Daniyan et al.

, Cogent Engineering (2018), 5: 1545409

https://doi.org/10.1080/23311916.2018.1545409

PRODUCTION & MANUFACTURING | RESEARCH ARTICLE

Design and simulation of a controller for an
active suspension system of a rail car
I. A Daniyan1*, K. Mpofu1 and D. F. Osadare2

Received: 01 August 2018
Accepted: 29 October 2018
First Published: 12 November 2018
*Corresponding author: I. A Daniyan,
Department of Industrial
Engineering, Tshwane University of
Technology, Pretoria, South Africa
E-mail: afolabiilesanmi@yahoo.com
Reviewing editor:
Jun Guo, Wuhan University of
Technology, China
Additional information is available at
the end of the article
Abstract: The quest to increase the performance of rail car in terms of stability,
safety and ride comfort has triggered research into the design, modelling and
dynamic simulation of the suspension system of a rail car. Advance control and
automation techniques can be used to optimize the performance of a rail car
suspension system. This work introduces the mathematical model of two-degree-of
-freedom and the use of the conventional proportional integral as well as derivative
(PID) controller for the active suspension system of a rail car. Dynamic simulation
via the use of MATLAB Simulink environment was employed for the design and
analysis of the effect of the controller with the control algorithm of PID controller
used to reduce motion of rail car body and the damper. The dynamic model for the
linear rail car suspension system was formulated while the PID controller type was
used to test the system’s performance while the Ziegler–Nichols tuning rules were
used to determine proportional gain, reset rate and derivative time of PID control-
lers. The results obtained show that the developed controller is highly efficient. The
simulated results of the comparison between the passive suspension system and
the active suspension system proved that, active suspension system with PID

ABOUT THE AUTHORS PUBLIC INTEREST STATEMENT

Dr. I. A. Daniyan is a Senior Lecturer in the
Department of Mechanical and Mechatronics
Engineering, Afe Babalola University Ado Ekiti,
Nigeria and a Postdoctoral Research Fellow at the
Department of Industrial Engineering, Tshwane
University of Technology, Pretoria, South Africa.
He was the supervisor of this research.
Prof. K. Mpofu is a Professor and research lea-
der at the Industrial Engineering Department,
Tshwane University of Technology in South Africa.
He has been involved in advanced manufacturing
research since 2007. His areas of specialization
include the innovative design of manufacturing
I. A Daniyan systems, facility layout and the design of manu-
facturing equipment. He holds a Doctoral degree
in Mechanical Engineering, a Master’s degree in
Manufacturing Systems and a Bachelor of
Engineering degree in Industrial and
Manufacturing Engineering. He is a member of
South African Institute of Industrial Engineers
(SAIIE).
Mr. D. F. Osadare holds a Bachelor of
Engineering Degree in Mechanical Engineering
from Afe Babalola University, Ado Ekiti, Nigeria.
He played a prominent role in this research.
The quest to increase the performance of rail car
in terms of stability, safety and ride comfort has
triggered research into the design, modelling and
dynamic simulation of the suspension system of
a rail car using advanced control. Every moving
rail car is exposed to different disturbances and
vibration which results in passengers’ discomfort,
energy losses, decreased durability of the car
itself and increased cost of maintenance. These
dynamic motions could also be unsafe, because
the tyres might wear out or lose their grip on the
rail way. Therefore, the main focus of this
research is to design an active rail car suspension
system with advance control mechanism to
ensure ride comfort by minimizing the effects of
such disturbances or complete isolation of the
car from the disturbances while enhancing
increased rail way holding capacity for a variety
of rail way conditions and rail car manoeuvres.

© 2018 The Author(s). This open access article is distributed under a Creative Commons
Attribution (CC-BY) 4.0 license.

Page 1 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

control improves ride comfort, gives lower or almost no peak over shoot (amplitude)
and faster settling time compared to the passive suspension system. This enhances
safety, good comfort for drivers and passengers as well as isolation of railcar from
noise, vibration and other disturbances.

Subjects: Industrial Design; Mechanical Engineering; Mechanical Engineering Design;

Manufacturing Engineering; Manufacturing Technology

Keywords: advance control; controller; damper; rail car; suspension system

1. Introduction
In the past years, there has been widespread interest in the use of advance control and automa-
tion techniques to optimize the performance of the suspension system of a rail car (Al-Zughaibi
and Davis, 2015; Bideleh & Berbyuk, 2016). Suspension systems have been widely applied to the
modern automobile and rail cars with complex control algorithms to reduce the effects of vertical
acceleration caused by road disturbances (Ulum, Affaf, & Suparwanto, 2017; Yao & Shi, 2013).
Every moving rail car is exposed to vibration which results in passengers’ discomfort, decreased
durability of the car itself and increased cost of maintenance. Different disturbances occur when
a rail car leans over during driving, cornering, diving or braking. Also, unpleasant vertical move-
ment, lateral, yawing and rolling vibrations of the rail car body can occur while driving over rail way
irregularities (Hasbullah & Faris, 2017; Oh et al., 2016). These dynamic motions do not only have an
adverse effect on comfort but can also be unsafe, because the tyres might wear out or lose their
grip on the rail way. The main task of a rail car suspension is to ensure ride comfort by isolating the
car from disturbances and enhance rail way holding capacity for a variety of rail way conditions
and rail car manoeuvres (Colombo, Gialleonardo, Facchinetti, & Bruni, 2014; Zhang, Han, Zhang, &
Yu, 2013). This in turn will directly contribute to the comfort and safety of the user. In general,
suspension system connects the wheel and the vehicle body by springs, dampers and some other
linkages (Shin, Lee, Yi, & Noh, 2015; Zong, Gong, Xuan, & Guo, 2013). A good suspension system
should provide a comfortable ride and good handling within a reasonable range of deflection
(Bideleh, Mei, & Berbyuk, 2016; Nguyen & Nguyen, 2017). From the system design perspective,
there are two main categories of disturbances on a rail car, namely railway and load disturbances
(Matamoros-Sanchez & Goodall, 2015; Nakajima, Shimokawa, Mizuno, & Sugiyama, 2014). Railway
disturbances have the characteristics of large magnitude in low frequency such as hills and small
magnitude in high frequency such as railway roughness. Load disturbances include the variation of
load induced by accelerating, braking and cornering. Therefore, in the design of a good suspension
system, importance is given to fairly reduce the disturbances via the use of suspension system
with proper cushioning that will be soft against railway disturbances and hard against load
disturbances. The vibration that occurs in most rail cars is undesirable, not only because of the
resulting unpleasant motions, the noise and the dynamic stresses which may lead to failure of the
suspension system but also because of energy losses and reduction in performance that accom-
pany the vibration (Eris, Ergen, & Kurtulan, 2015; Gohrle, Schindler, Wagner, & Sawodny, 2014;
Hasbullah, Faris, Darsivan, & Abdelrahman, 2015). The design of the rail car suspension system
always involves a compromise between ride comfort and handling (Ahmed, Hazlina, & Rashid,
2015; Pathare & Nimbalkar, 2014; Wang, Chen, & Yu, 2017). For good ride comfort, a compliant
suspension system is normally required while good handling demands a stiff suspension system to
control body roll (Sun, Deng, & Li, 2013). With normal passive suspension system, the character-
istics of the springs and dampers are fixed at the design stage and cannot be changed afterwards.
By using controllable springs and dampers, these characteristics can be changed while the rail car
is moving. It therefore becomes possible to have soft setting for good ride comfort whilst travelling
in a straight line on a good railway, while the suspension can be changed to a hard setting
moment later to give good handling when the rail car has to change direction as required for
lane changing or even accident avoidance. Controllable suspension systems can therefore reduce
or even eliminate ride comfort against handling compromise. Conventional suspension system is

Page 2 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

also known as a passive suspension system consisting of spring and damper mounted at each
wheels of the vehicle horizontally in parallel. According to Ha, Choi, and Lee (2012), the spring of
the vehicle suspension supports the vehicle body, absorb and store energy. The damper or shock
absorber is a component of the rail car suspension that is used to dissipate the vibration energy
stored in the spring and to control the impulse from the rail that is transmitted to the car. Other
purposes of suspension system are to isolate sprung mass from the unsprung mass vibration, to
provide directional stability during cornering and to manoeuvre while providing damping for the
high-frequency vibration-induced excitations (Herbst, 2013; Kumar, Sivakumar, Kanagarajan, &
Kuberan, 2018). Due to the lack of attitude control of the vehicle body and demand for better
ride comfort and stability, many researchers and academics in the automotive field are motivated
to consider the use of active suspensions. The electronically controlled active suspension system
can potentially improve ride comfort as well as the road handling and vehicle stability especially
for racing car (Kim, Choi, & Lee, 2014; Oh et al., 2016). A better suspension system with soft and
hard suspension should be controlled electronically in order to get ‘‘soft” against railway distur-
bance and “hard” against load disturbance (Hrovat, 1990). In order to increase the performance of
ride handling and stability of the vehicle, a new suspension system has been developed in which
the system is controlled electronically. There are two categories of advanced suspension system,
namely semi-active and active suspensions. In the automotive industries, semi-active and active
suspension systems are considered as new technologies that are used to improve ride quality and
handling of the vehicle with the semi-active suspension system showing some similarities with the
conventional suspension system. This kind of suspension has a spring and controllable damper in
which the spring element is used to store the energy meanwhile the controllable damper is used to
dissipate the energy. Some of the semi-active suspension systems use the passive damper and the
controllable spring, the controllable damper usually acts with limited capability to produce
a controlled force when dissipating energy (Hrovat, 1990; Nguyen, Choi, & Nguyen, 2014). The
advantages of using semi-active suspension are the operational cost which is less than active
suspension coupled with the fact that only small amount of energy is consumed (Jin, Xiao, Ling,
Zhou, & Xiong, 2013; Marzbanrada, Ahmadib, Zohoorc, & Hojjatd, 2004). Meanwhile in active
system, the components of spring or damper are replaced with an actuator controlled with the
aid of the feedback mechanism from the vehicle body. Technically, active suspension system is
used to control the movement of a vehicle using on-board controller by controlling the tire
movement during cornering, braking and accelerating. The method of the controller for active
suspension can be divided into four types based upon the control techniques, namely solenoid
actuated, hydraulic actuated, electromagnetic recuperative and magneto-rheological damper. For
semi-active with controllable damper, it acts like an actuator but with limited capability, to
produce a controlled force when dissipating energy and when supplying energy, it switches to
a zero damping state. The advantage of semi-active system is that the cost of operation is less
than the other type of advanced suspension system (Oraby, Aly, El-Demerdash, & Selim, 2007).

According to Arunachalam, Mannar Jawahar, and Tamilporai (2003), for the full active suspen-
sion system, energy is supplied into the system. It requires several components such as the
accelerometer, actuators, control unit, ACs, servo valves, high pressure tanks for the control
fluid, sensors for detecting the system (Allen, 2008). According to Ebrahimi (2009), they are
expensive but have better performance than the passive suspension system. There are two
types of configurations which are dependent upon the link of passive part (spring and damper)
and active part (controller). If these two parts are linked in parallel, they are known as high-
bandwidth configuration. The advantages of high-bandwidth to low-bandwidth lies in the ability to
exercise control at the higher frequency while the suspension system is working continuously as
a passive suspension in the event of an actuator not working as an active part. Many researchers
have carried out early works and researches on advanced suspension in the past few decades in
order to improve the stability and ride handling performance of the rail car using advanced
methods such as: proportional–integrator–derivational (PID) controller plus linear quadratic reg-
ulator (LQR) vibration controller (Minakaldas & Arefsoliman, 2014); LQR control design
(Arefsoliman, 2011; Nguyen, Nguyen, & Kim, 2000); fuzzy scheme (Abdelhady, 2003; Shao-Juan,

Page 3 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Zhong-Hua, & Chen, 2014); impedance control system (Fateh & Alavi, 2009); neural network
control system (Eski and Yildirim, 2009); spatial vehicle model (Demic, 2003); fuzzy sliding mode
controller (Lin, Lian, Chung-Neng, & Wun-Tong, 2009); fuzzy reasoning and a disturbance observer
(Semiha & Huseyin, 2008; Yoshimura & Takagi, 2004); fuzzy logic controller (Chen, Liu, Tong, & Lin,
2008; Chiou & Liu, 2009; Du & Zhang, 2009; Qiu, Feng, & Yang, 2009). Another control strategy has
been done by Yamamoto, Sugai, Kanda, and Buma (2014) using PID controller. This new control
strategy is proposed for half car model that has four-degree-of-freedom (DOF) and as a result, the
PID controller is capable to minimize the vehicle body acceleration. Although, from most of the
research work on advanced suspension either active or semi-active suspension, it was found that
an advanced active suspension system is able to increase the performance of ride and handling
stability as well as safety of the vehicle especially the rail car. Recent study by Hany, Al Adl, Samir,
and Younes (2008), regarding an active suspension system of a quarter car model using MATLAB
Simulink and IPG Car Maker software based on PID control scheme also confirms this fact. Recent
works on improving the dynamic performance of the vehicle have been conducted by Nurkan,
Sakman, and Guclu (2008) on quarter car model of an active suspension system of light passenger
vehicle using PID controller in order to improve ride comfort. Jinzhi, Fan, and Jun (2002) have
performed the controller design for active vehicle suspension by using GA-based PID and fuzzy
logic in controlling four DOF ground vehicles based on pitch plane of half car model. In their work,
two loops controller are able to minimize the vertical acceleration and pitch acceleration of the
vehicle body as well as to improve vehicle ride comfort performance with better system robust-
ness. The study for roll plane of half car model with four DOF has also been performed by
Yoshimura and Watanabe (2003). The controller is proposed in order to prevent the rollover
situation and as a result, there were significant performances as compared with the passive
system without rollover prevention. Zhu, Du, Zhang, and Wang (2014) developed a more sophis-
ticated mathematical linear model for a roll-plane active hydraulically interconnected suspension
system for verification of the new model. Data comparisons between simulations and experiments
indicate high consistent response of the model and the real system thus validating the robustness
and accuracy of the new mathematical model.

The aim of this work is to study and control the behaviour of a rail car suspension system when
cornering or hitting railway disturbances. Although many researchers have reported the design
and simulation of active suspension systems for both automobile and rail vehicles but the dynamic
simulation and control of the effects of rail vehicle acceleration, suspension travel and rail holding
ability on the active suspension system have not been sufficiently highlighted. The work focuses on
the design and dynamic simulation of a controller for the active suspension system of a rail car in
order to improve ride comfort, safety, performance of rail cars and minimize frequency of main-
tenance of rail car and railways via the isolation or compensation for unwanted disturbances as
well as unwanted vibrations.

2. Materials and method

This research methodology utilizes a simulation design via the use of MATLAB-Simulink to inves-
tigate the suspension system behaviour. The main reason for employing dynamic simulation is
that it can easily and adequately study the behaviour of a system during the design stage before
development of the real system. The dynamic simulation involves several steps. The first step is the
modelling process where a schematic diagram of suspension is sketched based upon the real
system of suspension. Then, the system is represented in MATLAB-Simulink based on the equation
of motion created from the schematic diagram. The bogie of a rail car is shown in Figure 1.

2.1. Mathematical model of the active suspension system of a rail car

The equations for the model motion were found by adding the vertical forces on the sprung mass
and unspring masses. The sprung mass is represented by ms , while the tire and axles are
represented by the unspring mass mu . The spring, damper and a variable force generating element
placed between the sprung and unspring masses constitute suspension. From this model, the
design was expended into a full car model. The goals of designing a suspension system are to

Page 4 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Figure 1. The bogie of a rail car

(WRC, 2017).

enhance driving comfort and good rail handling with the suspension system able to reduce the
vertical body acceleration and suspension travel. The passive suspension system cannot imple-
ment these goals because in order to have a good rail handling ability, more stiff suspension is
needed meanwhile for the stability and comfort, less stiff suspension is needed. To solve this
problem, an active suspension system was introduced in which an actuator is used to control
system. This report concentrates more on an active suspension in which an actuator is fixed
parallel with the spring and damper between sprung and unsprung mass of the vehicle. The
actuator is used to take a necessary action to supply and modulate the energy in order to improve
the performance of driving comfort and ride handling. The actuator is fixed parallel with the spring
and damper. In case the actuator is not functioning well or damaged, the suspension system will
still work as the passive system without a controller. The dynamic model is aimed at obtaining the
full state feedback response of non-linear quarter car suspension model. It presents the simulation
of the full state feedback response of a quarter car active suspension model with the suspension
deflection as the output and step inputs from the road disturbance and the actuator. Figure 2
shows a two- DOF system representing a free body diagram for the car rail model. The model
consists of the sprung mass ms and the unspring mass mu . The tire is modelled as a spring with

Figure 2. The free body diagram

for the rail car model (Ling
et al., 2014).

Page 5 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

stiffness Fkt . The suspension system consists of a passive spring Fks and a damper Fbs in parallel
with an active control force F.

From Figure 2 and Newton’s law, the dynamic equation was obtained as follows:

Let,

€1
Fms ¼ ms x (1)

€2
Fmu ¼ mu x (2)

3
Fks ¼ kls ðx2  x1 Þ þ knl
s ðx2  x1 Þ (3)

Fkt ¼ kt ðx2  wÞ (4)

Fbs ¼ bls ðx_ 2  x_ 1 Þ (5)

Let

bt ¼ 0 then; Fbt ¼ 0

þ " ∑F ¼ mx

Considering

ms

This implies that,

Fms ¼ FKS  F þ Fbs (6)

ms €x1 ¼ kls ðx2  x1 Þ þ knl 3 l _ _

s ðx2  x1 Þ  F þ bs ðx2  x1 Þ (7)

1 h l i
€1 ¼
x ks ðx2  x1 Þ þ knl 3 l _ _
s ðx2  x1 Þ  F þ bs ðx2  x1 Þ (8)
ms

Considering

ms

Fmu ¼ Fks þ F  Fbs þ Fkt (9)

€2 ¼ kls ðx2  x1 Þ  knl

mu x 3 l _ _
s ðx2  x1 Þ þ F  bs ðx2  x1 Þ þ kt ðx2  wÞ (10)

1 h l i
€2 ¼
x ks ðx2  x1 Þ  knl 3 l _ _
s ðx2  x1 Þ þ F  bs ðx2  x1 Þ þ kt ðx2  wÞ (11)
mu

For the state-space representation,

Let

x_ 1 ¼ x3 0 (12)

Page 6 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

x_ 2 ¼ x4 0 (13)

1 h l i
x_ 3 ¼ €x1 ¼ k ðx2  x1 Þ þ knl 3 l _ _
s ðx2  x1 Þ  F þ bs ðx2  x1 Þ (14)
ms s

1 h l i
x_ 4 ¼ €x2 ¼ ks ðx2  x1 Þ  knl 3 l _ _
s ðx2  x1 Þ þ F  bs ðx2  x1 Þ þ kt ðx2  wÞ (15)
mu

Using the state-space representation,

€ ¼ f ðxÞ þ gu
x (16)

The state, input and output vector are expressed as Equations (17)–(19), respectively.

x ¼ ½x1 x2 x3 x4 T (17)

u ¼ ½u w (18)

y ¼ ½ y i:e y ¼ x2  x1 (19)

where yw is the lateral displacement (m), ψ m is the yawing angle (deg.), ϕw is the rolling angle (deg.),
YrL;R is the lateral displacement along the horizontal axis (m), ZrL;R is the vertical displacement (m),
ϕrL;R is the torsional angle (deg.), X; Y; Z are the system axes with coordinates xI ; yI and zI .

ms is the sprung mass (kg); mu is the unspring mass (kg), kls is the linear component of the spring
constant (N/m), knl
s is the non-linear component of the spring constant (N/m), kt is the wheel spring
constant (N/m), bls is the linear component of the damping coefficient (Ns/m), bnl s is the non-linear
component of the damping coefficient (Ns/m), y ¼ x2  x1 is the suspension deflection (m), x1 is
the body deflection (m), x2 is the wheel deflection (m), w is the road disturbance (m), u is the
system input, F ¼ Axp is the actuator force (N), Fks is the spring force acting on the body (N), Fbs is
the damping force acting on the body (N) and Fkt is the spring force acting on the wheel (N).

2.2. Design using Simulink

Designing an automatic suspension system for a rail car is a control problem. The suspension
system is designed and a model is used to simplify the problem to a one-dimensional spring-
damper system. In this case, m1 , m2 and w are the sprung mass, unsprung mass as well as the
vertical displacement of rail, respectively.

The model parameters are given as follows:

(1) Sprung or body mass (m1) = 120,000 kg

(2) Unsprung or wheel assembly mass (m2) = 1125 kg
(3) Spring constant or stiffness of suspension system (k1) = 10,0000 N/m
(4) Spring constant or stiffness of wheel assembly (k2) = 800,000 N/m
(5) Damping constant or coefficient of suspension system (b1) = 3050 Ns/m
(6) Damping constant or coefficient of wheel assembly (b2) = 25,000 Ns/m

The simulated model is shown in Figure 3.

2.3. The description of model for rail disturbance input

In this case, rail bump is declared as the rail disturbance input for the model. For a rail car, only
a single wheel is declared to hit the rail bump. The rail bump is assumed to have 0.05 m height
presented as a step input w, in MATLAB-Simulink.

Page 7 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Figure 3. The simulated model.

2.4. Controller design using proportional–integral–derivative (PID)

PID control stands for proportional, integral and derivative control. The controls are designed to
automatically eliminate steady-state errors and adjust to the system’s requirements. In order to
avoid the small variation of the output at the steady-state, the PID controller is also designed so
that it reduces the errors by the derivative nature of the controller. A PID controller is depicted in
Figure 4. The set-point is the reference point where the measurement is pre-set to be and the error
generated is the difference between the set-point and measurement representing deviation from
the pre-determined values (Error = set-point − measurement) while the variable being adjusted to
the desired range is called the manipulated variable which is usually equal to the output of the
controller. The output signal of the PID controllers often responds to changes in the measurements
and reference points over time. Increasing the controller gain will make the loop go unstable.
Integral action was included in controllers to eliminate this offset. With integral action, the
controller output is proportional to the amount of time the error is present. Integral action
eliminates offset. Controller Output = (1/Integral) (Integral of) e(t) d(t). With derivative action,
the controller output is proportional to the rate of change of the measurement or error calculated
by the rate of change of the measurement with time hence the derivative action compensates for
changes in measurement by preventing further changes in the measurement when compared to
the proportional action. When a load or set-point change occurs, the derivative action causes the
controller gain to move the “wrong” way when the measurement gets near the set-point hence it
is often used to avoid overshoot. For the purpose of this study, the error is taken to be the
difference between the measured acceleration and desired acceleration. The block diagram of
the PID controller is shown in Figure 4.

The parameters for the design of PID controller for active suspension system are: body accel-
eration, suspension travel and rail holding.

The design of controller is expressed by Equation (20):

Page 8 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Figure 4. The block diagram of

the PID controller.

Ki t deðtÞ
uc ¼ Kp eðtÞ þ ò eðtÞdt þ Kd Td (20)
Ti 0 dt

The signal (U) will pass through the controller that computes the derivative and integral of error
signal. The signal U after passing the controller is equal to the summation of proportional gain (Kp)
multiplied by the magnitude of the error and integral gain (Ki) time the integral of the error and
derivative gain (Kd) multiplied by the derivative of the error. The signal U will be sent to the system
and the output of Y will be obtained. The signal then will be sent again to the sensor in order to
check an error occurs in the system. The PID controller was designed in the MATLAB environment
using the transfer function directly and employed applied to generate a continuous-time control.
Thus, the tuning of PID controller was done using following steps: generation of transfer function;
importation of the obtained parameters to linear time-invariant system to confirm the tuning base
and tuning the PID controller.

The PID controller is shown in Figure 5.

In designing a controller for active suspension, it is necessary to know the standard form of the
control structure. The standard form of control structure consists of three items: reference, con-
troller (PID) and subsystem. The controller reference is used as a target of the control design to
keep the rail car body stable when in contact with external disturbance. It is necessary to tune the
PID iteratively in order to obtain better control action for the system. In order to achieve the
system requirements, it is necessary to make sure the proposed controller working properly by

Figure 5. PID controller.

Page 9 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

tuning the controller properly. In other words, this tuning process must be fine-tuned in order to
achieve the required set such as decreasing rise time and settling time, and eliminating the
steady-state error.

In case of PID controller, there are several methods used in tuning this three-term controller.
Those methods are process reaction curve method and ultimate cycle method. There is another
method used for tuning the PID controller, namely the “trial and error” method to get the value of
Kp, Ki and Kd. The potential for improved ride comfort and better road holding using PID controller
design was examined.

3. Results and discussion

In this report, the trial and error method was used in order to clearly determine the function and
the characteristics of proportional (P), integral (I) and derivative. Based on the characteristic PID
controller as shown in Table 1, it can be concluded that the Kp, Ki and Kd are dependent of each
other.

Table 1 also shows the summarization of the PID controller characteristic. It can be noted that
there is a relationship between these three-term controllers because if one of the controllers is
changed, it gives an effect to the other two. Increasing and decreasing the value of the propor-
tional gain (Kp) only give an effect to the decreasing a rise time and the steady-state error. Tuning
an integral gain (Ki) can also reduce the rise time of the system and able to eliminate the steady-
state error. Meanwhile, a derivative gain (Kd) gives an effect of decreasing an overshoot and
settling time, and has a small change for rise time and steady-state error. The combination of
these controllers increases the stability of the system, improves the transient response as well as
eliminates the steady-state error. The way to finding the values of Kp, Ki and Kd is from the trial and
error method according to Zeigler–Nichols tuning rules. Zeigler–Nichols proposed the rules for
determining the proportional gain Kp, integral time
Ti and derivative time Td based on the transient response characteristics of a given system. The
tuning values of PID controller using Zeigler–Nichols are presented in Table 2.

3.1. Analysis of the suspension system using dynamic simulation

Simulation based on the mathematical model for quarter car using dynamic simulation was
performed. The rail disturbances produce a push on the rail car constituting additional weight on
the rail car body. According to the Newton’s third law of motion, a reaction force to this push
produces an equivalent but opposite weight on the rail car body. This force is transmitted through
the entire rail car body causing mechanical stresses to build up. The performances of the suspen-
sion system in terms ride comfort and ride handling were observed, where road disturbance is
assumed as the input for the system. Parameters observed are the car body travel, wheel deflec-
tion and the car body acceleration for quarter car. Force generated by damper in case of active PID
controlled system was also observed for controller accuracy with small amplitude value for vehicle
body acceleration (Figure 6). The excitation force gets cancelled by the generated force by the
actuator indicating the efficiency of the PID controller. By comparing the performance of the
passive and active suspension system using PID control technique, it is clearly realized that the
active suspension gives lower or almost no peak over shoot (amplitude) and faster settling time.

Table 1. Effects of controller parameters on closed loop system

S/N Controller Rise time Over-shoot Setting time
response
1 Kp Decrease Increase Small change
2 Ki Decrease Increase Increase
3 Kd Small change Decrease Decrease

Page 10 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Table 2. Zeigler–Nichols tuning rules

S/N Type of Kp Ki Kd
controller
1 P 0.5 Kcr 1 0
2 PI 0.45 Kcr 0.83 Pcr 0
3 PID 0.6 Kcr 0.5 Pcr 0.125 Pcr

Figure 6. Effect of rail vehicle

body acceleration on the sus-
pension system.

Body travel can reduce the amplitude and settling time compared to passive suspension system as
shown in Figure 6. The nature of the wave motion is that of a damped system.

The PID controller design approach was examined for the active and passive system in terms of
the suspension travel. From Figure 7, small amplitude value for suspension travel and wheel
deflection was observed for the suspension travel for active suspension system. The suspension
travel is the distance the suspension spring elongates or compresses when load is removed or
applied, respectively. In other words, it is a function of the spring deflection as a result of rail
disturbances and irregularities. Alternate application and removal of loads may cause stress to
build up around the wheel which can produce unwanted wheel deflection. The suspension travel in
the active suspension system was found to reduce to more than half or almost null of its value in
the passive system (Figure 7). Initially there was variation in the deflected distances but over the
period of 50 s, the suspension travel maintained a constant suspension travel distance of −3 mm.
In contract, the passive suspension system demonstrated large deflections upon the application
and removal of load resulting from rail disturbances. By implication, the smaller the deflection, the
higher the ride comfort and vice versa. By introducing an active element in the suspension, it is
possible to reach a better compromise than using purely passive elements.

From Figure 8, there is significant improvement in road holding ability of the active suspension
system compared to the passive system especially for random rail surface. The rail holding ability
is the tendency of the rail car to move steadily along the designated track without sloshing. The
value of displacement which is a function of the rail holding ability with respect to time was
negligibly small at the starting point and maintained stability (0 mm) for the period of 50 s
compared to passive system which shows variation throughout the time interval from −2 to
2 mm. This points to the fact that the controlled active suspension system could sufficiently
prevent irregular movements of the rail car thus ensuring stability, smooth and vibration-free
operation on the track as opposed to the passive system that demonstrated sloshing during
operation.

Page 11 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

Figure 7. Effect of suspension

travel on suspension system.

Figure 8. Effect of rail holding

ability on the suspension
system.

4. Conclusion and recommendations

4.1. Conclusion
The objective of this wok was achieved. A quarter rail car model with two- DOF was modelled and
PID controller was designed for its active suspension system in order to minimize the effect of rail
disturbances. Also, the hydraulic dynamics were also considered and simulated leading to the
formulation of a dynamic model for the linear quarter rail car suspension system. The PID
controller was used to test the system’s performance while the Ziegler–Nichols tuning rules were
used to determine proportional gain, reset rate and derivative time of PID controllers. The simu-
lated results proved that, active suspension system with PID control improves ride comfort requir-
ing less space for movement. However, there is significant improvement in rail holding ability
observed especially for random road surface. Besides its relative simplicity in design and the
availability of well-known standard hardware, the viability of PID controller as an effective tool
in developing active suspension system was established.

Page 12 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

4.2. Recommendations
Work-study that has been done on controlled active suspension does not consider the dynamics of
the force actuators. In this work, the mathematical modelling for quarter rail car does not include
the dynamic model of the actuators. Thus, the study of these actuators is important so that it can
give real-time performance. The parameters used also can affect the output performance either
for active or passive suspension systems. Therefore, using different type of parameters can give
different output performance. The consideration of the presence of non-linearities such as hard-
ening spring, a quadratic damping force and the tyre lift-off phenomenon in a real suspension
system is important to establish a proper non-linear quarter car suspension system. Fuzzy logic
systems can also be used to approximate these non-linear systems. The development of novel
direct fuzzy back-stepping control methods to handle the control design problems for the systems
is also recommended.

Funding
The authors received no direct funding for this research.
Author details
I. A Daniyan1
E-mail: afolabiilesanmi@yahoo.com
K. Mpofu1
E-mail: mpofuk@tut.ac.za
ORCID ID: http://orcid.org/0000-0003-3429-7677
D. F. Osadare2
E-mail: dapoosadapo@gmail.com
1
Department of Industrial Engineering, Tshwane
University of Technology, Pretoria, South Africa.
2
Department of Mechanical and Mechatronics
Engineering, Afe Babalola University, Ado Ekiti, Nigeria.
Citation information
Cite this article as: Design and simulation of a controller
for an active suspension system of a rail car, I. A Daniyan,
K. Mpofu & D. F. Osadare, Cogent Engineering (2018), 5:
1545409.
References
Abdelhady, M. (2003). A fuzzy controller for automotive
active suspension systems. SAE Technical Paper,
2003-01-1417. Retrieved from https://doi.org/
10.4271/2003-01-1417
Ahmed, M. I., Hazlina, M. Y., & Rashid, M. M. (2015).
Modeling a small-scale test rig of quarter railway
vehicle suspension system. International Journal of
Robotics and Mechatronics, 2(4), 149–153.
Allen, J. A. (2008, August) Design of active suspension
control based upon use of tubular linear motor and
quarter car model (MSc. Thesis). Texas A&M
University.
Al-Zughaibi, A., & Davies, H. (2015). Controller design for
active suspension system of quarter car with
unknown mass and time-delay. International Journal
of Mechanical, Aerospace, Industrial, Mechatronic and
Manufacturing Engineering, 9(8), 1484–1489.
Arefsoliman, M. A. (2011). Adaptive LQR control strategy
for active suspension system. SAE Technical Paper,
2011-01-0430.
Arunachalam, K., Mannar Jawahar, P., & Tamilporai, P.
(2003). Active suspension system with preview
control-A review. Technical Paper, 2003-28-0037.
Bideleh, S. M. M., & Berbyuk, V. (2016). Global sensitivity
analysis of bogie dynamics with respect to suspen-
sion components. Multibody System Dynamics, 37(2),
145–174. doi:10.1007/s11044-015-9497-0
Bideleh, S. M. M., Mei, T. X., & Berbyuk, V. (2016). Robust
control and actuator dynamics compensation for
railway vehicles. Vehicle System Dynamics, 37(2),
145-174.
Chen, B., Liu, X., Tong, S., & Lin, C. (2008). Observer-based
stabilization of T–S fuzzy systems with input delay.
IEEE Transaction on Fuzzy System, 16(3), 652–663.
doi:10.1109/TFUZZ.2007.903329
Chiou, J.-S., & Liu, M.-T. (2009). Using fuzzy logic controller
and evolutionary genetic algorithm for automotive
active suspension system”. International Journal of
Automotive Technology, 10(6), 703–710. doi:10.1007/
s12239-009-0083-4
Colombo, E. F., Gialleonardo, E. D., Facchinetti, A., &
Bruni, S. (2014). Active car body roll control in railway
vehicles using hydraulic actuation. Control
Engineering Practice, 31, 24–34. doi:10.1016/j.
conengprac.2014.05.010
Demic, M. (2003). Identification of vibration parameters
for motor vehicles. Vehicle System Dynamics, 27,
68–88.
Du, H., & Zhang, N. (2009). Fuzzy control for nonlinear
uncertain electrohydraulic active suspensions with
input constraint. IEEE Transaction on Fuzzy System,
17(2), 343–356. doi:10.1109/TFUZZ.2008.2011814
Ebrahimi, B. (2009). Development of hybrid electromagnetic
dampers for vehicle suspension systems (PhD thesis).
University of Waterloo, Waterloo, Ontario, Canada.
Eris, O., Ergen, A. F., & Kurtulan, S. (2015). A modified
delayed resonator for active suspension systems of
railway vehicles. International Federation of
Automatic Control, 281–285.
Eski, İ., & Yıldırım, Ş. (2009). Vibration control of vehicle
active suspension system using a new robust neural
network control system. Journal of Simulation
Modelling Practice and Theory, 17(5), 778–793.
doi:10.1016/j.simpat.2009.01.004
Fateh, M. M., & Alavi, S. S. (2009). Impedance control of an
active suspension system. Journal of Mechatronics,
19(1), 134–140. doi:10.1016/j.
mechatronics.2008.05.005
Gohrle, C., Schindler, A., Wagner, A., & Sawodny, O.
(2014). Design and vehicle implementation of pre-
view active suspension controllers. IEEE Transactions
on Control Systems Technology, 22(3), 1135–1142.
doi:10.1109/TCST.2013.2272342
Ha, S. H., Choi, S. B., & Lee, K. S. (2012). Ride quality
evaluation of railway vehicle suspension system
featured by magnetorheological fluid damper.
Advanced Science Letters, 12, 209–213. doi:10.1166/
asl.2012.2740
Hany, S. H., Al Adl, M. R., Samir, M. E., & Younes, K. Y. (2008).
Active suspension of highway truck seat using generic
algorithms. SAE Technical Paper, 2008- 04-14.
Hasbullah, F., & Faris, W. F. (2017). Simulation of distur-
bance rejection control of half-car active suspension
system using active disturbance rejection control
with decoupling transformation. IOP Conference

Page 13 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409
Series: Journal of Physics: Conference Series, 949,
012025. doi:10.1088/1742/6596/949/1/012025
Hasbullah, F., Faris, W. F., Darsivan, F. J., &
Abdelrahman, M. (2015). Ride comfort performance
of a vehicle using active suspension system with
active disturbance rejection control,”. International
Journal of Vehicle Noise Vibration, 11(1), 78–101.
doi:10.1504/IJVNV.2015.067995
Herbst, G. (2013). A simulative study on active distur-
bance rejection control (ADRC) as a control tool for
practitioners. Electronics, 2(3), 246–279. doi:10.3390/
electronics2030246
Hrovat, D. (1990). Optimal active suspension structures
for quarter-car vehicle models. International
Federation of Automatic Control, 26(5), 845–860.
Jin, X.-S., Xiao, X.-B., Ling, L., Zhou, L., & Xiong, J.-Y. (2013).
Study on safety boundary for high-speed trains run-
ning in severe environments. International Journal of
Rail Transportation, 1(1–2), 87–108. doi:10.1080/
232483782013790138
Jinzhi, F., Fan, Y., & Jun, L. (2002). An investigation to
controller design for active vehicle suspension by
using GA-based PID and fuzzy logic. SAE Technical
Paper, 2002- 03-04. doi:10.1044/1059-0889(2002/
er01)
Kim, H. C., Choi, S. B., & Lee, G. S. (2014, March 9).
Performance analysis of a semi-active railway vehicle
suspension featuring MR dampers. SPIE smart struc-
tures and materials + nondestructive evaluation and
health monitoring (pp. 905711). San Diego, CA.
Bellingham, WA: International Society for Optics and
Photonics (SPIE).
Kumar, P. S., Sivakumar, K., Kanagarajan, R., & Kuberan, S.
(2018). 2799 Adaptive neuro fuzzy inference system
control of active suspension system with actuator
dynamics. Journal of Vibroengineering, 20(1),
541–549. doi:10.21595/jve.2017.18379
Lin, J., Lian, R. J., Chung-Neng, H., & Wun-Tong, S. (2009).
Optimal active suspension structures for quarter-car
vehicle models. Journal of Mechatronics, 19(7),
1178–1190. doi:10.1016/j.mechatronics.2009.03.009
Ling, L., Xiao, X.-B., Xiong, J.-Y., Zhou, L., Wen, Z.-F., &
Jin, X.-S. (2014). A three-dimensional model for cou-
pling dynamics analysis of high speed train-track
system. Journal of Applied Physics and Engineering,
Pp, 1, 21.
Marzbanrada, J., Ahmadib, G., Zohoorc, H., & Hojjatd, Y.
(2004). Stochastic optimal preview control of
a vehicle suspension. Journal of Sound and Vibration,
275, 973–990. doi:10.1016/S0022-460X(03)00812-5
Matamoros-Sanchez, A. Z., & Goodall, R. M. (2015). Novel
mechatronic solutions incorporating inerters for rail-
way vehicle vertical secondary suspensions. Vehicle
System Dynamics, 53(2), 13–136. doi:10.1080/
00423114.2014.983529
Minakaldas, M. S., & Arefsoliman, M. A. (2014). Influence
of active suspension preview control on vehicle ride
and braking performance. SAE Technical Paper,
2014-01-0862.
Nakajima, T., Shimokawa, Y., Mizuno, M., & Sugiyama, H.
(2014). Air suspension system model coupled with
leveling and differential pressure valves for railroad
vehicle dynamics simulation. Journal of
Computational and Nonlinear Dynamics, 9, 1–9.
doi:10.1115/1.4026275
Nguyen, S. D., Choi, S. B., & Nguyen, Q. H. (2014). An
optimal design of interval type-2 fuzzy logic system
with various experiments including magnetorheolo-
gical fluid damper. IMechE Part C. Journal of
Mechanical Engineering Science, 228(17), 3090–3106.
doi:10.1177/0954406214526585
Nguyen, S. D., & Nguyen, Q. H. (2017). Design of active
suspension controller for train cars based on sliding
mode control, uncertainty observer and neuro-fuzzy
system. Journal of Vibration and Control, 23(8),
1334–1353. doi:10.1177/1077546315592767
Nguyen, T. T., Nguyen, V. G., & Kim, S. B. (2000). “Control
of active suspension system by using H∞ theory.
Transaction on Control Automation, and Systems
Engineering, 2(1), 1–6.
Nurkan, Y., Sakman, L., & Guclu, R. (2008). Different con-
trol applications on a vehicle using fuzzy logic
control. Journal of Sadhana, 33(1), 15–25.
doi:10.1007/s12046-008-0002-9
Oh, J.-S., Shin, Y.-J., Koo, H.-W., Kim, H.-C., Park, J., &
Choi, S.-B. (2016). Vibration control of a semi-active
railway vehicle suspension with magneto-rheological
dampers. Advances in Mechanical Engineering, 8(4),
1–13. doi:10.1177/1687814016643638
Oraby, W. A. H., Aly, M. A., El-Demerdash, S. M., &
Selim, A. M. (2007). Influence of active suspension
preview control on the vehicle lateral dynamics. SAE
Technical Paper, 2007-01-234.
Pathare, Y. S., & Nimbalkar, S. R. (2014). “Design and
development of quarter car suspension test rig
model and its simulation. International Journal of
Innovative Research and Development, 3(3), 157–170.
Qiu, J., Feng, G., & Yang, J. (2009). A new design of delay-
dependent robust H∞ filtering for discrete-time T–S
fuzzy systems with time varying delay. IEEE
Transaction on Fuzzy System, 17(5), 1044–1058.
doi:10.1109/TFUZZ.2009.2017378
Semiha, T., & Huseyin, A. (2008). Aspects of achievable
performance for quarter-car active suspensions.
Journal of Sound and Vibration, 311(1–2), 440–460.
doi:10.1016/j.jsv.2007.09.014
Shao-Juan, L., Zhong-Hua, H., & Chen, Y.-Z. (2014).
Automobile active suspension system with fuzzy
control. Journal of Central South University of
Technology, 11(2), 206–209.
Shin, D., Lee, G., Yi, K., & Noh, K. (2015). Motorized
vehicle active suspension damper control with
dynamic friction and actuator delay compensation
for a better ride quality. Proceedings of the
Institution of Mechanical Engineers Part D: Journal
of Automobile Engineering, 1–16. doi:10.1177/
0954407015598670
Sun, S., Deng, H., & Li, W. (2013). Improving the critical
speeds of high-speed trains using magnetorheologi-
cal technology. Smart Materials & Structures, 22,
115012. doi:10.1088/0964-1726/22/11/115012
Ulum, Z., Affaf, M. S., & Suparwanto, A. (2017, August
9–11). Active suspension systems design of a light
rail vehicle using MPC with preview information
disturbance. 5th international conference on instru-
mentation, control, and automation (ICA) (pp. 18–23).
Yogyakarta, Indonesia.
Wang, G., Chen, C., & Yu, S. (2017). Robust non-fragile
finite-frequency H static output-feedback control for
active suspension systems. Mechanical Systems and
Signal Processing, 91, 41–56. doi:10.1016/j.
ymssp.2016.12.039
Wolf Railway Consulting WRC (2017). Principle course. 1–81.
Yamamoto, A., Sugai, H., Kanda, R., & Buma, S. (2014).
Preview ride comfort control for electric active suspen-
sion (eActive3). SAE Technical Paper, 2014-01-0057.
Yao, J. L., & Shi, W. K. (2013). Development of a sliding
mode controller for semi-active vehicle
Page 14 of 15
Daniyan et al., Cogent Engineering (2018), 5: 1545409
https://doi.org/10.1080/23311916.2018.1545409

suspensions. Journal of Vibration and Control, 19,
1152–1160. doi:10.1177/1077546312441045
Yoshimura, T., & Takagi, A. (2004). Pneumatic active sus-
pension system for a one-wheel car model using fuzzy
reasoning and a disturbance observer. Journal of
Zhejiang University, 5(9), 1060–1068. doi:10.1631/
jzus.2004.1060
Yoshimura, T., & Watanabe, K. (2003). Active suspension
of a full car model using fuzzy reasoning based on
single input rule modules with dynamic absorbers.
International Journal of Vehicle Design, 31(1), 22–40.
doi:10.1504/IJVD.2003.002045
Zhang, B. L., Han, Q. L., Zhang, X. M., & Yu, X. (2013).
Sliding mode control with mixed current and
delayed states for offshore steel jacket platforms.
IEEE Transactions on Control Systems Technology,
22, 1769–1783. doi:10.1109/TCST.2013.2293401
Zhu, S., Du, H., Zhang, N., & Wang, L. (2014). Development
of a new model for roll-plane active hydraulically
interconnected suspension. SAE International Journal
of Passenger Cars—Mechanical Systems, 7(2),
447–457. doi:10.4271/2014-01-0053
Zong, L. H., Gong, X. L., Xuan, S. H., & Guo, C. Y. (2013).
Semi-active H∞ control of high-speed railway
vehicle suspension with magnetorheological
dampers. Vehicle System Dynamics, 51(5),
600–626. doi:10.1080/00423114.2012.758858

© 2018 The Author(s). This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.
You are free to:
Share — copy and redistribute the material in any medium or format. Adapt — remix, transform, and build upon the
material for any purpose, even commercially.
The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made.
You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
No additional restrictions
You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.

Cogent Engineering (ISSN: 2331-1916) is published by Cogent OA, part of Taylor & Francis Group.
Publishing with Cogent OA ensures:
• Immediate, universal access to your article on publication
• High visibility and discoverability via the Cogent OA website as well as Taylor & Francis Online
• Download and citation statistics for your article
• Rapid online publication
• Input from, and dialog with, expert editors and editorial boards
• Retention of full copyright of your article
• Guaranteed legacy preservation of your article
• Discounts and waivers for authors in developing regions
Submit your manuscript to a Cogent OA journal at www.CogentOA.com

Page 15 of 15