[MBDyn-users] Question about initial conditions in wind turbine analysis

Pierangelo Masarati masarati at aero.polimi.it
Fri Mar 23 09:35:32 CET 2012

On 03/23/2012 07:32 AM, Manoj K. Dhadwal wrote:
> Dear all,
> Dear Mr. Masarati,
> In my wind turbine model [MBDyn version 1.4.2], if I set initial rotor
> speed greater than zero, I get the error "divide by zero, probably due to
> singularity in rotation parameters". The model works fine if the initial
> rotor speed is zero. I am using cosine half drive to smoothly ramp the
> rotor speed.
> Any possible explanations about this?

Likely, if you start with non-zero initial velocity, the analysis does 
not converge.  You're only seeing the final error (likely, some rotation 
parameter grew unbounded and some division by zero would have occurred). 
  You can check the progress of the iterative solution by enabling some 
minimal output using, in the "initial value" block,

     output: iterations;

You can inspect each element of the solution using

     output: solution;

A brief description of the meaning of each element will appear on the 
right of the value.

If your model assembles correctly and reasonably close to equilibrium, 
the analysis can usually start with non-zero angular velocity; I 
routinely do with flexible helicopter rotors.

Cheers, p.

Pierangelo Masarati
Associate Professor
Dipartimento di Ingegneria Aerospaziale
Politecnico di Milano

More information about the MBDyn-users mailing list