Process model with second order disturbance model(Identified with System identification toolbox) convert into Simulink block diagram(not to import with Idmodel)

Hello everyone ...

I am new user and please assist me in my question .

I have identified model with System Identification toolbox which is process model with only 1 pole and 2 order disturbance model . I have 2 input as temperature data from sensor and output also temperature .It is multiple input single output model .With described model I have good fit and residual analysis is also perfect .I have used system Identification GUI.

As per question I want to use process model equation[ y=Gu+(C/D)e ] with simulink basic block (not with idmodel).I am looking for help in it . Also looking for more information in disturbance model.

Thank in advance

Answer 1

[Rajiv Singh]

Get the numerator and denominator polynomials corresponding to the treansfer function G using tfdata:

[num, den] = tfdata(model)

Use num, den in a built-in continuous-time Transfer Fcn block.

C, D polynomials corresponding to the noise component can be extracted from the identified model using its "NoiseTF" property.

C = model.NoiseTF.num

D = model.NoiseTF.den

Use C,D in a second Transfer Fcn block that is connected in parallel to the primary transfer function block for G (defined using num/den). You will need to do this only if the noise component is nontrivial (C, D different from 1).

Source: you need 2 input signals. The first one is the measured input signal "u(t)". This can be provided, for example, using a Signal Generator, an IDDATA source block, or a "From Workspace" block.

The second one should provide the noise signal "e(t)". For this, you can use a Band-Limited White Noise block. Set the Noise power to model.NoiseVariance and Sample time to a value that corresponds (at least roughly) to the sample time of the data that was used for identification of the model.

[Roshan Sanjay Bagul]

Thanks, @Rajiv Singh for answering my question ...

I have attached an image of the Simulink model . Please let me know if it is correct.

Actually, I am quite new to the disturbance model. Could you please provide a relevant theory article related to the disturbance model with the transfer function model, if you don't mind? It would be better for me to understand it correctly and deeply.