How can I extract the coeffs. of a first order TF model?

Hello everyone,

I have a very simple system (b/(z+a)) implemented in SIMULINK.

In the DAQ (data aqc.) phase I have 2 arrays 1000x1 (u, y) for which I want to calculate using OE, ARX and ARMAX functions the coefficents of the system. The idea is that I already know the coeffs (a, b) and I will compare the results or the estimated errors.

What I am strugling to do is to extract the coeffs of the functions.

So for example when I run: oe(Date_pt_estimare, [nb nf nk]); -> I get:

est_oe =
Discrete-time OE model:  y(t) = [B(z)/F(z)]u(t) + e(t)
  B(z) = 4.9 z^-1                                     
                                                      
  F(z) = 1 + 0.69 z^-1                                
                                                      
Sample time: 0.001 seconds
  
Parameterization:
   Polynomial orders:   nb=1   nf=1   nk=1
   Number of free coefficients: 2
   Use "polydata", "getpvec", "getcov" for parameters and their uncertainties.

Status:                                                   
Estimated using OE on time domain data "Date_pt_estimare".
Fit to estimation data: 100%                              
FPE: 1.917e-31, MSE: 1.904e-31    

So what I want to do is to extract the coeff. of B(z) and F(z). Something like:

Extracted_param1 = function(oe(Date_pt_estimare, [nb nf nk], 1); 0< Extracted_param = 4.9

Extracted_param2 = function(oe(Date_pt_estimare, [nb nf nk], 1); 0< Extracted_param = 0.69

I hope I made me clear!

Below you can find my code.

%% Date initiale
h = 1e-3; % [ms] > pasul de esantionare
K = 29; % [-] > nr. catalog
N = 1000; % [-] > nr. esantioane
a = (0.4+0.01*K); 
b = (2+0.1*K);
fprintf('Valoarea parametrilor este: a = %.2f si b = %.2f, pentru numarul de la catalog K = %d\n', a, b, K);

%% 2.Determinarea oridnelor caracteristice sistemului
na = 1;
nb = 1;
nc = 1;
nf = 1;
nk = 1;

%% 3.Prelucrarea datelor achizitionate
Date_pt_estimare = iddata(y, u, h);

%% 4.Estimarea parametrilor sistemului discret
est_oe = oe(Date_pt_estimare, [nb nf nk]);
est_arx = arx(Date_pt_estimare, [na nb nk]);
est_armax = armax(Date_pt_estimare, [na nb nc nk]);

Answer 1

[Voicila Iulian-Teodor]

Solved!

est_oe.B(2), est_oe.F(2), est_arx.B(2), est_arx.A(2), est_armax.B(2), est_armax.A(2)