Hello,
I've built a model, consisting of two bodies which are connected with a bushing. The body below is fixed with the ground and the body on the top is penetrated with a force as you can see in the picture (Einmassenschwinger).
I want to run a co-simulation with Simulink. For that I created a Simulink diagram, where an sin-Inputsignal is generated and sent to the NX-Motion-Plant. As an reaction of the forceinput, NX measures a displacement in a point an gives it as an output back to Matlab. I wrote a script, where the Frequency-respnose function is calculated and the bode-diagramm is plotted.
So when I start the simulation with those standard rigid bodies everthing works out. As soon as I upload flexible bodies into the model, see the other pictutre and restart the simulation with Simulink as host, I get the following error-message in Matlab:
=== Error Message ========================
1. NX/Motion Model is invalid
=> Modify NX/Motion Model and restart simulation
2. Bigger waiting time is needed to open NX/Motion Model
=> Increase Waiting time and restart the simulation
==========================================
=== Error Message ========================
The number of plant input of Simulink is not equal to the number of plant input of NX/Motion.
The number of plant output of Simulink is not equal to the number of plant output of NX/Motion.
==========================================
In the FE-modell of the bodies I used RBE2 elements and as constraints for the connection-points I used free DOF, excludung the fixationpoint to the bottom where I used fixed DOF.
By the way: when I use NX as host, the cosimulation in both versions, rigid aswell as flexible, works out.
So do you have any idea where I could find my mistake?
Did I create the flexible bodys in a wrong way in the FE-modelling? Or can't NX and MATLAB work together when using flexible bodies in a simulation?
I would be glad if anybody has a solution.
Thanks,
Linda
I've built a model, consisting of two bodies which are connected with a bushing. The body below is fixed with the ground and the body on the top is penetrated with a force as you can see in the picture (Einmassenschwinger).
I want to run a co-simulation with Simulink. For that I created a Simulink diagram, where an sin-Inputsignal is generated and sent to the NX-Motion-Plant. As an reaction of the forceinput, NX measures a displacement in a point an gives it as an output back to Matlab. I wrote a script, where the Frequency-respnose function is calculated and the bode-diagramm is plotted.
So when I start the simulation with those standard rigid bodies everthing works out. As soon as I upload flexible bodies into the model, see the other pictutre and restart the simulation with Simulink as host, I get the following error-message in Matlab:
=== Error Message ========================
1. NX/Motion Model is invalid
=> Modify NX/Motion Model and restart simulation
2. Bigger waiting time is needed to open NX/Motion Model
=> Increase Waiting time and restart the simulation
==========================================
=== Error Message ========================
The number of plant input of Simulink is not equal to the number of plant input of NX/Motion.
The number of plant output of Simulink is not equal to the number of plant output of NX/Motion.
==========================================
In the FE-modell of the bodies I used RBE2 elements and as constraints for the connection-points I used free DOF, excludung the fixationpoint to the bottom where I used fixed DOF.
By the way: when I use NX as host, the cosimulation in both versions, rigid aswell as flexible, works out.
So do you have any idea where I could find my mistake?
Did I create the flexible bodys in a wrong way in the FE-modelling? Or can't NX and MATLAB work together when using flexible bodies in a simulation?
I would be glad if anybody has a solution.
Thanks,
Linda
