3.6. H2AND HANALYSISAND SYNTHE SIS 105
g3 = ctrlplot(step,g3,{line style=2});
g3 = plot(g3,{title="Step responses",...
legend=["Kinf";"K2";"input"]})?
24680 10
0.2
0.4
0.6
0.8
1
0
1.2
Step responses
Kinf
K2
input
3.6.2 System Norm Calculations
Functions are providedfor ca lculating the H2andHnorms of Dynamic Systems. In
the H2case, this involves the solution of a Lyapunov equation. A bisection method,
involving the calculation of eigenvalues of a scaled Hamiltonian matrix, is required for
the Hnorm calculation.
The Xµfunction for the two norm calculation is called h2norm. The syntax and
operation are self explanatory.
The calculation of the Hnorm involves the iterative solution of a Riccati equation.
The technique is a generalization of the the theoretical result given in Lemma 2 in
Section 2.3.3. As a result, a tolerance can be sp ecified, and the calculation gives upper
and lower bounds. The function is hinfnorm. The syntax is illustrated below.
[out,omega] = hinfnorm(system,tol)