usage of FreeKiteSim

79 views
Skip to first unread message

Ramiro Saraiva

unread,
Oct 24, 2014, 10:37:41 AM10/24/14
to free-k...@googlegroups.com
Hello, my name is Ramiro Saraiva and I am currently researching AWE applications as part of my Post Graduation course in Control and Systems Engineering at the Federal University of Santa Catarina in Brazil.

I have managed to install and run the FreeKiteSim in my computer. In order to test some control strategies, I have implemented a communication interface between Matlab and the simulation software using the zmq socket library. The freekitesim sends cyclically the current state of the kite to matlab, and matlab sends the setpoints for the steering, winch speed and depower.

However, I have realized that, for some reason, the simulation do not accept to reel in the tether. When I send a positive setpoint for the winch speed, it works fine, but when the setpoint is negative the winch does not follow it. For negative setpoint of speed, it seems that the winch tries do keep the traction force at a given level and, to achieve that, it reels out the cable.

Is there any tip how to deal with this?

Best regards,

Ramiro

Uwe Fechner

unread,
Oct 27, 2014, 6:13:19 AM10/27/14
to free-k...@googlegroups.com
Dear Ramiro,

it is difficult for me to find a bug without having access to your code.

If you cannot publish it at the moment, perhaps you can send me a private email, or you try to write a minimal example
that shows the problem and put it into a gist repository (https://gist.github.com/).

Best regards:

Uwe

Ramiro Saraiva

unread,
Oct 27, 2014, 7:52:40 AM10/27/14
to free-k...@googlegroups.com
Dear Uwe,

Thanks for the prompt answer. This issue happens even with the original code, available in bitbucket. Running the FreeKiteSim as simulator, the axis #1 of the joystick defines the steering setpoint, and the axis #2, the winch speed  setpoint. For a positive setpoint, when the winch axis is saturated in '+1' for example,  the winch speed stabilizes at +8 m/s, i.e. 8 times the axis input, as is defined in the code. But when it is given a negative command, -1, the winch stabilizes between  +1 and + 3 m/s. It seems that this speed stabilizes when traction force is around 700 N. Is this the normal behaviour of the simulation?

Best Regards,

Ramiro

Uwe Fechner

unread,
Oct 27, 2014, 2:31:18 PM10/27/14
to Ramiro Saraiva, free-k...@googlegroups.com
Dear Ramiro,

Jan will investigate this issue tomorrow.

Best regards:

Uwe
--
You received this message because you are subscribed to the Google Groups "Free KiteSim" group.
To unsubscribe from this group and stop receiving emails from it, send an email to free-kitesim...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Uwe Fechner

unread,
Oct 28, 2014, 8:27:25 AM10/28/14
to free-k...@googlegroups.com
We checked your problem.

The model behaves correctly:
If you increase the reel-in speed slowly (0-100% in 2 seconds or more) and limit the reel-in speed to a value
of  3 m/s, then everything fine.

If you reel in faster, then you overload the motor and the motor swaps the direction and starts to operate as
generator again.  (Have a look at the speed - torque diagram at: http://en.wikipedia.org/wiki/Induction_motor)

To avoid that, you could do two things:
1. implement a winch controller that limits the force to 4000 N (nominal force of the winch)
2. depower the kite before you start to reel in

Currently the depower settings (angle of attack) cannot be controlled with the joystick.

This will be changed in the future.

Best regards:

Uwe

Ramiro Saraiva

unread,
Oct 30, 2014, 9:20:35 AM10/30/14
to free-k...@googlegroups.com
Thanks for the support.

Best Regards,

Ramiro

Uwe Fechner

unread,
Oct 30, 2014, 3:08:08 PM10/30/14
to free-k...@googlegroups.com
We already added code, that enables the control of the depower settings with a joystick,
just the documentation is missing. It will be added next week.

Uwe
--
Reply all
Reply to author
Forward
0 new messages