Resampling

Sometimes the measured data is sampled within too small a sampling time Ts (sampling frequency too high). A proper sampling time is about Ts <= T/10, where T is the system's dominant time constant. You can resample the datasets in your project. The down-sampled datasets will also benefit the later modeling by reducing the training time.

Note
Resampling affects all training and test datasets in your project.

Proceed as follows to resample the datasets.

1. In the main window, select Data → Resampling.

   The "Resampling" window opens.

   

2. In the "Resampling Properties" area, do one of the following:

   • In the "Sample Time" field, enter the desired sample time in seconds.

   • In the "Resampling Factor" field, enter a factor relative to the original sample time.

     For example, a resampling factor of 2 doubles the absolute time difference between two samples of the original data.

   The resampling property you did not edit is updated automatically. The new and the original training and test data are calculated, and the fields in the "Data Sizes" area show the numbers of samples using the current sample time as well as the original one. The Details button opens a window that shows the same numbers for each dataset.

3. Click on OK.

   The datasets are resampled based on linear interpolation.