Determine States, State Transition Matrix

Thread Starter

jegues

Joined Sep 13, 2010
733
The problem statement, all variables and given/known data

An LTI system is given in state-space form,

\(\left( \begin{array}{cc} \dot{x_{1}} \\ \dot{x_{2}} \end{array} \right) = \left( \begin{array}{cc} -1 & 0.5 \\ 1 & 0 \end{array} \right) \left( \begin{array}{cc} x_{1} \\ x_{2} \end{array} \right) + \left( \begin{array}{cc} 0.5 \\ 0 \end{array} \right) u
\)

A unit-step signal is applied to the input of the system. If,

\(x_{1}(0) = 1, \quad x_{2}(0) = 0\)

determine the state of the system after t = 0.1 sec.

The attempt at a solution

\(\underline{\dot{x}} = \underline{A} \underline{x} + \underline{B}u\)

\(\mathcal{L} \Rightarrow s \underline{X(s)} - \underline{x(0)} = \underline{A}\underline{X(s)} + \underline{B} U(s)\)

\(\Rightarrow \underline{X(s)} = (s\underline{I} - \underline{A})^{-1} \underline{x(0)} + (s\underline{I} - \underline{A})^{-1}\underline{B}U(s)\)

Working through the simplification I obtain,

\(\left( \begin{array}{cc} X_{1}(s) \\ X_{2}(s) \end{array} \right) = \left( \begin{array}{cc} \frac{s}{s^{2}+s-0.5}\\ \frac{1}{s^{2}+s-0.5} \end{array} \right) + \left( \begin{array}{cc} \frac{0.5}{s^{2}+s-0.5}\\ \frac{0.5}{s(s^{2}+s-0.5)} \end{array} \right)
\)
Thus,
\(
X_{1}(s) = \frac{s}{s^{2}+s-0.5} + \frac{0.5}{s^{2}+s-0.5}
\)
\(
X_{2}(s) = \frac{1}{s^{2}+s-0.5} + \frac{0.5}{s(s^{2}+s-0.5)}
\)

How can I put this in a form that I can easily pull out of the s-domain back into time domain using inverse Laplace transform? If weren't for the -0.5 in the denominator I think I could work something out by reworking it into one of the following two forms,
\(
\frac{\omega}{(s+a)^{2} + \omega^{2}} \quad \text{ or } \quad \frac{s+a}{(s+a)^{2} + \omega^{2}}
\)

Any ideas? Did I make a mistake in my simplification perhaps?
 

blah2222

Joined May 3, 2010
582
I didn't really go through all the math but couldn't you just complete the square in the denominators to get it in the form you listed below:

\(

s^2 + s - \frac{1}{2} = s^2 + s + \frac{1}{4} - \frac{3}{4} = (s + \frac{1}{2})^2 - \frac{3}{4} = (s + \frac{1}{2})^2 - (\frac{\sqrt{3}}{2})^2

\)

Then for that one term in X2(s) with the s term multiplied by the quadratic term you could expand it using partial fractions then get to something you could invert.

Hope that helped.
 

Thread Starter

jegues

Joined Sep 13, 2010
733
I didn't really go through all the math but couldn't you just complete the square in the denominators to get it in the form you listed below:

\(

s^2 + s - \frac{1}{2} = s^2 + s + \frac{1}{4} - \frac{3}{4} = (s + \frac{1}{2})^2 - \frac{3}{4} = (s + \frac{1}{2})^2 - (\frac{\sqrt{3}}{2})^2

\)
The denominator in the forms I've listed has \(+\omega^{2}\) in the denominator, so that \(- (\frac{\sqrt{3}}{2})^2
\) term won't cut it.
 
Top