Bec403 M4

Control Systems (BEC403)
Module- 4:
STABILITY ANALYSIS AND ROOT LOCUS
4.1. Stability:
• A system is stable of the output of the system is under control
Absolutely Stable Unstable Marginally stable
In an LTI system, Bounded output Unbounded output Undamped/Sustained
for a Bounded oscillation with constant
input system gives amplitude and frequency
No input No output
Roots of All roots have -ve If one of root has +v real part or Non-repeated purely
characteristic real part repeated purely imaginary roots imaginary roots
equation
Poles location in All poles are at left If one pole is at right side of Overlapping poles on
s-Plane side of imaginary imaginary axis or overlapping poles imaginary axis
axis on imaginary axis
• Conditionally stable system gives bounded output up to some threshold input

• Zero -input/Asymptotic stability:
o Sometimes system gives the output based on the initial conditions without inputs.
o If the zero input response of the system subjected to finite initial conditions reaches to zero as time
approaches infinity the system is said to be zero input stable otherwise it is called zero input unstable.
o |c(t)|≤M≤∞ for all t≥𝑡0 and lim |𝑐(𝑡)|=0
𝑡→∞
o BIBO stable system must be asymptotic stable

• Relative stability
ECE, GMIT, Bharathinagara Page 1

o System is said to be relatively more stable if settling time for that system is less than that of other
systems
o Lesser the settling time system is more stable
o Relative stability increases as the closed loop poles moves away from left half of imaginary axis of s-
plane
Note:
𝐶(𝑠) 𝑏0 𝑠𝑚 +𝑏1 𝑠𝑚−1 +⋯..+𝑏𝑚
• Polynomial form of Transfer function , 𝑅(𝑠)= 𝑎0 𝑠𝑛 +𝑎1 𝑠𝑛−1 +⋯..+𝑎𝑛
• Characteristics equations, 𝑎0 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + ⋯ . . +𝑎𝑛 =0

4.2. Necessary but insufficient conditions for stability (i.e. no roots in right of s-plane)
• All the coefficients of characteristics have same sign
• No coefficients of s should not vanish
4.3. Hurwitz Criterion:
• Necessary and sufficient conditions for roots to be in left of s-plane is all the sub-determinants Dk
(k=1,2,3,…,n)formed by Hurwitz determinant H, must be positive.
𝑎0 𝑎2 𝑎4 … . 𝑎2𝑛−1
𝑎1 𝑎3 𝑎5 … . 𝑎2𝑛−2
|0 𝑎0 𝑎2 𝑎4
𝑎0 𝑎2 … . 𝑎2𝑛−3 | 𝑎0 𝑎2
• H= 0 𝑎1 𝑎3 … . 𝑎2𝑛−4 | , D1=|𝑎0 | , D2=|𝑎1
𝑎
𝑎3 | , D3=| 1
𝑎3 𝑎5 |
| 0 𝑎0 𝑎1
⋮ ⋮ ⋮ …. ⋮
0 0 0 … … … … 𝑎 𝑛
• Disadvantages:
o For higher order systems solving higher order determinants is tedious
o Roots contributed to instability cant be found
o Marginal stability cant be found
4.4. Routh-Hurwitz Criterion:
• Routh’s array from characteristic equation
𝑠𝑛 𝑎 0 𝑎2 𝑎4
𝑠 𝑛−1 𝑎1 𝑎3 𝑎5
𝑠 𝑛−2 | 𝑏1 𝑏2 𝑏3
|
⋮ 𝑐1 𝑐2 𝑐3
𝑠0 𝑑1 𝑏2 𝑏3

𝑎1 𝑎2 −𝑎0 𝑎3 𝑎1 𝑎4 −𝑎0 𝑎5 𝑎1 𝑎6 −𝑎0 𝑎7

• 3rd row: 𝑏1 = , 𝑏2 = , 𝑏3 = ,……
𝑎1 𝑎1 𝑎1
𝑏1 𝑎3 −𝑎1 𝑏2 𝑏1 𝑎5 −𝑎1 𝑏3 𝑏1 𝑎7 −𝑎1 𝑏4
• 4th row: 𝑐1= , 𝑐2 = , 𝑐3 = ,……
𝑏1 𝑏1 𝑏1
• The necessary and sufficient conditions to make the system stable( roots to be in left of s-plane) is all the
terms in the first column of Routh’s array must have same sign i.e. no sign change.
• If the sign change
o System is unstable
o No. of sign changes equals number of roots lying in right half of s-plane
• Special Case-1: if one of the term in first column is ‘0’ it should be replaced by constant ‘Ɛ’
𝑠𝑛 𝑎0 𝑎2 𝑎4 𝑠𝑛 𝑎0 𝑎2 𝑎4
𝑠 𝑛−1 𝑎1 𝑎3 𝑎5 𝑠 𝑛−1 𝑎1 𝑎3 𝑎5
𝑠 𝑛−2 | 𝑏1 𝑏2 𝑏3 ➔ 𝑠 𝑛−2 | 𝑏1 𝑏2 𝑏3
| |
⋮ 0 𝑐2 𝑐3 ⋮ Ɛ 𝑐2 𝑐3
⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮
• Special Case-2: if one of the row is completely ‘0’ then
o form an auxiliary equation using the row above zeroth row
o Differentiate auxiliary equation
o The coefficients of differentiated auxiliary equation are used to replace zeroth row
𝑠𝑛 𝑎0 𝑎2 𝑎4 Auxiliary equation, 𝑠𝑛 𝑎0 𝑎2 𝑎4
𝑠 𝑛−1 𝑎1 𝑎3 𝑎5 𝑠 𝑛−1 𝑎1 𝑎3 𝑎5
𝑠 𝑛−2 || 𝑥 𝑦 𝑧 ➔ A(s)=0 → xsn-2+ysn-4+zsn-6+..=0 ➔𝑠 𝑛−2 || 𝑥 𝑦 𝑧
⋮ 0 0 0 ⋮ 𝑝 𝑞 𝑟
⋮ ⋮ ⋮ Differentiate Auxiliary equation ⋮ ⋮ ⋮
⋮ ⋮
dA/dS=0 → psn-1+psn-3+rsn-5+..=0
• RH Criteria is used
o To find relative stability
o Determine the range of system gain K
o Marginal K
o Frequency of sustained oscillations
4.5. Root Locus:
• Locus formed by the roots of the characteristic equation by varying system gain K from zero to infinity.
• characteristic equation of the closed loop control system: 1+G(s)H(s)=0 solving this gives the set a roots
• G(s) can have system gain ‘K’ and if this K is made variable then different set of poles (or roots) will be obtained
• Angle and Magnitude condition:
o 1+G(s)H(s)=0
o G(s)H(s)=-1
o Angle condition: ⌊G(s)H(s) = ±(2n+1) π= ±(2n+1)180o=±odd (π or 180o)
o Magnitude condition: |G(s)H(s)| = 1

o Any point on s-plane which satisfies the angle condition will be on root locus
o After the confirmation of existence of point on root locus K can be found by
o Method 1: using magnitude condition at a point
Product of phasor length between a point to all the poles
o Method 2: K can also be found by K= Product of phasor length between a point to all the pole
• Rules of Root Locus: P=No. of poles, Z=No. of zeros, N=No. of branches

1. Rule 1: Root locus is always symmetric about real axis
2. Rule 2: Branches in root locus always starts from pole and ends at zero.
No. of root Locus branches (N) Pole-Branch-Zero
P=Z P or Z N branches starts at pole and ends at zero
P>Z P (P-Z) branches lead to infinity
P<Z Z (Z-P) branches start from infinity and reaches
3. Rule-3: A point on real axis is a part of root locus if there are odd number of poles and zeros to the right side of
the point.
Note:
• Points considered to check the part of root locus should be between the poles or zeroes and not points on
Poles and zeroes
• Asymptote: Is a path/guideline leading to infinity
(2n+1)π
4. Rule-4: Asymptote are symmetric about real axis and Angle of Asymptote= where n=0,1,2,…,(P-Z-1)
P−Z
∑ Real part of poles−∑ Real Parts of zeros
5. Rule-5: All the asymptotes intersect the real axis at centroid, σ = P−Z
Note:
• σ is always real value and may/may not be a part of root locus
• Breakaway points:
o Point on the root locus where multiple poles of characteristic equation occur for one K
o Point from where branches break into complex from real
o Break-in point: Point from where branches break into real from complex
180𝑜
6. Rule-6: Root Locus branches always leave breakaway points at an angle of ± where n→No. of branches
𝑛
approaching at breakaway points

6a. Breakaway points may exist
• Between adjacently placed poles on real axis which is part of root locus
• Between adjacently placed zeros on real axis which is part of root locus
• To the left of zero which lies on real axis and no poles and zeros to the left of this zero
6B. Steps to find breakaway points:

• Let the characteristics equation, 1+G(s)H(s)=0
• Separate the terms involving K and ‘s’ and K=F(s)
𝑑𝐾
• Differentiate above equation w.r.t. ‘s’ equates to zero, 𝑑𝑠 =0
• Roots of the above gives breakaway points

7. Rule-7: Intersection points of root locus with imaginary axis are roots of auxiliary equation while solving RH
criteria for Kmar
Note: Root locus intersects with imaginary axis if Kmar is + ve and does not if Kmar is -ve then root locus
8. Rule-8: Angle at which branch departs from complex pole is called angle of departure, d=180o-
Angle at which branch arrives to complex zero is called angle of arrival, a =180o+
Here, = ∑ ∅p - ∑ ∅z
Where, ∑ ∅p → Sum of angles made by all poles with the pole where d is calculated
∑ ∅z → Sum of angles made by all poles with the pole where d is calculated
4.6. Procedure to sketch Root locus:
1. Find G(s)H(s)
2. Find P, Z, N, |P-Z|
3. Find the section on real axis which are part of root locus (using Rule-3) and confirm no. of breakaway points
(using 6A)
4. Find angle of asymptotes (Rule-4)
5. Find the centroid (Rule-5)
6. Find breakaway points (Rule-6B)
7. Find intersection of root locus with imaginary axis (Rule-7)
8. If there are complex poles or zeroes find angle of departure/arrival (Rule-8)
9. Sketch Root locus
10. Check the stability using root locus

Bec403 M4

Uploaded by

Copyright:

Available Formats

Bec403 M4

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bec403 M4

Uploaded by

Copyright:

Available Formats

Control Systems (BEC403)

• Conditionally stable system gives bounded output up to some threshold input

o BIBO stable system must be asymptotic stable

ECE, GMIT, Bharathinagara Page 1

• Characteristics equations, 𝑎0 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + ⋯ . . +𝑎𝑛 =0

ECE, GMIT, Bharathinagara Page 2

𝑎1 𝑎2 −𝑎0 𝑎3 𝑎1 𝑎4 −𝑎0 𝑎5 𝑎1 𝑎6 −𝑎0 𝑎7

o Magnitude condition: |G(s)H(s)| = 1

• Rules of Root Locus: P=No. of poles, Z=No. of zeros, N=No. of branches

approaching at breakaway points

6B. Steps to find breakaway points:

• Roots of the above gives breakaway points

ECE, GMIT, Bharathinagara Page 5

You might also like