AreParamsTunable (true/false) : true
Options for 'numericalpert' algorithm
NumericalPertRel : 1.000000e-05
NumericalXPert : []
NumericalUPert : []
The following command performs a linearization using the zero-order hold method.
Because the linearization includes the Zero-Order Hold block, the sample time of the
linearization is 0.01.
sys_zoh = linearize(model,io,opt);
The following commands change the rate conversion method to the Tustin (Bilinear
transformation) method and then linearize using this method. The sample time of this
linearized model is also 0.01.
opt.RateConversionMethod = 'tustin';
sys_tust = linearize(model,io,opt);
It is also possible to create a continuous-time linearized model by specifying the sample
time as 0 in the options object. The rate conversion method still creates a discrete-time
linearized model but then converts the discrete-time model to a continuous-time model.
opt.SampleTime = 0;
sys_c = linearize(model,io,opt);
The Bode plots for the three linearizations show the effects of the two rate conversion
methods. In this example, the Tustin rate conversion method gives the most accurate
representation of the phase response of the continuous system and the zero-order hold
gives the best match to the magnitude response.
p = bodeoptions('cstprefs');
p.YLimMode = {'manual'};
p.YLim = {[-100 0];[-180 -30]};
p.Grid = 'on';
bodeplot(sys_c,sys_zoh,sys_tust,p);
h = legend('sys_c','sys_zoh','sys_tust','Location','SouthWest');
h.Interpreter = 'none';
Linearization Using Different Rate Conversion Methods