th International Conference on Advanced Model Measurement Technology for the Maritime Industry. [PDF]

• Brouwer, J., Tukker, J., Klinkenberg, Y., & van Rijsbergen, M. (2019). Random uncertainty of statistical moments in testing: Mean. Ocean Engineering, 182(April), 563–576. https://doi.org/10.1016/j.oceaneng.2019.04.068

It allows to easily detect transient portion of a signal and measure the statistical uncertainty with that portion removed.

Can be installed like any python package, for example:

pip3 install --user https://github.com/WavEC-Offshore-Renewables/pyTST/archive/master.zip

This package provides both a command line tool as well as a python library (for more flexibility).

If the signal data looks like:

# time   signal
  1     0.280910708014E-03 
  2     0.280910708014E-03
  3     0.345576259768E-03
...

the following can be used

TST-cli --time-col=0 --signal-col=1 example_data_filename

See TST-cli -h for more details on the capabilities.

Signal data can be loaded from a file:

from pyTST import pyTST

tst = pyTST()

tst.load_data_file("example_data_filename", signal_column=1, time_column=0, tstep=0.05)

tst.compute_TST(step_size=10)
tst.export_to_txt("TST_analysis.dat")
# tst.import_from_txt("TST_analysis.dat")
tst.plot()

Or provided via python arrays:

import numpy as np
from pyTST import pyTST

# Signal creation
t = np.linspace(1,1000, 5000)

signal = np.sin(t)

# Add initial transiant effect
signal[0:100] += np.linspace(1,0, 100)


tst = pyTST()
tst.load_data_array(signal_array=signal, time_array=t)

tst.compute_TST(step_size=10)
tst.export_to_txt("TST_analysis.dat")
# tst.import_from_txt("TST_analysis.dat")
tst.plot()

For more info, the library is documented via docstrings:

from pyTST import pyTST
help(pyTST)

This code can be cited with:

@software{lemaire_sebastien_2021_4428158,
  author       = {Lemaire, Sébastien and
                  Klapwijk, Maarten},
  title        = {pyTST},
  month        = jan,
  year         = 2021,
  publisher    = {Zenodo},
  version      = {v1.0},
  doi          = {10.5281/zenodo.4428158},
  url          = {https://doi.org/10.5281/zenodo.4428158}
}