Archived OpenModelica forums. Posting is disabled.

Alternative forums include GitHub discussions or StackOverflow (make sure to read the Stack Overflow rules; you need to have well-formed questions)


Forgot password? | Forgot username? | Register
Page Start Prev 1 Next End

Shattering (Zero Effect) problem? Another bouncing ball Example

#6835  Permalink Top

Shattering (Zero Effect) problem? Another bouncing ball Example

Mar-16-16 02:16:32

Hi Guys,

I encountered an unpredictable behavior when running the simulation.

Let's start with the simple and successful example:

Code:


model BouncingBall

    constant Real g = 9.81;

    parameter Real c = 0.9;

    parameter Real radius = 0.1;

    Real height(start=1);

    Real velocity(start=0);

equation

    der(height) = velocity;

    der(velocity) = -g;

    when height <= radius then reinit(velocity, -c*pre(velocity)); end when;

end BouncingBall;

and following script:

Code:


loadFiles({"BouncingBall.mo"});

getErrorString(); 

simulate( BouncingBall,startTime=0,stopTime=5,outputFormat="plt"); getErrorString();

plot({height});

quit();

I can successfully get the plot I wanted.

However, if I use the same script but change the code to followings (all I did is to "abstract" the equation for re-use purpose), then the ball falls straightly into the ground, why is the code failed?

Code:


block derHeight

     Real height;

     Real velocity;

equation

    der(height)=velocity;

end derHeight;



 block derVelocity

     Real velocity;

     Real g;

equation

    der(velocity)=-g;

end derVelocity;



 block reinitVelocity

     Real height;

     Real radius;

     Real velocity;

     Real c;

equation

    when height <= radius then reinit(velocity, -c*pre(velocity)); end when;

end reinitVelocity;



model BouncingBall

    constant Real g = 9.81;

    parameter Real c = 0.9;

    parameter Real radius = 0.1;

    Real height(start=1);

    Real velocity(start=0);



    derHeight e1;

    derVelocity e2;

    reinitVelocity e3;    

equation

    velocity=e2.velocity;

    velocity=e1.velocity;

    velocity=e3.velocity;



    height = e1.height;

    height = e3.height;



    g = e2.g;

    radius=e3.radius;

    c = e3.c;

end BouncingBall;

Strange though, following code only abstract "e1" and it simulates successfully!

Code:


block derHeight

    Real height;

    Real velocity;

equation

    der(height)=velocity;

end derHeight;



model BouncingBall

    constant Real g = 9.81;

    parameter Real c = 0.9;

    parameter Real radius = 0.1;

    Real height(start=1);

    Real velocity(start=0);



    derHeight e1;

equation

    velocity=e1.velocity;

    height = e1.height;



    der(velocity)=-g;

    when height <= radius then reinit(velocity, -c*pre(velocity)); end when;

end BouncingBall;

Even more strange, following code only abstract "e2" but the simulation failed!

Code:


block derVelocity

    Real velocity;

    Real g;

equation

    der(velocity)=-g;

end derVelocity;



model BouncingBall

    constant Real g = 9.81;

    parameter Real c = 0.9;

    parameter Real radius = 0.1;

    Real height(start=1);

    Real velocity(start=0);



    derVelocity e2;

equation

    velocity=e2.velocity;

    g = e2.g;

    der(height)=velocity;

    when height <= radius then reinit(velocity, -c*pre(velocity)); end when;    

end BouncingBall;

Any helps? Thank you very much!
Miles

Edited by: milesma - Mar-16-16 02:27:01
#6836  Permalink Top

Re: Shattering (Zero Effect) problem? Another bouncing ball Example

Mar-16-16 05:36:25

The ball will typically fall through the ground if the velocity does not bounce it high enough, so you need a condition to just let the ball remain to 0 velocity when the numbers get too small.

#6837  Permalink Top

Re: Shattering (Zero Effect) problem? Another bouncing ball Example

Mar-16-16 05:39:06

if I remove "pre" function in my failed "abstract" version, so it looks like:

Code:




 when height <= radius then reinit(velocity, -c*velocity); end when;

then the problem solved, I can get the exact plot as the the following code:

Code:


model BouncingBall

    constant Real g = 9.81;

    parameter Real c = 0.9;

    parameter Real radius = 0.1;

    Real height(start=1);

    Real velocity(start=0);

equation

    der(height) = velocity;

    der(velocity) = -g;

    when height <= radius then reinit(velocity, -c*pre(velocity)); end when;

end BouncingBall;

Now the questions:
1. what exactly role is the "pre" function playing in my example? Is it necessary?
2. why the version with "pre" on the "abstract" version does not work?

#6838  Permalink Top

Re: Shattering (Zero Effect) problem? Another bouncing ball Example

Mar-16-16 05:46:03

removed duplicate reply

Edited by: milesma - Mar-16-16 05:46:57
#6839  Permalink Top

Re: Shattering (Zero Effect) problem? Another bouncing ball Example

Mar-16-16 05:46:37

sjoelund.se wrote:

The ball will typically fall through the ground if the velocity does not bounce it high enough, so you need a condition to just let the ball remain to 0 velocity when the numbers get too small.

Thanks for your reply, from the book "Principles of Object-oriented modeling and simulation with modelica 3.3" second edition by Peter Fritzson, I understood the case you mentioned (Zero Effect). However, my problem is at the first bounce, it will falls into the ground (as if there is no ground). You can run the script and code to have a look.

I found "pre" is the cause of my problem. Remove it will just solve it. However, I don't know the side effect of removing "pre", there must be reason to have "pre" there?

Thanks
Miles

Page Start Prev 1 Next End
Previous topic
Next topic
There are 0 guests and 0 other users also viewing this topic
You are here: