WECCCOMP – Forums › WECCCOMP forum › PTO dynamics
 This topic has 3 replies, 2 voices, and was last updated 6 years ago by John V. Ringwood.

AuthorPosts

11th May 2018 at 1:16 pm #1402
Dear Organizers,
I have some doubts as to the correctness of the PTO dynamics you have included in the WECCCOMP simulator.
At the moment it is defined in the ‘Cmd torque to measured torque dynamics’ block as a discretetime transfer function withNumerator = [ 0.9821 1.9618 0.9797];
Denominator = [1.0000 1.9978 0.9978];
Sample_time = 1;Two minor remarks to begin with :
* Sample_time = 1, means that the sample time is inherited from the simulation time step, that is 0.001 s. I would have specified the sample time directly, to make the code cleaner (and allow to obtain correct results even when the simulation time step is change). By the way, only the newest versions of Simulink are tolerant enough to allow the use of inherited sample time in these cases.
* When using a discretetime function, for the sake of accuracy, one should keep as many decimal digits as possible. 4 digits could be too few when the sampling period is short. If the original transfer function was obtained by system identification, the best option would have been to store and retrieve the coefficients from the workspace via a .mat file.Beside that, the resulting dynamics is really peculiar :
– There is a direct term between input and output, whose value is 0.9821.
– The steadystate gain is not 1 (it is 1.091)
– Beside the instantaneous effect of the direct term, there is a quite slow time constant.This means that if I apply an unitary step to the commanded PTO torque, the response will instantaneously rise to 0.9821, than it will settle to 1.091 after 1.52 s. Which, in my opinion, cannot really represent the physical behavior of a forcecontrolled linear motor.
Could you please comment on my remarks?
Thanks
Kind regards,
Paolino Tona
11th May 2018 at 2:56 pm #1403Paolo,
Thanks for your comment – all reasonable points. We need to have a short internal discussion to see the best way to proceed.John
18th July 2018 at 2:26 pm #1633Hi John,
I am getting back to you concerning PTO dynamics. A few weeks after your reply, a new version of the model appeared on GitHub with an updated cmdtoforce transfer function. The new transfer function still has a direct transfer term but also shows an oscillating behavior, with a (5%) settling time of about 1s.
This modeling choice still puzzles me. Has the transfer function been obtained by system identification using experimental data from the real PTO? Would it be possible to study these data to better understand your choice and get prepared to the implementation of our controller?
Thanks
Kind regards,
Paolo
8th August 2018 at 4:24 pm #1678Paolo,
The structure of the model is inspired by the physical system, with the parameters determined by data fitting. I believe the people at Aalborg/NREL/Sandia ahve good faith in this torque loop model. However, Giorgio Bacelli is happy to provide further details on the origin of the model is you really need it. The direct feedthrough comes as a result of the fact that the force measured by the load cell is not exactly the force generated by the motor/gen but is the force exerted by the motor slider on the arm, while the input to the model is the desired/commanded force. The underlying issue is that the motor and slider/arm inertia are nonnegligable above a certain frequency. If the direct feedthrough is causing you some problems (e.g. with algebraic loops, etc), you could put a z^{1} (delay of 0.001s)……….. 
AuthorPosts
 You must be logged in to reply to this topic.