Block Diagram Models, Signal Flow Graphs and Simplification ...

128 downloads 110 Views 2MB Size Report
Dynamic Systems and Control. Lavi Shpigelman. Block Diagram Models, ... Block Diagram Reduction. (with SISO components). +. G2. +. +. G3. G4. H3. H1. H2.
Block Diagram Models, Signal Flow Graphs and Simplification Methods

Block Diagram Manipulation Rules X1

X2

G1

X3

G2

X1 = X1

Dynamic Systems and Control Lavi Shpigelman

X1 + ± X2

G

(of rational functions ).

• Mainly relevant where there is a cascade of information flow

X3 ±

X3

G

=

X2

X1

X2

G

X1

G

= X2

X2

G

X2

variables.

• Are equivalent to a set of linear algebraic equations

+

Block Diagram Manipulation Rules

Block Diagram Models • Visualize input output relations • Useful in design and realization of (linear) components • Helps understand flow of information between internal

G

X3

G2 G1

X1

G

X1

X2

X1 =

X2

G G-1

X1

Block Diagram Reduction

Block Diagram Manipulation Rules X1

G

+

X1 +

X3 ±

=

±

X2

(with SISO components)

G

X3

G-1

U(s) +

-

X2

G1

+

-

H2 ! G4 G2

+ +

G3

G4

Y(s)

H1

+

X2

G

-

=

X1

(I+GH)-1G

X2

H3

H

Block Diagram Reduction

Block Diagram Reduction

(with SISO components)

(with SISO components) H2 ! G4

H2 U(s) +

-

G1

+

-

G2

+ +

G3

G4

Y(s)

U(s) +

-

G1

+

-

G2

H1 H3

+ +

G3 G4 H1

H3

Y(s)

Block Diagram Reduction

Block Diagram Reduction

(with SISO components)

U(s) +

-

G1

+

-

(with SISO components)

H2 ! G4 G3 G4 !!!!!! 1-G3 G4H1

G2

Y(s)

U(s) +

-

H3

Block Diagram Reduction

(with SISO components)

-

G1

+

Y(s)

H3

Block Diagram Reduction

U(s) +

G2 G3 G4 !!!!!!!!!!!! 1-G3 G4H1+ G2 G3H2

G1

(with SISO components)

H2 ! G4

-

G2 G3 G4 !!!!!! 1-G3 G4H1

H3

Y(s)

U(s) +

G1G2 G3 G4 !!!!!!!!!!!! 1-G3 G4H1+ G2 G3H2

-

H3

Y(s)

Block Diagram Reduction

Block Diagram Vs. SFG

(with SISO components)

x1 + ± U(s)

G1 G2 G3 G4 !!!!!!!!!!!!!!!!!!!!!!! 1-G3 G4H1+ G2 G3H2+ G1 G2 G3 G4H3

G

x2

x1

1

x1

• Blocks

x2

±H(s)

H

Y(s)

G(s)

Edges (aka branches)

(representing transfer functions)

• Edges + junctions

Vertices (aka nodes)

(representing variables)

Signal Flow Graphs • Alternative to block diagrams • Do not require iterative reduction to find

transfer functions (using Mason’s gain rule)

• Can be used to find the transfer function

between any two variables (not just the input and output).

• Look familiar to computer scientists (?)

Algebraic Eq representation • x = Ax + r

x1 = a11x1+a12x2+r1 x2 = a21x1+a22x2+r2

• y(s) = G(s)u(s)

G11(s) u1(s) G21(s) G12(s) u2(s) G22(s)

y1(s)

y2(s)

1

x2

Definitions

Another SFG Example a32

y2 = a12y1 + a32y3

a12 y1

y2

y3 a32

y3 = a23y2 + a43y4

a12

y5

y4

y5

a23

y1

y2

y3

a43

a32 a12

y4 = a24y2 + a34y3+a44y4

y4 a43

a23

y1

y2

a44 a34

y3

y4

y5

a24

a43

a32

y5 = a25y2 + a45y4

a12

a23

y1

y2

a44 a34

y3

a45 y4

y5

• Input: (source) has only outgoing branches • Output: (sink) has only incoming branches • Path: (from node i to node j) has no loops. • Forward-path: path connecting a source to a sink • Loop: A simple graph cycle. • Path Gain: Product of gains on path edges • Loop Gain: Product of gains on loop Loops: Loops that have no vertex • Non-touching in common (and, therefore, no edge.)

a24 a25

Mason’s Gain Rule (1956)

Input / Output • Input (source) has only

a32

outgoing edges

• Output (sink) has only incoming edges

a12 y1

a23 y2

• any variable can be

made into an output by adding a sink with “1” edge

Given an SFG, a source and a sink, N forward paths between them and K loops, the gain (transfer function) between the source-sink pair is !Pk!k Tij = !!!! !

y2 1 a12 y1

y3

a32 a23

y2

1 y3

y3

Pk is the gain of path k, ! is the “graph determinant”: ! = 1- !(all loop gains) + !(products of non-touching-loop gain pairs) - !(products of non-touching-loop gain triplets) + ... !k = ! of the SFG after removal of the kth forward path

Mason’s Rule for Simple Feedback loop

A Feedback Loop Reduces Sensitivity To Plant Variations

P1 = G(s) L1 = -G(s)H(s)

1

x1

x1

! = 1 - (-G(s)H(s))

x2

G(s)

1

x2

-H(s)

G8 G2

G3

G4

G5

G6

y(s)

-H1

-H4 -H2

Forward paths:

-H3

P1 = G1G2G3G4G5G6 P2 = G1G2G7G6 P3 = G1G2G3G4G8

Feedback loops: L1 = -G2G3G4G5H2 L4 = -G7H2G2 L7 = -G1G2G7G6H3

G(s)

G=20000 y(s)/u(s)=20000/(1+20000*0.01)=99.50

G7

G1

1

-H(s)

P1 !1 G(s) G(s) T(s) = !!! = !! = !!!!! ! ! 1+G(s)H(s)

1

u(s)

G=10000 y(s)/u(s)=10000/(1+10000*0.01)=99.01

!1 = 1

x(s)

y(s)/u(s)=G/(1+GH)

L2 = -G5G6H1 L3 = -G8H1 L5 = -G4H4 L6 = -G1G2G3G4G5G6H3 L8 = -G1G2G3G4G8H3

Loops {3,4},{4,5} and {5,7} don’t touch ! = 1-(L1+L2+L3+L4+L5+L6+L7+L8)+(L3L4+L4L5+L5L7) !1 = !3 = 1 , !2 = 1-L5 = 1 - G4H4 y(s) P1+P2!2+P3 T(s) = !! = !!!!!` x(s) !

= 0.01

1

y(s)