Session G-09

MODELLING AND SIMULATION OF AN ACTIVE VEHICLE

SUSPENSION SYSTEM
Milind Ranade and Jiang Zhou
Department of Mechanical Engineering
Lamar University, Beaumont, TX 77705
[email protected]

Abstract
The transmission of road generated vibrations to a vehicle body is treated as a source-path-
receiver problem. The suspension system acts as a path for transmitting forces and motions from
the body to the wheels and vice-versa. Springs and dampers are passive components of
suspension system which act as shock absorbers and energy dissipaters respectively, thereby
putting limitations on transmission of these forces and motions. However, suspension
performance can be greatly improved by using an active suspension system with adaptive
feedback control technique.

This paper presents design of a feedback controller called as notch filter for an active suspension
system. Root Locus approach is used as a design tool. In active suspension systems, sensors are
used to measure the accelerations of sprung mass and unsprung mass and the analog signals from
the sensors are sent to a controller. The controller is designed to take necessary actions to
improve the performance abilities already set. The controller amplifies the signals which are fed
to the actuator to generate the required forces to form closed loop system (active suspension
system).The performance of this system is then compared with that of the open loop system
(passive suspension system). A quarter vehicle suspension model with two degree of freedom
system consisting of one sprung mass and one un-sprung mass is considered for simplicity of the
analysis. MATLAB/SIMULINK is used to simulate the ride characteristics of the vehicle
suspension. Effect of variation of design parameters such as stiffness, damping constant on the
performance of suspension and in turn on the ride characteristics of the vehicle is also studied.

1. Introduction

Every vehicle moving on a randomly profiled road is exposed to vibrations which are harmful
both for the passengers in terms of comfort and for the durability of the vehicle itself. Therefore
the main task of a vehicle suspension is to ensure ride comfort and road holding for a variety of
road conditions and vehicle maneuvers. This in turn would directly contribute to the safety.

In recent years, considerable interest appeared in the use of active vehicle suspensions, which
can overcome some of the limitations of passive suspension systems1. Demands for better ride
Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

1
comfort and controllability of road vehicles has motivated many automotive industries to
consider the use of active suspensions1. The electronically controlled active suspension systems
can potentially improve the ride comfort as well as the road handling of the vehicle
simultaneously. In passive systems, the vehicle chassis is supported by only springs and
dampers. While in active systems, the springs and dampers are replaced, in parted or fully by
actuators. These act as force producers according to a certain control strategy, using the feedback
from the vehicle. Semi-active suspension systems are considered to be derived from active
systems, with the actuator replaced by controllable damper and a spring in parallel. Semi-active
suspension system can only change the viscous damping coefficient of the shock absorber it does
not add additional energy to the suspension system2.

In the following topic, a mathematical model of a conventional suspension system is built and
equations of motion are derived. Performance of the system is analyzed for various values of
system parameters. Further a SIMULINK model for this system is developed. Next topic explains
need of having a controller and then it explains design of a semi-active controller. The method of
design is illustrated with the help of root locus technique. Finally, displacement and acceleration
of the semi-active system is compared to that of passive suspension system.

2. Conventional Suspension System

A good vehicle suspension system should have satisfactory road holding ability, while still
providing comfort when riding over bumps and holes in the road1. When the vehicle is
experiencing any road disturbance (i.e. pot holes, cracks, and uneven pavement), the vehicle
body should not have large oscillations and the oscillations should dissipate quickly. Section 2-1
shows mathematical modeling and derivation of equation of motion of the conventional (passive)
suspension system, section 2-2 shows SIMULINK model of the same. With the help this model,
effect of change of spring stiffness and damping coefficient on the system response is discussed
in section 2-3. Root Locus plot for this system is shown in section 2-4 which explains the need of
having a controller in order to improve performance of the system.

2-1: Mathematical Model and Equation of motion

Figure 1 shows a quarter vehicle model of two degree of freedom system consisting of a sprung
mass and an unsprung mass. Sprung mass is a portion of the vehicle’s total mass which is
suspended above suspension system. The sprung mass typically includes the body, frame, the
internal components, passengers and cargo but does not include the mass of the components
suspended below the suspension components. This includes the wheels, wheel bearings, brake
rotors, calipers and/or caterpillars tracks, if any, which are part of the vehicle's unsprung mass.
Equation of motion of the system is,

. (1)

Assuming that the initial conditions are zero, Laplace transform of the above equation becomes,

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

2
 . (2)

Fig.1: Quarter Vehicle Model of a Passive (Conventional) Suspension System

Let ,

δ = det ,

or,
.

Pre-multiplying by A−1 on both sides

. (3)

A passenger in the vehicle feels displacement with respect to the disturbance, i.e., x1-w, which is
difficult to measure. Since the deformation of the tire, i.e, x2-w, is negligible, displacement of the
sprung mass with respect to the tire, x1-x2, instead of x1-w is used as output in the problem.
Therefore, transfer function of this system becomes
Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

3
The road disturbance, w, in this problem is simulated by a step input. This step could represent
the vehicle coming out of a pothole. The problem now can be represented with the help of a
simple block diagram as shown below in figure 2.

Fig. 2: Block diagram of Open Loop System

X1(s)-X2(s) is relative displacement of sprung mass with respect to that of the unsprung mass in Laplace
domain and is the output of the system.

For the purpose of analysis and simulation, the following values for stiffness, damping
coefficient and mass are considered. = 2500 kg, = 320 kg, = 350 N-s/m, =15,020 N-
4
s/m, = 80,000 N/m, = 500,000 N/m . A MATLAB program is developed to obtain the system
response to 0.1 m step disturbance as shown in figure 3.

response of passive suspension system to a step disturbance

0.1

0.08

0.06

0.04
displacement (m)

0.02

-0.02

-0.04

-0.06

-0.08

-0.1
0 5 10 15 20 25 30
time (sec)

Fig.3: Relative Displacement of Sprung Mass to 0.1 m Step Disturbance

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

4
2-2: SIMULINK Model of Open Loop System (Passive Suspension)
In order to faciliate the parameter study, a Simulink block-diagram based model is developed as
shown in figure 4. Unit step input of magnitude 0.1 m is given to the system. Output x1-x2 is
taken to the scope and to the MATLAB workspace.Similarly, net acceleration a1-a2 is also given
in scope as well as in the MATLAB workspace.Advantage of a Simulink model is that changes in
the system parameters can be made very easily and corresponding change in the output of the
system is obsereved immidiately.

Fig. 4: SIMULINK Model of Open Loop System (passive suspension)

2-3: Response of the System to Variation in System Parameters

Based on the SIMULINK model shown in figure 4, study of variation of spring constant and
damping coefficient is carried in the following section.
(A) Effect of change of stiffness of suspension spring (k1)

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

5
Stiffness of suspension spring is changed and response of the system to 0.1 m step disturbance is
observed as shown in figure 5 and figure 6. Three different values of stiffness of suspension
spring are considered 1.6×103 N/m, 80×103 N/m and 400×103 N/m. It is clearly evident from
figure 5 that larger the value of stiffness of spring, smaller is the displacement and conversely.
However, figure 6 shows that greater the value of spring stiffness, larger is the acceleration
experienced by passengers. Therefore value of spring stiffness should neither be too large nor
should be too small. A golden mean is desirable. It can also be seen that there is significant
change in frequency of vibration of the system, with change in stiffness of spring.

displacement of open loop system for variation in stiffness of spring

0.1
k 1=1.6e3
0.08
k 1=80e3

0.06 k 1=400e3

0.04
displacement (m)

0.02

-0.02

-0.04

-0.06

-0.08

-0.1
0 5 10 15 20 25
time (sec)

Fig.5: Relative Displacement of Sprung mass as value of k1 varied

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

6
 4
x 10 Acceleration of open loop system for variation in stiffness of spring
4
k 1=1.6e3
3 k 1=80e3
k 1=400e3
2

1
acceleration (m/s 2)

-1

-2

-3

-4

-5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time (sec)

Fig.6: Net Acceleration Experienced by Sprung Mass for various values of k1

(B) Effect of change of damping coefficient of the suspension (b1):

Figure 7 and figure 8 show acceleration and displacement of the sprung mass respectively, as b1
is increased and decreased by 50 times of the original value. The original value being 350 N-s/m,
the other two are 17500 N-s/m and 7 N-s/m.
It can be seen from the figure that the vibrations of the sprung mass dissipate faster as value of
the damping coefficient is increased. Settling time can be reduced to, as much as, less than 3
seconds by increasing damping coefficient to 17500 N-s/m. Greater the damping coefficient
better is the damping effect and the system quickly becomes stable. It can also be seen that there
is no change in frequency of vibration of the system at all, with change in damping coefficient.
The same trade-off between displacement and acceleration is observed even in this case. For
large damping coefficients large acceleration is experienced by the passenger where as
displacement of the passenger is small.

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

7
 displacement of open loop system for variation in damping coefficient
0.1
b1=7 N-s/m
0.08
b1=350 N-s/m
0.06 b1=17500 N-s/m

0.04
displacement (m)

0.02

-0.02

-0.04

-0.06

-0.08

-0.1
0 5 10 15 20 25
time (sec)

Fig. 7: Relative Displacement of Sprung mass as value of b1 is varied (50 times)

4
x 10 Acceleration of open loop system for variation in damping coefficient
3
b1=7 N-s/m
b1=350 N-s/m
2
b1=17500 N-s/m

1
acceleration (m/s )
2

-1

-2

-3
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
time (sec)

Fig. 8: Net Acceleration Experienced by Sprung Mass as value of b1 is varied (50 times)

Figure 9 and figure 10 show acceleration and displacement of sprung mass respectively, as value
of damping coefficient (b1) is increased and decreased by 10 times of the original value. The
Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

8
original value being 350 N-s/m, the other two values are 3500 N-s/m and 35 N-s/m. This shows
that the damping effect is small when there is small value of damping coefficient.
displacement of open loop system for variation in damping coefficient
0.1
b1=35 N-s/m
0.08
b1=350 N-s/m
0.06 b1=3500 N-s/m

0.04
displacement (m)

0.02

-0.02

-0.04

-0.06

-0.08

-0.1
0 5 10 15 20 25
time (sec)

Fig. 9: Relative Displacement of Sprung Mass as value of b1 is varied (10 times)

4
x 10 Acceleration of open loop system for variation in damping coefficient
2.5
b1=35 N-s/m
2
b1=350 N-s/m
1.5 b1=3500 N-s/m

1
acceleration (m/s 2)

0.5

-0.5

-1

-1.5

-2

-2.5

-3
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
time (sec)

Fig. 10: Net Acceleration of Sprung mass as value of b1 is varied (10 times)

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

9
2-4: Root Locus of Passive Suspension System

Root Locus
Root Locus for conventional suspension system
40

30

20

10
Imaginary Axis

-10

-20

-30

-40
-40 -30 -20 -10 0 10 20 30 40
Real Axis

Fig.11: Root Locus of uncompensated system whose open loop transfer function is G(s)
Figure 11 shows root locus plot of the system whose open loop transfer function is G(s).There is
a pair of complex conjugate open loop poles laying very close to imaginary axis in complex
plane. This makes the system marginally stable. It is needed to shift the root locus towards left in
the complex plane. There is a need to change the shape of the root locus and hence need of a
controller.

3. Feedback Control System (Active Suspension):

Objective is to design a feedback controller so that the output (x1-x2) has an overshoot less than
+/-5% and a settling time shorter than 5 seconds. For example, when the vehicle runs onto a 10
cm high step, the vehicle body will oscillate within a range of +/- 5 mm and return to a smooth
ride within 5 seconds. The main idea of root locus design is to estimate the closed-loop response
from the open-loop root locus plot. By adding zeros and/or poles to the original system (adding a
compensator), the root locus and thus the closed-loop response can modified.

In this system, the transfer function of the plant contains one of the two pairs of complex-
conjugate poles which lie close to the imaginary axis in the s-plane. These are dominant open
loop poles of the system. This results in an undesirable closed-loop system that is unstable or
Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

10
only lightly damped. One way to control this system is to design a controller with zeros near the
undesirable, lightly-damped poles of the plant. These zeros can attenuate the effect of these
poles. The poles of the controller can then be placed in a more desirable position. Such a
controller is called a notch filter. The developed design allows the suspension system to behave
differently in different operating conditions, without compromising on road-holding ability. The
effectiveness of this control method has been explained by data from time domains.
Section 3-1 shows a mathematical model of the active suspension system. Equations of motion
are derived and a controller (compensator) is designed consequently. Section 3-2 explains root
locus of compensated system. Further, characteristics of the system response are observed in
section 3-3.

3-1: Model of the Vehicle Suspension with Active Control

In active suspension systems actuators are added to the passive components as shown in Figure
12, u being force from the controller. The system’s advantage is that even if the active actuator
or the control system fails, the passive components continue to operate. Figure 13 shows block
diagram of feedback control system.
Equation of motion of the system is

. (5)

Fig.12: Quarter vehicle model of an active suspension system

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

11
 Fig.13: Block diagram of feedback control (the active suspension) system

Assuming that the initial conditions are zero, Laplace transform of the above equation becomes,

. (6)

Transfer function of the Plant is

Transfer function of the Controller is

Eq. (8) can be written in the form of

where, Kc = 0.2 is gain of the controller and = 3+4i, = 3 4i are the two zeros and = ,
= are the two poles.

The controller is essentially a Notch Filter which consists of two lead compensators. By adding
two zeros close to the dominant open loop poles of the plant, their effect is nullified.
Furthermore, two poles are added away from the imaginary axis to get quicker response.

3-2: Root Locus of the Compensated System

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

12
It can be seen from the root locus plot of the compensated system shown in figure 14 that the
system has a higher critical gain value. Thus the system is no more marginally stable and is in
fact stable up to gain value of 14.

Root Locus
Root Locus of compensated (active) suspension system
100

80

60

40

20
Imaginary Axis

-20

-40

-60

-80

-100
-100 -50 0 50 100 150
Real Axis

Fig.14: Root locus of compensated system whose open loop transfer function is G(s) × Gc(s)

The complex-conjugate poles near the imaginary axis have been nearly canceled and more part
of the root locus is now in the left half plane than that of passive system. This means that higher
gains (kc) can be used, while maintaining stability.

3-3: Characteristics of the system response

Figure 15 shows response of the active suspension system to unit step disturbance of magnitude
10 cm. It can be seen from the figure that the system settles down within two seconds (i.e.< 5
sec) and maximum overshoot is about 4 mm (i.e.<+/-5%).This system satisfies the design
requirements and hence is one of the possible solutions of the problem. Acceleration of the
sprung mass, shown in figure 16, is limited to 140 m/s2 and it vanishes within half a second. This
is a significant improvement considering that acceleration experienced by a passenger using
conventional suspension system is about 2 x 104 m/s2 (see Fig 6).

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

13
 -3
x 10 displacement of compensated system to 0.1 m step disturbance
2

0
Displacement (m)

-1

-2

-3

-4

-5
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time (sec)

Fig.15: Relative displacement of Sprung Mass [x1(t) − x2(t)] to 0.1 m step disturbance

acceleration of compensated system to 0.1 m step disturbance

160

140

120

100
Acceleration (m/s 2)

80

60

40

20

-20
0 0.05 0.1 0.15
Time (sec)

Fig 16: Acceleration experienced by sprung mass when subjected to 0.1 m step disturbance
Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

14
4. Conclusion
1. Active suspension system has been developed by using a notch filter. R, L, C components of
the notch filter can be chosen in such a way that the transfer function of the notch filter becomes
same as the transfer function of the controller (Gc).
2. It has been found that the system is very sensitive to changes in parameters. Active control
System becomes unstable for values of gain more than 14.3 and therefore increasing gain of the
system may not result in better performance.
3. Active suspension system has improved ride comfort. Acceleration experienced by a
passenger using the system is reduced by over 140 times the original value.

5. References
1. M. Senthil kumar, S. Vijayarangan; Analytical and experimental studies on active suspension system of light
passenger vehicle to improve ride comfort, ISSN 1392 – 1207; Mechanika, 2007.
2. J. Y. Wong; Theory of Ground Vehicles; John Wiley and Sons, Inc; Hoboken, New Jersey;2008; 4th edition,
chapter 7.
3. E. Esmailzadeh and H. Bateni; Optimal Active Vehicle Suspensions with Full State Feedback Control;
Suspensions and their Relationship to Vehicle Roll Stability; SAE, Inc; Warrendale, PA, 1992; SP-940.
4. Carnegie Mellon Control Tutorials for MATLAB.

Proceeding of the 2010 ASEE Gulf-Southwest Annual Conference, McNeese State University
Copyright © 2010, American Society for Engineering Education

15