Design of Lag-Lead compensator using Root Locus method

S. Senthil Kumar, Assistant Professor

Dept. of Aeronautical Engineering, KCT, Coimbatore 641049

Lag-Lead compensator
 Combination of Lead and Lag compensators.
 Can be implemented using passive components (simple RLC circuit) as an alternative to the
ideal PID controller (active circuits).
 The purpose of Lead compensator is to improve the transient response.
 The purpose of Lag compensator is to improve the steady-state response (to reduce the steady-
state error).
 Hence, the purpose of Lag-lead compensator is to improve both the transient response and the
steady-state response.
 We have to improve steady-state error and transient response independently.
 For an approach, we can improve the steady-state error first and then follow with the design to
improve the transient response. A disadvantage of this approach is that the improvement in
transient response in some cases yields some decay in the improvement of the steady-state error,
which was designed first.
 In other case, the improvement in transient response yields further improvement in steady-state
errors. Thus, a system can be overdesigned with respect to steady-state errors (Overdesign is
usually not a problem unless it affects cost or produces other design problems).
 Here, I am going to design compensator for transient response first and then design for steady-
state error.
 First, the lead compensator has to be designed to improve the transient response. Next, the
improvement in steady-state error still required has to be evaluated. Finally, we have to design
the lag compensator to meet the steady-state error requirement.

Design of Lag-Lead compensator

The following steps summarize the design procedure for a lag-lead compensator:
1. Evaluate the performance of the uncompensated system to determine how much improvement in
transient response is required.
2. Design the lead compensator to meet the transient response specifications. The design includes
the zero location, pole location, and the loop gain.
 3. Simulate the system to be sure all requirements have been met.
4. Redesign if the simulation shows that requirements have not been met.
5. Evaluate the steady-state error performance for the lead-compensated system to determine how
much more improvement in steady-state error is required.
6. Design the lag compensator to yield the required steady-state error.
7. Simulate the system to be sure all requirements have been met.
8. Redesign if the simulation shows that requirements have not been met.

Lag-Lead compensator design: Example

K
Design a lag-lead compensator for the system given in Figure 1 with G  s   so that the
s  s  5 s  11

system will operate with 30% overshoot and a reduction in peak time by a factor of 2. Further, the
compensated system will exhibit a 30 times reduction in steady-state error for a ramp input.

Figure 1: Uncompensated system for the given example

Step 1: First we evaluate the performance of the uncompensated system in MATLAB.

Figure 2: Root locus for uncompensated system

Searching for the 30% overshoot in Figure 3 (zoomed diagram of Figure 2), we find the dominant poles
at –1.45 ± j 3.78 with a gain of 215, and the corresponding damping ratio  = 0.358.

Figure 3: Root locus for uncompensated system (zoomed for finding 30% overshoot)

Now, the step response of the uncompensated system with the gain value of 215 is shown in Figure 4.

Figure 4: Step response of the uncompensated system with K = 215

From Figure 4, the peak time is observed as 0.919 seconds. (Theoretically, the peak time can be
calculated from the imaginary part of the dominant pole (that is, 3.78j). We know that the dominant

poles are given by the roots n  jd , where d  n 1   2 . Now, the peak time (Tp) is given by

 
Tp    0.831 seconds). Also, the steady-state error for the uncompensated system with gain
d 3.78
value of 215 for the unit ramp input is shown in Figure 5 (obtained using MATLAB lsim command) and
is observed as 0.256 (you can zoom the unit ramp response graph further in MATLAB).

Figure 5: Unit ramp response of the uncompensated system with K = 215 (at time t = 1000 sec)

Step 2: Next, a lead compensator has to be designed to reduce the peak time by a factor of 2, that is,
from 0.919 seconds to 0.46 seconds (theoretical value = 0.416 seconds).
The dominant poles for the desired peak time of 0.46 seconds can be obtained by theoretical means.
From the peak time relation, the imaginary part of the desired dominant pole, d is obtained as


d  = 7.552.
0.416
Now, for finding the real part of the desired dominant pole, a line has to be drawn to the left of the
imaginary axis in s-plane with respect to the origin at an angle,   cos 1    cos 1  0.358  = 69.0226°

as shown in Figure 6. From that, the real part of the desired dominant pole can be obtained as follows:
7.552 7.552
tan  69.0226    n   2.9
n tan  69.0226 
 Figure 6: Right angled triangle formed in s-plane to find the dominant pole

Therefore, the dominant poles for the desired transient response specification of peak time = 0.416
seconds are –2.9±7.552j. Now, we start the design of lead compensator. The selection of lead
compensator zero will be arbitrary. While selecting the zero for the compensator, the care should be
taken to ensure the pole-zero cancellation for the system in closed-loop operation.
And from various simulations in MATLAB (not shown here), placing the compensator zero
between first open-loop pole of s = 0 and second open-loop pole of s = –5 will have little chance of
pole-zero cancellation. Therefore in this example, we select the location of the compensator zero
coincident with the open-loop pole at –5. This choice will eliminate a zero and leave the lead-
compensated system with three poles, the same number that the uncompensated system has.
Now, we complete the design of lead compensator by finding the location of the compensator pole using
the root locus property as shown in Figure 7 by connecting the dominant pole to all the uncompensated
poles and zeros.

Figure 7: Angles formed between the dominant pole and all other poles and zeros
The angles formed between the dominant pole and all other poles and zeros can be obtained as follows:
1 = 180 – 69.0226 = 110.977°
7.552
tan  2   3.5962   2 =74.46° ( 3 also equal to 74.46°)
5  2.9
7.552
tan  4   0.93235   4  43°
11  2.9
Now, the angle contribution required from the compensator pole (pc) in order to make the root locus to
pass through the desired dominant pole can be obtained as
Angle contribution = 180° – (sum of angles from the dominant pole to all other poles)
+ (sum of angles from the dominant pole to all other zeros)
= 180  1   2   4   3 = 180 – (110.977+74.46+43) + 74.46 = 26.02°

Figure 8: Evaluating the lead compensator pole (pc)

Using the geometry shown in Figure 8,

7.552
 tan  26.02 
pc  2.9
from which the location of the compensator pole, pc, is found to be –18.37.
Therefore, the transfer function of the designed lead compensator is given by
s5
Glead (s) =
s  18.37
Now, the loop gain K for the lead-compensated system can be determined by either manual calculations
or from root locus graph of the lead-compensated system. The complete root locus of the lead-
compensated system Glead (s) G(s) is shown in Figure 9.
 Figure 9: Root locus for lead-compensated system

Figure 10: Root locus for lead-compensated system

(zoomed for finding the gain for corresponding dominant pole)

From the root locus graph obtained from MATLAB simulation (zoomed in Figure 10), it is evident that
the dominant pole for the lead-compensated system with 30% overshoot shows –2.88±7.52j with a gain
value of 1530. Thus, the simulation value not exactly matches with the calculated dominant pole value
of –2.9±7.552j. Therefore, I decided to calculate the loop gain K through manual calculations.
Determination of loop gain K for the lead-compensated system:
We know that the location of the dominant pole is a closed-loop pole location. And the closed-
loop poles can be obtained directly from the characteristic equation for various values of K. Here, we
know the dominant pole as –2.9±7.552j and the corresponding K value can be obtained as follows:
The characteristic equation for a system is given by 1+G(s) H(s) = 0. Since the given system in Figure 1
is unity feedback system, H(s) = 1.
Therefore, the characteristic equation is simply 1+G(s) = 0. The transfer function G(s) of the lead-
compensated system (including the loop gain K) is given by
K  s  5 K
G s  
s  s  5  s  11 s  18.37  s  s  11 s  18.37 

K K
From characteristic equation, 1  0   1
s  s  11 s  18.37  s  s  11 s  18.37 

 K    s  s  11 s  18.37   = 1545.5+3.5291 j

s 2.9  7.552 j

2 2
Therefore, K  1545.5   3.5291 = 1545.5 (remember, K = 1530 from simulation)

Steps 3 and 4: Now, we have to check the lead-compensator design with loop gain of 1545.5 in
MATLAB simulation. The step response of the lead-compensated system is shown in Figures 11 and 12.

Figure 11: Step response of the lead-compensated system with K = 1545.5 (theoretical value)
 Figure 12: Step response of the lead-compensated system with K = 1530 (simulation value)

Figure 13: Comparison of step response for uncompensated system and lead-compensated system

In Figure 11, the peak time is observed as 0.466 seconds (1.97 times reduction) with the theoretical loop
gain value of 1545.5. But in Figure 12, the peak time is observed as 0.461 seconds (two times
reduction) with the loop gain value of 1530 obtained from simulation (See Figure 10). Thus, it is evident
from the simulation that the lead-compensated system is having two times reduction in peak time (with
K = 1530) and operating with overshoot of less than 30% and the result is satisfactory. The comparison
of step responses of uncompensated system and lead-compensated system is shown in Figure 13.

Step 5: In order to meet the desired steady-state performance, which is 30 times reduction in steady-
state error for a ramp input which is equal to actual error of 0.256 (uncompensated system) divided by
30 = 0.0085. Now, the steady-state error of the lead-compensated system can be obtained either
manually or from simulation.
The steady-state error for the lead-compensated system with gain value of 1530 for the unit ramp
input is shown in Figure 14 and is observed as 0.132 (you can zoom the unit ramp response graph
further in MATLAB to find the difference).

Figure 14: Unit ramp response of lead-compensated system with K = 1530 (at time t = 1000 sec)

Theoretical calculations to find the steady-state error:

1530
The transfer function of the lead-compensated system is G  s   and it is a type-1
s  s  11 s  18.37 

1
system. For type-1 system with ramp input, the steady-state error = , where K v is the velocity error
Kv
1530
constant and is given by K v  Lt s G  s   Lt = 7.5716.
s 0 s 0  s  11 s  18.37 

Therefore, steady-state error = 1 / 7.5716 = 0.132. Thus, the steady-state error is reduced only around
two times, so further improvement is required to reduce it to the desired value of 0.0085.
Step 6: To design a lag compensator for the desired steady-state performance, the pole selection should
be arbitrary and is very close to origin. For the selected pole value, the corresponding zero has to be
determined. Below, I am going to explain two methods for finding the compensator zero (zc) for the
arbitrary selection of compensator pole (pc).
Method 1:
We find the steady-state error for the lead-compensated system (uncompensated for steady-state
performance) as 0.132 and its corresponding error constant is 7.5716. But the desired or required steady-
state error is 0.0085. Therefore, the desired error constant is obtained as 1 / 0.0085 = 117.647. Now, the
ratio of the desired error constant to the actual error constant is the required ratio of the compensator
zero to the compensator pole. That is,
zc 117.647
 = 15.538
pc 7.5716
Here, for lag compensator for the actual lead-compensated system, I am selecting the pole arbitrarily as
pc = 0.01 (very close to origin). Therefore, the lag compensator zero can be obtained directly as
zc  15.538  0.01 = 0.1554.

s  0.1554
Hence, the transfer function of the lag compensator is Glag (s) = . Now, the loop gain K of the
s  0.01
lag-lead compensated system can be obtained from root locus graph of the system. The root locus of the
lag-lead compensated system is shown in Figure 15.

Figure 15: Root locus for lag-lead compensated system

 Figure 16: Root locus for lag-lead compensated system (zoomed near the origin)

Figure 17: Root locus for lag-lead compensated system

(zoomed for finding the loop gain for corresponding transient specification)

From the root locus graph obtained from MATLAB simulation (zoomed in Figure 17), it is evident that
the dominant pole for the lead-compensated system with 30% overshoot shows –2.84±7.42j with a gain
value of 1520. Now, we can discuss the second method of finding the compensator zero for the arbitrary
selected pole of pc = 0.01 to get the desired steady-state error of 0.0085.
Method 2:
This method provides the theoretical means of determining the compensator zero as well as the
loop gain K without the root locus graph. We know the transfer function of the lead-compensated
K
system is G  s   and after adding the lag compensator, the system becomes a lag-
s  s  11 s  18.37 

K  s  zc 
lead compensated one and its transfer function is given by G  s   .
s  s  11 s  18.37  s  pc 

Now, for the compensator pole, the value is selected as pc = 0.01, therefore the open loop transfer
K  s  zc 
function of the lag-lead compensated system becomes G  s   .
s  s  11 s  18.37  s  0.01

We know that the steady-state error is given by

s R s
ess  Lt , here H  s  = 1
s0 1  G  s  H  s 

1 1
1 s = Lt s
For unit ramp input, R  s   , therefore ess  Lt
s 2 s  0 1  G  s  s 0 K  s  zc 
1
s  s  11 s  18.37  s  0.01

Here, the desired steady-state error = 0.0085, therefore

1  s  11 s  18.37  s  0.01
0.0085  Lt s  Lt
s0 K  s  zc  s 0 s  s  11 s  18.37  s  0.01  K  s  z 
c
1
s  s  11 s  18.37  s  0.01

1118.37  0.01
 0.0085   K zc = 237.73
K zc
Now, the characteristic equation of the lag-lead compensated system is 1+G(s) = 0
K  s  zc  K  s  zc 
 1 0   1
s  s  11 s  18.37  s  0.01 s  s  11 s  18.37  s  0.01

 K  s  zc     s  s  11 s  18.37  s  0.01 

K s  K zc    s  s  11 s  18.37  s  0.01 

K s    s  s  11 s  18.37  s  0.01   K zc

   s  s  11 s  18.37  s  0.01   237.73

  s  s  11 s  18.37  s  0.01   237.73

 K
s
s 2.9  7.552 j
   s 4  29.38 s 3  202.3637 s 2  2.0207 s   237.73
 = 1552.1+26.248 j
s
s 2.9 7.552 j

2 2
Therefore, K  1552.1   26.248  = 1552.32 (Compare it with the value K = 1520 obtained in

method 1 from root locus graph!!!).

237.73 237.73
Finally, we have K zc = 237.73, therefore, zc   = 0.1531.
K 1552.32
s  0.1531
Hence, the transfer function of the lag compensator is Glag (s) = .
s  0.01
After the design of lag compensator, the complete open-loop transfer function of the lag-lead
compensated system (including the loop gain K) from method 2 is given by
1552.32  s  0.1531
G  s 
s  s  11 s  18.37  s  0.01

Steps 7 and 8: Now, we have to check the performance of lag-lead compensator design with loop gain
of 1552.32 in MATLAB simulation. The step response of the lag-lead compensated system is shown in
Figure 18.

Figure 18: Step response of the lag-lead compensated system with K = 1552.32

In Figure 18, the peak time is observed as 0.463 seconds (approximately two times reduction) and
operating with overshoot of 31% (somewhat acceptable as it is too close to 30%) and the result is
satisfactory. The comparison of step responses of uncompensated system and lag-lead compensated
system is shown in Figure 19.

Figure 19: Comparison of step response for uncompensated system and lag-lead compensated system

Now, the steady-state performance of the lag-lead compensated system has to be verified with
simulation.

Figure 20: Unit ramp response of lag-lead compensated system with K = 1552.32 (at t = 1000 sec)
The steady-state error for the lag-lead compensated system with gain value of 1552.32 for the unit ramp
input is shown in Figure 20 and to obtain the exact steady-state error, the graph in Figure 20 has to be
zoomed further as shown in Figure 21.

Figure 21: Unit ramp response of lag-lead compensated system showing the final value of 999.9915

From Figure 21, it is evident that the lag-lead compensated system gives a final value of 999.9915 for
the unit ramp input and thus the steady-state error = 1000 – 999.9915 = 0.0085, and the result is
perfectly satisfied the requirement.

Summary of transfer functions:

1
Transfer function of actual uncompensated system, G  s   , K = 215
s  s  5 s  11

s5
Transfer function of lead compensator, Glead  s   , K = 1530
s  18.37

Transfer function of lag-lead compensator, Glag lead  s  

 s  5  s  0.1531 , K = 1552.32
 s  18.37  s  0.01

Reference textbook
 Norman S. Nise, ‘Control Systems Engineering’, Sixth Edition, John Wiley and Sons, 2011.

About the concepts explained here, feel free to ask your queries to: senthil.avionics@gmail.com