Scribd
Upload
193 views22 pages

Chapter 8 State Space Analysis

This document discusses state space analysis and state variable modeling of systems. It begins by explaining why state variable models are useful, including for computer-aided analysis, providing more internal information about systems, and enabling optimal design procedures. It then defines the state of a system and introduces the standard state equation and output equation forms. An example RLC circuit is modeled using different analysis methods. The document also covers formal solutions using transforms, examples of state variable modeling of different systems, and the controllable and observable canonical forms.

Uploaded by

Ali Ahmad
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
193 views22 pages

Chapter 8 State Space Analysis

This document discusses state space analysis and state variable modeling of systems. It begins by explaining why state variable models are useful, including for computer-aided analysis, providing more internal information about systems, and enabling optimal design procedures. It then defines the state of a system and introduces the standard state equation and output equation forms. An example RLC circuit is modeled using different analysis methods. The document also covers formal solutions using transforms, examples of state variable modeling of different systems, and the controllable and observable canonical forms.

Uploaded by

Ali Ahmad
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 22

Chapter 8 (page# 535) Chapter 8 (page# 535) State Space Analysis State Space Analysis

Reading assignment: Section 8.2 (page 536 to 545) 8.2.5 (page 551 to 554)

Why State-Variable Models Computer-aided analysis works better for state models than the transfer function approach. State variable model provide more internal information about the plant, allowing more complete control. Optimal (best) design procedures are mostly based on the use of state-variable models. The state variable models are required for digital simulation. System State Definition The state of a system is defined as The state of a system at any time t0 is the amount of information at t0 that, together with all inputs for t t0, uniquely defines the behavior of the system for all t t0.

State Equations Standard Form


System matrix State vector State equation Output equation

(t ) = Ax (t ) + Bu (t ) x y (t ) = Cx (t ) + Du (t )

u(t) = input vector = (n 1) vector of input functions y(t) = output vector = (p 1) vector of defined outputs.
x ( t ) is the time derivative of x(t) x(t) is the state vector, an (n 1) vector of the states A is the system matrix, an (n n) matrix of the coefficients B is an (n r) input matrix where r is the number of inputs C = (p n) output matrix D = (p r) matrix representing direct coupling between the input and output. (Neglected)

State Equation
Initial Conditions

= Ax + Bu x y = Cx
+ +

1 = s
A

x = Ax + Bu y = Cx

solution response

Example 2 RLC Circuit


R + vI N _ v1 1 L i v2 2 + + v _ C vOUT _ i1

Reference node

The system has two state variables, the inductor current (i) and the capacitor voltage (v). We can obtain the system equations by use of 1) mesh analysis, 2) nodal analysis, or 3) mixed (both) analysis. We would nor-mally choose mesh analysis since there is only one mesh; how-ever, we have a new constraint avoiding integrals. The results of each approach are given on the following page.

System Equations
Nodal Analysis Mesh Analysis Mixed v1 v IN 1 t t di di 1 + ( v v ) dt = 0 v IN + iR + L + v = 0 1 2 v IN + iR + L + idt = 0 R L 0 dt dt C 0 dv 2 1 t dv 1 t ( v v ) dt + C =0 v i + C =0 2 1 idt OUT = L 0 dt dt C 0

All three results are valid models but only the equations obtained from the mixed analysis can be represented as state equations in the standard form. The state equations are given below:

di R 1 1 = i v + v IN dt L L L dv i = dt C
This form is frequently used in digital simulation.

State Equations
The state equations, in standard form, for the series RLC circuit are:

R i L = 1 C v C vOUT = v

1 i 1 L + L vIN v 0 C 0

Solution Using Transforms


vOUT(t) can be obtained by taking the inverse:
1 LC VOUT ( s ) = 2 = R 1 s ( s + 0.5 + j 0.866)( s + 0.5 j 0.866) s + s+ L LC VIN ( s )

The solution for vOUT(t) will have the form

vOUT (t ) = V0 + V1e ( 0.5 + j 0.866)t + V2e ( 0.5 j 0.866)t


The constants V0, V1, and V2 are evaluated using partial fractions:
1 V0 = =1 (0.5 + j 0.866)(0.5 j 0.866) ,
V1 =

1 = 0.5 + j 0.2887 (0.5 j 0.866)( j1.732)

V2 =

1 = 0.5 j 0.2887 (0.5 j 0.866)( j1.732)

vOUT (t ) = 1 e 0.5t cos(0.866t ) 0.5774e 0.5t sin(0.866t )

Formal Solution We may use the standard form directly:


The characteristic equation for a i i 1 L + L vIN set of state equations is given by = 1L C vC 0 v 0 |I A| where A is the system C matrix, I is the identity matrix, and the values are the eigenvalues. The matrix |I A| 1 R 1 R is given by s + 1 0 L L L L I A = = 1 1 0 1
C 0 C s
R 1 2 I A = + + =0 Taking the determinant of this matrix yields L LC

This agrees with the previous results (the denominator of the trans-form).

State variable Modeling

d2y dy M +B + Ky = f (t ) 2 dt dt
d2y dy M = B Ky + f (t ) 2 dt dt

1 G(s) = Ms 2 + Bs + K Let x1 (t ) = y

1 = x2 x 2 = Bx2 Kx1 + f Mx
B K f 2 = x x2 x1 + M M M
1 0 x = k x 2 M y = [1 0] x 1 0 B x + 1 f M M

dy (t ) dx(t ) 1 (t ) x2 (t ) = = =x dt dt dx2 (t ) dy 2 (t ) 2 = x = dt dt 2

Example: Consider the system described by the coupled differential equations

1 + k1 y 1 + k 2 y1 = u1 + k3u2 y

2 + k 4 y2 + k5 y 1 = k6u1 y
u1 & u2 are inputs, y1 & y2 are outputs. ki ; i=1,.6 are system parameters.

Let x1 = y1 1 = x 1 x2 = y x3 = y2 2 = k 2 x1 k1 x2 + u1 + k3u2 x 3 = k5 x2 k 4 x3 + k 6u1 x


output equation y1 = x1 y2 = x3
1 0 1 x x 2 = k 2 k1 3 k5 x 0 1 0 0 y= x 0 0 1 0 0 1 0 x + k4 k 6 0 k3 f 0

Simulation Diagram

Given

0 1 x( t ) = x( t ) + 8 2 y( t ) = [ 4 1] x;

1 u( t ) 1

Control canonical (Controllable) & Observer Canonical (Observable) Matrix A,B & C are constructed from the transfer function. Next - Examples

Next slides consider general standard Transfer Function

Y ( s) b2 s + b1s + b0 = 3 U ( s ) s + a2 s 2 + a1s + a0
2

b2 b1 b0 + 2+ 3 Y ( s) s = s s U ( s ) 1 + a2 + a1 + a0 s s2 s3

Y G = U 1 + GH

Dividing each term by the highest order of s yields is canonical form

& x 0 1 x = 0 & 2 & 3 x a0 y = [ b0 b1

1 0 a1

0 x1 0 1 x2 + 0 u a2 x3 1

x1 b2 ] x2 + [ 0] u x3

Control canonical form block diagram.

Example

b2 b1 b0 2 8 6 + 2+ 3 + 2+ 3 C ( s) 2 s 2 + 8s + 6 s s s = s s s = 3 = R( s ) s + 8s 2 + 26s + 6 1 + a2 + a1 + a0 1 + 8 + 26 + 6 s s2 s3 s s2 s3 The form is:


is controllable canonical form

& 1 0 x1 0 x 0 1 x =0 x + 0 r & 0 1 2 2 & 3 x 6 26 8 x3 1 x1 y = [ 6 8 2 ] x2 + [0]r. x3 System controllable but not observable. Why? Ans: State model is based on controllable canonical form, this can be confirmed by another method later in the next lecture.

Y ( s) b2 s + b1s + b0 = 3 U ( s ) s + a2 s 2 + a1s + a0
2

b2 b1 b0 + 2+ 3 Y ( s) s = s s U ( s ) 1 + a2 + a1 + a0 s s2 s3

Y G = U 1 + GH

called observable canonical

a X 1 2 X 2 = a1 a X 3 0

Y = [1 0 0]

1 0 X1 b 0 0 1 X 2 + b1 u 0 0 X3 b 2

Observer canonical form block diagram.

called controllable canonical

called observable canonical

0 X 1 X 2 = 0 a X 3 0 Y = [b 0 b1

b2 ]

1 0 a1

0 X1 0 1 X 2 + 0u a2 X3 1

a X 1 2 X 2 = a1 a X 3 0

Y = [1 0 0]

1 0 X1 b 0 0 1 X 2 + b1 u 0 0 X3 b 2

Controllable but not observable

observable but not controllable

G(s) =

s +1 s +1 1 = 2 = (s + 1)(s + 3) s + 4s + 3 s + 3

Controllable canonical

Observable canonical

0 x = 3

1 0 x + u 4 1

0 = x 1

3 1 x + u 4 1

x 1 y = [1 1] x 2

x1 y = [ 0 1] x 2

You might also like