Hi!
There are 4 sets of data in the attached mat file:
Data_estimate_dry
Data_validate_dry
Data_estimate_wet
Data_validate_wet

Note that for the wet, the output data (y1) decreases at a much faster rate than the dry.
I used the following to estimate the transfer function based on dry set of data:
Opt = tfestOptions(‘Display’,’on’);
mtf = tfest(Data_estimate_dry, 2,2,13,Opt);
Used another set to check the fit:
figure, compare(Data_validate_dry,mtf)

The resulting fit looks good with 88% fit.
Then I used wet data to run it through the estimated transfer function:
figure, compare(Data_validate_wet,mtf)

As expected, the fit is not good. However, during the decay phase, I expected to see a big different between simulated data and the wet data.

Is this due to different input signals for wet and dry data?Hi!
There are 4 sets of data in the attached mat file:
Data_estimate_dry
Data_validate_dry
Data_estimate_wet
Data_validate_wet

Note that for the wet, the output data (y1) decreases at a much faster rate than the dry.
I used the following to estimate the transfer function based on dry set of data:
Opt = tfestOptions(‘Display’,’on’);
mtf = tfest(Data_estimate_dry, 2,2,13,Opt);
Used another set to check the fit:
figure, compare(Data_validate_dry,mtf)

The resulting fit looks good with 88% fit.
Then I used wet data to run it through the estimated transfer function:
figure, compare(Data_validate_wet,mtf)

As expected, the fit is not good. However, during the decay phase, I expected to see a big different between simulated data and the wet data.

Is this due to different input signals for wet and dry data? Hi!
There are 4 sets of data in the attached mat file:
Data_estimate_dry
Data_validate_dry
Data_estimate_wet
Data_validate_wet

Note that for the wet, the output data (y1) decreases at a much faster rate than the dry.
I used the following to estimate the transfer function based on dry set of data:
Opt = tfestOptions(‘Display’,’on’);
mtf = tfest(Data_estimate_dry, 2,2,13,Opt);
Used another set to check the fit:
figure, compare(Data_validate_dry,mtf)

The resulting fit looks good with 88% fit.
Then I used wet data to run it through the estimated transfer function:
figure, compare(Data_validate_wet,mtf)

As expected, the fit is not good. However, during the decay phase, I expected to see a big different between simulated data and the wet data.

Is this due to different input signals for wet and dry data? tfest, system identification toolbox MATLAB Answers — New Questions

​