Higher order systems

Higher than first order systems can be built with multiple integrators, as shown here for a third order system:

opam7.gif

From the diagram, we can get

  $\displaystyle \left\{ \begin{array}{l}
Y_3(s)=Y_2(s)/s \Longrightarrow Y_2(s)=...
... Y_1(s)=Y_0(s)/s \Longrightarrow Y_0(s)=Y_1(s)s=Y_3(s)s^3
\end{array} \right.
$ (106)
But we also have
  $\displaystyle Y_0(s)=X(s)-(k_1Y_1(s)+k_2Y_2(s)+k_3Y_3(s)) $ (107)
i.e.,
  $\displaystyle X(s)=Y_0(s)+k_1Y_1(s)+k_2Y_2(s)+k_3Y_3(s)=(s^3+k_1s^2+k_2s+k_3) Y_3(s) $ (108)
we get the transfer function
  $\displaystyle H(s)=\frac{Y_3(s)}{X(s)}=\frac{1}{s^3+k_1s^2+k_2s+k_3}
$ (109)

Second order system by 2 integrators

opam8.gif

From the diagram, we can get

  $\displaystyle \left\{ \begin{array}{ll}
Y_2(s)=-c_2Y_1(s)/s \Longrightarrow Y_...
...^2Y_2(s)/c_1c_2 \\
Y_0(s)=k_0 X(s)+k_1Y_1(s)+k_2Y_2(s)
\end{array} \right.
$ (110)
substituting the first two equations into the last one, we get
  $\displaystyle \frac{s^2}{c_1c_2} Y_2(s)=k_0X(s)+k_1(-\frac{s}{c_2})Y_2(s)+k_2Y_2(s) $ (111)
from which we obtain the transfer function as
  $\displaystyle H(s)=\frac{Y_2(s)}{X(s)}=\frac{k_o}{\frac{s^2}{c_1c_2}+\frac{s}{c_2}s-k_2}
=\frac{k_oc_1c_2}{s^2+k_1c_1s-c_1c_2k_2}
$ (112)
which is a second order system. In particular, if $c_1=c_2=c$, we have
  $\displaystyle H(s)=k_0\frac{c^2}{s^2+c k_1s-k_2c^2}
$ (113)
Comparing this with the canonical 2nd order system transfer function
  $\displaystyle H(s)=\frac{\omega_n^2}{s^2+2\zeta \omega_n s+\omega_n^2}
$ (114)
we see that we can let $c=\omega_n$ and $k_1=2\zeta$. Moreover, $k_2<0$, i.e., the feedback from the output should be negative. $k_0$ is a constant scalar which can take any value.