STABILITY AND ROOT LOCUS TECHNIQUES:

(stability concepts - conditions for stability - routh, hurwitz stability criterion –relative
stability analysis – root locus concepts – construction of root loci – effect of root pole
additions on root loci – root contours – systems with transportation lag)

 STABILITY:
 Small change in the system input(in initial conditions/system parameters)
doesn’t result in large change in output.
 A liner time invariant system  stable if the following is satisfied
 If input = Bounded , output = Bounded
 If input = 0 , output = 0 (irrespective of initial conditions)
ASYMPTOTIC STABILITY

 UNSTABILITY:
 If input = Bounded , Output = Unbounded
 Output of unstable system  increase to a certain limit(non linear) and then
may break down
 For nonlinear systems  defining stability  difficult
(Reason : Existence of multiple equilibrium states)

 DETERMINING WHETHER A SYSTEM IS STABLE / UNSTABLE:

 Determining the roots of the system’s characteristic equation
 ABSOLUTELY STABLE with respect to a parameter  system is stable for all
values of this parameter
 CONDITIONALLY STABLE with respect to a parameter system is stable only for
certain bounded ranges of values of this parameter
 Root moves farther away from imaginary axis – improved relative stability of
system
 INFERENCE:
 Roots of characteristic equation have –ve real parts  STABLE---1
 Roots of characteristic equation have +ve real parts UNSTABLE
 Repeated root on “jw” axis UNSTABLE
 1 + (presence of one / more non repeated roots on “jw”
axis)LIMITEDLY
STABLE
  Stability can be established by determining the roots of characteristic equation
 Relative stability
 quantitative measure of stability of a system
 how fast the transients die out in a system
 important criteria to access stability
 measured by relative settling time of each root

NECESSARY CONDITIONS FOR STABILITY:

 all coefficients of characteristic equation – real & of same sign

 no coefficient = 0
 absence/ negativeness of any coefficient of characteristic equation – unstable/partially
stable
 1st step in analyzing stability of a system – examine characteristic equation

CRITERION TO FIND STABILITY(SUFFICIENT CONDITIONS):

 HURWITZ STABILITY CRITERION(in terms of determinants):

 For stability of a “n”th order system – “n” determinants formed from coefficients
of characteristic equation – positive
 ROUTH /CRITERION(in terms of array formulation):
  based on ordering the coefficients of the characteristic equation into an
arrayRouth array.
 Necessary and sufficient condition for stability each element of 1st column of
routh array- +ve
 Unstable – sign change of elements of 1st column
 No of sig changes – no of roots of characteristic equation in right half of s-
plane(UNSTABLE)
 DIFFICULTY:
 If any element in 1st column = 0
 Unable to proceed – presence of infinity in the successive
elements of 1st column
 SOLUTION:
Replace that 0 by a constant(say β) – proceed with
building the routh array – finally after finishing routh array
construction replace β by 0 – find stability of system
Replace “s” by “1/z” – proceed with routh array
construction – no of sign changes indicate the no of poles
of system in right side of z-plane
 if any row of routh array = 0
 represents the existence of symmetrical roots in s-plane
 SOLUTION:
Coefficients of the row previous to 0-row forms auxillary
polynomial(always of even order) provides location and
no of the summetrical roots
Replace the coefficients of 0-row by coefficients of
polynomial “x”
X – derivative of auxillary polynomial
o If symmetrical roots = complex
conjugate(oscillatory nature)  results in
oscillations
 LIMITATION:
 Doesn’t provide any information about values of roots
 Doesn’t distinguise between real and complex roots

EQUIVALENCE OF HURWITZ AND ROUTH CRITERION:

 Positiveness of Hurwitz determinantpositiveness of elements of 1st column of routh

array
RELATIVE STABILITY:

if a system  stable

 Find relative stability by determining the settling time of the dominant roots of its
characteristic equation
 Dominant roots :
 roots influencing the transient response of system
 closest to imaginary axis in left half of s-plane
 decide the speed of system
 governs overshoot, settling time , damping
 shape the system’s behavior
 HOW TO IDENTIFY:
For real roots – one with smallest magnitude
For complex conjugate roots – one with smallest magnitude real part


CONSTRUCTION OF ROOT LOCUS:

 A transfer function will be given

 Locate poles and zeros from the given transfer function
 Plot the poles (X) and zeros(0) in the graph of s plane
 Find root locus on real axis
 Use a test point
 A root locus will start at a pole and will end at a zero / infinity
 Randomly choose test point position .
 If the no of poles to the right of test pointodd root locus exists
 If the no of poles to the right of test pointeven root locus doesn’t exists
 find asymptodes(lines passing parallelly very close to axes till infinity without touching
the axes )
 no of asymptodes = no of poles
 find angle of asymptodes
− + 𝟏𝟖𝟎(𝟐𝒒+𝟏)
 𝒂𝒏𝒈𝒍𝒆 𝒐𝒇 𝒂𝒔𝒚𝒎𝒑𝒕𝒐𝒅𝒆𝒔 = 𝒏𝒐 𝒐𝒇 𝒑𝒐𝒍𝒆𝒔−𝒏𝒐 𝒐𝒇 𝒛𝒆𝒓𝒐𝒔
Where q0, 1,…..,(no of poles-no of zeros-1)
 find the centroid
 asymptodes starts from the centroid with this angle
𝒔𝒖𝒎 𝒐𝒇 𝒑𝒐𝒍𝒆𝒔−𝒔𝒖𝒎 𝒐𝒇 𝒛𝒆𝒓𝒐𝒔
 𝒄𝒆𝒏𝒕𝒓𝒐𝒊𝒅 = 𝒏𝒐 𝒐𝒇 𝒑𝒐𝒍𝒆𝒔−𝒏𝒐 𝒐𝒇 𝒛𝒆𝒓𝒐𝒔
 Mark the centroid in graph
  Keeping centroid as a center , mark the asymptode angles
 Join the centroid and marking of asymptode angles with dotted line
 Find break away point and break in point
 Break away point  in between 2 adjacent poles – root locus starting from the
adjacent poles meets at a point inbetween them
 Break in point in between 2 adjacent zeros - root locus ending at the adjacent
zeros meets at a point inbetween them
 To find need characteristic equation of system
 Write the characteristic equation interms of “k”
 Differentiate “k” with respect to “s” and equate to 0
 The answer must be a point on the root locus
 Find the crossing point on imaginary axis
 The point on the imaginary axis at which the root locus starting from the pole
cuts the imaginary axis and moves parallel to the asymptodes to infinity
 Presence of complex poles  angle of departure
 𝒂𝒏𝒈𝒍𝒆 𝒐𝒇 𝒅𝒆𝒑𝒂𝒓𝒕𝒖𝒓𝒆 = 𝟏𝟖𝟎 + 𝑿 − 𝒀
 X= 𝒔𝒖𝒎 𝒐𝒇 𝒂𝒏𝒈𝒍𝒆𝒔 𝒕𝒉𝒆 𝒄𝒐𝒎𝒑𝒍𝒆𝒙 𝒑𝒐𝒍𝒆 𝒎𝒂𝒌𝒆𝒔 𝒘𝒊𝒕𝒉 𝒓𝒆𝒎𝒂𝒊𝒏𝒊𝒏𝒈 𝒑𝒐𝒍𝒆𝒔
 Y= 𝒔𝒖𝒎 𝒐𝒇 𝒂𝒏𝒈𝒍𝒆𝒔 𝒕𝒉𝒆 𝒄𝒐𝒎𝒑𝒍𝒆𝒙 𝒑𝒐𝒍𝒆 𝒎𝒂𝒌𝒆𝒔 𝒘𝒊𝒕𝒉 𝒓𝒆𝒎𝒂𝒊𝒏𝒊𝒏𝒈 𝒛𝒆𝒓𝒐𝒔
 Presence of complex zeros  angle of arrival
 𝒂𝒏𝒈𝒍𝒆 𝒐𝒇 𝒂𝒓𝒓𝒊𝒗𝒂𝒍 = 𝟏𝟖𝟎 + 𝑿 − 𝒀
 Y= 𝒔𝒖𝒎 𝒐𝒇 𝒂𝒏𝒈𝒍𝒆𝒔 𝒕𝒉𝒆 𝒄𝒐𝒎𝒑𝒍𝒆𝒙 𝒑𝒐𝒍𝒆 𝒎𝒂𝒌𝒆𝒔 𝒘𝒊𝒕𝒉 𝒓𝒆𝒎𝒂𝒊𝒏𝒊𝒏𝒈 𝒑𝒐𝒍𝒆𝒔
 X= 𝒔𝒖𝒎 𝒐𝒇 𝒂𝒏𝒈𝒍𝒆𝒔 𝒕𝒉𝒆 𝒄𝒐𝒎𝒑𝒍𝒆𝒙 𝒑𝒐𝒍𝒆 𝒎𝒂𝒌𝒆𝒔 𝒘𝒊𝒕𝒉 𝒓𝒆𝒎𝒂𝒊𝒏𝒊𝒏𝒈 𝒛𝒆𝒓𝒐𝒔

LAG COMPENSATOR:

 Given G(s),Phase margin,ess(steady state error)

 Find gain value k from kv
 Velocity error constant kv=1/ess
 Kv=lim s0 sG(s)H(s)
 Put s=jw
 Draw table
 Plot magnitude plot and phase plot
 Find phase margin
 found phase margin less than given
 gamma required=gamma desired + epsilon
epsilon5/10
  phi = gamma required-180
 draw a line from phi to phase plot
 meets at a frequency wgcn
 for lag compensator find beta
 draw a line from wgcn cuts gain plot at a valueagcn
 agcn=20logbeta
 find beta
 find zero of lag compensator
 z=1/T=wgcn/10
 find T
 equation for transfer function  (s+(1/T))/(s+(1/bT))beta(1+sT)/(1+sTbeta)

LEAD COMPENSATOR:

 same process till plotting phase plot and magnitude plot

 got phase margin < required
 find max phase lead angle (phi-m)=desired phase margin-got phase margin
+epsilon
 alpha = (1-sin phi-m)/( 1+sin phi-m)
 find magnitude corresponding to this max lead angle
 m=-20log(1/(root alpha))
 draw a line from m to magnitude plot get corresponding omega
 find T=1/(m(root aplha))
 find transfer function of lead compensator=(s+(1/T))/(s+(1/alpha
T))=alpha(1+sT)/(1+sTalpha)
