How to choose initial parameters for grey-box model for estimation, which can give stable initial model?

Hi, I am using "greyest" function to estimate a model which has 50 paramters. I use "idgrey" to define the estimation model. However, when I use "greyest" to identify the parameters, it shows the following error -

Error using iddata/greyest (line 40) The initial computation of the loss function failed. The initial model, if specified, may be unstable. Consider setting the "EnforceStability" option to TRUE. Also make sure that the parameter bounds do not make the model unstable.

I understand the error, it tells me that the initial values of parameters I chose leads to an unstable model. I tried to set "EnforceStability" to true also, but with no avail. Also tried to set 'InitialState' to 'estimate', no success. Does anyone know how to choose initial parameters of a state-space model in such a way that resulting model is stable?

[honggildong]

Did you find a solution to your problem yet?

[Tobias Fischer]

Did you find a solution to your problem yet?:)

Answer 1

[佳伟 王]

I also have this problem. And what confused me is that, I can't set the "EnforceStability" option to TRUE.

Here is my code :

optEst = greyestOptions('InitialState','estimate','EnforceStability','true');

And matlab shows the errors:

Error using greyestOptions,the value of the "EnforceStability" option must be TRUE or FALSE.

I don't know why, and I think my model is stable.