Journal of Physics: Conference Series
PAPER • OPEN ACCESS
Design of half-car active suspension system for passenger riding comfort
To cite this article: Y Susatio et al 2018 J. Phys.: Conf. Ser. 1075 012030
View the article online for updates and enhancements.
This content was downloaded from IP address 106.192.61.8 on 18/10/2020 at 11:32
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
Design of half-car active suspension system for passenger
riding comfort
Y Susatio1, L Oktaviana1, N K Rizki1, E Listijorini2, T R Biyanto3
1
Vibration and Acoustic Laboratory, Departement of Engineering Physics, Institut
Teknologi Sepuluh Nopember, Surabaya, Indonesia
2
Department of Mechanical Engineering, Universitas Sultan Ageng Tirtayasa
3
Instrumentation and Control Laboratory, Departement of Engineering Physics,
Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia
Abstract. In this paper, the active suspension system is designed to reduce vibration on the
passenger seat caused by change in road-surface shapes or disturbance. PID control is utilized
in the design of this active suspension system, and Direct Synthesis method is proposed to tune
PID parameters. Direct Synthesis method optimize PID tuning parameters to achieve desired
output response in different road surface. In this study, three interference signals are used as
disturbance from surface road i.e step, impulse, and sinus. The aim of the suspension control
design is to obtain the control parameters that conform to ISO 2631. The simulation results
show that value of τc 0.0002 for step disturbance provide aRMS 0.086 m/s2 and damping ratio
0.21. For impulse input the value of τc 0.002 provide aRMS 0.112 m/s2 and damping ratio of
0.22. For sinus input value of τc 0.00005 provide aRMS 0.09 m/s2.
1. Introduction
Nowadays, riding comfort becomes the main focus in the automotive industries. The passengers feel
the vibration effects due to interaction of the vehicle and road surfaces. Many researchers said that
vibration can degrade human fitness due to discomfort and fatigue [1]. There are several standards
about whole body vibration that aim for restricting the allowable magnitude of vibration on human
body such as ISO 2631 [2].
The vehicle suspension system is responsible for ride comfort and road holding as the suspension
carries the vehicle body and transmit all forces between the body and the road. A classical car
suspension or passive suspension system consists of a spring (coil spring, air spring or leaf spring) and
a damping element. The spring and damping coefficients are chosen according to comfort, road
holding, and handling specifications. However, conventional suspensions can achieve a trade-off
between ride comfort and road holding since their spring and damping coefficients cannot be
adaptively tuned according to driving efforts and road conditions. They can achieve good ride comfort
and road holding only under the designed conditions. To avoid the tradeoff, intelligent suspension
systems have been investigated since the 1980s with the development of microprocessor, sensor, and
actuator technologies; their spring and damping coefficients can be controlled. The active suspension
usually requires a substantial amount of external energy to generate the required control forces. As
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
intelligent suspension systems, active suspensions belong to the controlled suspension system, which
consists of sensors, controllers, plants and actuators. Therefor the active suspension can improve
riding comfort against vibration due to road surfaces change [3].
Recent years, a lot of studies about control method used in active suspension have been engaged.
PID controller still widely used because of its simple structure and robust performance in linear model
system. The application of PID on active suspension has been developed by many researchers and the
result shows the response of active suspension of sprung mass decrease overshoot peak as compare
with its passive suspension [4] [5]. However, it is necessary to build controller which will bear with
any road changes. In this research Direct Synthesis method is used to optimize PID tuning parameters
to achieve desired output response in different road surfaces. The presented controller is tested with
three road surfaces i.e step, impulse and sinus. The robustness of control system will performed within
simulation in MATLAB SIMULINK.
2. Mathematical Modelling
The 5DOF active suspension shown in Figure 1. The suspension, tire, passenger seat are modeled by
linear springs in parallel with dampers. The Ys expresses seat displacement while Y expresses sprung
mass vertical displacement (bounce) at the center of gravity and Ө is the pitch displacement of sprung
mass. Front and rare tire displacement are denoted by Y 1 and Y2. The device U set at the front
suspension as the actuator which can be controlled. The Z1 and Z2 are the disturbance which hit the
vehicle at front and rare tire.
Figure 1. Half-car active suspension model
The equation of motion of half-car active suspension system is being carried out by using Newton
second law of motion and written bellow as
𝑚𝑠 𝑦̈𝑠 + 𝑐𝑠 (𝑦̇𝑠 + 𝑎𝜃̇ − 𝑦̇ ) + 𝑘𝑠 (𝑦𝑠 + 𝑎𝜃 − 𝑦) = 0 (1)
𝑚𝑏 𝑦̈ + (𝑐1 + 𝑐2 + 𝑐𝑠 )𝑦̇ − 𝑘2 𝑦2 − 𝑘𝑠 𝑦𝑠 + (𝑐1 𝑙1 − 𝑐2 𝑙2 − 𝑐𝑠 𝑎)𝜃̇ − 𝑐1 𝑦̇ 1 − 𝑐2 𝑦̇ 2 + (𝑘1 + 𝑘2 + 𝑘𝑠 )𝑦 +
(𝑘1 𝑙1 − 𝑘2 𝑙2 − 𝑘𝑠 𝑎)𝜃 − 𝑐𝑠 𝑦̇ − 𝑘1 𝑦1 = 𝑈 (2)
𝐼𝜃̈ + (𝑐1 𝑙1 − 𝑐2 𝑙2 + 𝑐𝑠 𝑎)𝑦̇ − 𝑘𝑠 𝑎𝑦𝑠 − 𝑐𝑠 𝑎𝑦̇ + (𝑐2 𝑙22 + 𝑐1 𝑙12 + 𝑐𝑠 𝑎2 )𝜃̇ + 𝑐1 𝑙1𝑦̇ 1 + 𝑐2 𝑙2𝑦̇ 2 +
𝑘2 𝑙2𝑦2 + (𝑘2 𝑙2 − 𝑘1 𝑙1 + 𝑘𝑠 𝑎)𝑦 + 𝑘1 𝑙1𝑦1 + (𝑘2 𝑙22 + 𝑘1 𝑙12 + 𝑘𝑠 𝑎2 )𝜃 = 𝑈𝑙1 (3)
̇ + 𝑘 )𝑦 = 𝑘 𝑧
𝑚𝑡2 𝑦̈ 2 − 𝑘2 𝑦 + 𝑘2 𝑙2𝜃 − 𝑐2 𝑦̇ + 𝑐2 𝑙2𝜃̇ + 𝑐2 𝑦̇ 2 + (𝑘 (4)
2 𝑡2 2 𝑡2 2
̇
𝑚𝑡1 𝑦̈ 1 + 𝑐1 𝑦̇ − 𝑘1 𝑥𝑏 − 𝑘𝑠𝑓 𝑙2𝜃+(𝑘𝑠𝑓 + 𝑘𝑡𝑓 )𝑦 − 𝑐1 𝑙1𝜃̇ + 𝑐̇ 1 𝑦̇ 1 = 𝑘𝑡1 𝑧1 + 𝑈 (5)
2
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
In order to optimize the simulation and to analyze the half-car suspension behavior, the parameters
are input as follow:
Table 1. Suspension parameter of half-car
Description Value
Driver Seat Mass (ms) 75 Kg
Sprung Mass (mb) 505.1 Kg
Front unsprung mass (mt1) 28.58 Kg
Rare unsprung mass (mt2) 54.3 Kg
Moment of inertia of the vehicle (I) 651 Kg.m
Rare body length from the CG (l2) 1.468 m
Front body length from the CG (l1) 1.098 m
Driver seat length from the CG (a) 0.7 m
Spring stiffness of front and rare tire (kt1, kt2) 155900 N/m
Spring stiffness of front and rare unsprung mass (k1, k2) 15000 N/m
Damping coefficient of front nad rare unsprung mass (c1, c2) 1828 N.s/m
Spring stiffness of driver seat (ks) 15000 N/m
Damping coefficient of driver seat (cs) 150 N.s/m
2.1. Riding Comfort Standard Condition
ISO 2631 has been recognized as minimum standard for whole body vibration on passenger as
described below:
Table 2. ISO 2631 standard for riding comfort
RMS (Root Mean Square) Acceleration Ride comfort Criteria
a<0.315 m/s2 Comfortable
0.315<a<0.63 m/s2 A little Uncomfortable
2
0.5<a<1m/s Fairy Uncomfortable
0.8<a<1.6m/s2 Uncomfortable
1.25<a<2.5m/s2 Very Uncomfortable
a>2.5m/s2 Extremely Uncomfortable
Another riding comfort standard comes from Optimum G, an England automotive consultant,
claims that passenger vehicles generally use a damping ratio (ξ) of approximately 0.2- 0.25 for
maximizing riding comfort [6]. Furthermore, the controlled responses should be conformed to those
riding standards.
3. PID Tuned by Direct Synthesis
The first step to tune PID parameters using direct synthesis method is recognizing open loop
behaviour of the system. Previous study has successfully arranged the equation to tune Kp, Ki, and Kd
on the second order system shown in the equation (6)-(8) [7]:
𝑘𝑐 = (𝜏1 + 𝜏2 )/𝑘𝑝 𝜏𝑐 (6)
𝜏𝑖 = (𝜏1 + 𝜏2 ) (7)
𝜏𝑑 = 𝜏1 𝜏2 /(𝜏1 + 𝜏2 ) (8)
𝜏𝑐 is the parameter in direct synthesis method that can be changed to reach the best response of the
system. When the disturbance shape change, system may be unsteady and unrobust. Therefor in this
3
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
present attempt 𝜏𝑐 parameter used to build an adaptive control system to any different road surfaces.
Table 3, table 4, and table 5 show PID gain parameters for step input, impulse input and sinus input
with several varieties of 𝜏𝑐 . The range of 𝜏𝑐 would be different for each input shapes since this value
depend on the settling time and gain of each system.
Table 3. Gain parameters for step input
Gain τc
parameter 1 0.1 0.01 0.001 1E-04 2E-04
Kp 9 93 927 9268 92680 46340
Ki 3 34 336 3358 33580 16790
Kd 6 64 639 6395 63949 31974
Table 4. Gain parameters for impulse input
Gain τc
parameters 1 0.1 0.01 0.001 0.002
Kp 39.5 395.3 3952.6 39526.4 19763.2
Ki 9.1 91.1 910.7 9107.5 4553.7
Kd 42.9 428.9 4288.6 42886.2 21443.1
Table 5. Gain parameters for sinus input
Gain τc
parameters 0.001 0.0001 0.00005
Kc 10174.42 101744.2 203488.4
Ki 4069.767 40697.67 81395.35
Kd 6359.012 63590.12 127180.2
4. MATLAB simulation and result analysis
The MATLAB SIMULINK simulation is prepared for the passenger seat displacement and
acceleration responses to analyse the achievement of riding comfort. The goals of this study will show
the comparison between active suspension system response and its passive suspension for each road
shapes.
4.1. Step road profile
The first road profile is step input with 0.08 m high. Figure 2 shows passenger seat displacement
response while figure 3 shows passenger seat acceleration response.
Figure 2. Seat displacement response due to Figure 3. Seat acceleration response due to step
step input with τc value 0.0002 input with τc value 0.0002
4
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
Figures show the maximum overshoot of the respon from 0.0227 m of passive to 0.0021 m of
active, and lessens the aRMS from 0.39 m/s2 to 0.08 m/s2 and gives the controlled response with 0.21
damping ratio. So that the active response meets all qualifications of riding comfort standard.
4.2. Impulse road profile
The second road profile formed as impulse with 0.08 m high and hit by the car for 0.06 seconds while
it is traveling at 20 km/hour. Figure 4 shows passenger seat displacement response and figure 5 shows
acceleration response.
Figure 4. Seat displacement response due to Figure 5. Seat acceleration response due to step
impulse input with τc value 0.002 impulse with τc value 0.002
Figures observe that the result of active suspension system degrade maximum overshoot from
0.087 m of passive to 0.0012 m of active. Active suspension system with direct synthesis obtain aRMS
0.112 m/s2 and damping ratio (ξ) 0.22, these two results absolutely meet riding comfort standard.
4.3 Sinus road profile
The third road profile formed as sinus wave. It assumed that the road is approximated as sinusoidal in
cross section with amplitude of 0.035 m and the wavelength (λ) of 5 m. The car is traveling at 20
km/hour.
The period and cycle frequency for the harmonic input is as follow
𝑍1 = 0.035 sin 17.27𝑡 (9)
Figure 6 shows passenger seat displacement response and figure 7 shows passenger seat
acceleration response.
Figure 6. Seat displacement response due to Figure 7. Seat acceleration response due to
sinus input with τc value 0.00005 sinus input with τc value 0.00005
5
Regional Conference on Acoustics and Vibration 2017 (RECAV 2017) IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1075 (2018)
1234567890 ‘’“” 012030 doi:10.1088/1742-6596/1075/1/012030
Figures observe that active suspension gives the better result. With maximum overshoot depression
from 0.010 m of passive to 0.003 to active. And reach aRMS 0.099 m/s2. This result successfully
conform to ISO 2631 standard.
5. Conclusion
This research presents that direct synthesis is the appropriate method for active suspension system.
aRMS and damping ration obtained by active suspension system can adapted to three different road
profiles and fulfill the requirements of passenger car riding comfort standard. It can be surely denoted
that direct synthesis is the robust control strategy for automobile model
References
[1] Gagorowski A 2010 Simulation Study on Stiffness of Suspension Seat in The Aspect of the
Vibration Assesment Affecting a Vehicle Driver Logistics and Transport vol 11 no 2 pp 55-62,
[2] ISO 2631 1997 Mechanical Vibration and Shock Evaluation of Human Body Exposure to Whole
Body Vibration: General Requirement (International Organization of Standarization)
[3] Honghai L, Huijun G and Ping L 2013 Handbook of Vehicle Suspension Control System,
(London: The Institution of Engineering and Technology) pp 1
[4] Al-Mutar W H and Abdalla T Y 2015 Quarter Car Active Suspension System Control Using PID
Controller tuned by PSO Iraq J. Electrical and Electronic Engineering vol 11 no 2 pp 151-154
[5] Al-amir H S and Al-zahra A T 2014 Design a Robust PID Controller of an Active Suspension
System International Journal of Mechanical and Mechatronics Engineering vol 14 no 2 pp
122-127
[6] Brisson S and Giaraffa M Tech and Tip Spring & Damper Ep Four Optimum G, Denvor.co
[7] Emad A 2015 Chapter 4. PID Control Design Advanced Topic in Process Control (Riyadh,
Saudi Arabia, King Saud University) pp 1-27
