Available online at www.sciencedirect.
com
ScienceDirect
Energy Procedia 105 (2017) 2653 – 2659
The 8th International Conference on Applied Energy – ICAE2016
Research on air suspension control system based on fuzzy
control
Gao Zepeng a,Nan Jinruia,b*,Liu Liana,Xu Xiaolina
a
Beijing technology and engineering university, No.5 YardˈZhong Guan Cun South StreetHaidian DistrictˈBeijing, 100081,China
b
Collaborative innovation center in Beijing, No.5 YardˈZhong Guan Cun South StreetHaidian DistrictˈBeijing, 100081,China
Abstract
In order to solve the suspension overshoot phenomenon existing in the static regulation process of the electric vehicle
with electric controlled air suspension, firstly, the relevant factors of air spring are analyzed and the characteristics of
the gasbag are simulated and verified in AMESim. Secondly, the model of the electric vehicle body is analyzed
according to the law of dynamics and the fuzzy control theory is used to set up the electric vehicle body model in
Simulink. In the case of unbalanced load, the effectiveness of the fuzzy controller is used to simulate the phenomenon
of "overshoot" in the system.
© 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review and/or
Selection peer-reviewofunder
under responsibility responsibility
the scientific of of
committee ICAE
the 8th International Conference on Applied Energy.
Keywoeds: Suspension overshoot phenomenon ;electronically controlled air suspension; fuzzy controller
Introduction
With the rapid development of electronic control technology and the wide application of vehicle
control system and the new air suspension produced continuously[1], it’s promoted that electronically
controlled air suspension with excellent performance in advanced vehicles[2] and technologies in related
fields have also become the focus of vehicle engineering research. Body height adjustment is one of the
main features of ECAS, and it can actively regulate and control the height of the vehicle body according
to actual driving conditions and operation requirements and it has important significance to improve the
ride comfort, passing quality, the control stability and the fuel economy of the vehicle.
Domestic and foreign scholars have carried out extensive research on air spring suspension and its
related fields. Yoyofuku Katsuya et al studied the relationship between the vibration frequency and the
spring response and analysed the effect of the pipeline and the gas chamber on the spring characteristics
was analyzed[3]. Homeycr et al optimize the air spring structure. Domestic research institutes and
*Nan Jinrui, Tel.: +86-010-68914070; fax: +86-010-68914070
E-mail address: nanjinrui@bit.edu.cn
1876-6102 © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 8th International Conference on Applied Energy.
doi:10.1016/j.egypro.2017.03.770
2654 Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659
universities focus on the transmission characteristics, control and simulation of air suspension vehicle
model[4-5]. Research on air spring suspension, the domestic and international main focus on air spring and
automobile suspension matching design, simulation analysis, control.
In this paper, we mainly study the performance of the air spring for the electronic controlled air
suspension system[6] and build the vehicle system model of air suspension based on the characteristics
analysis of the system. We should make simulation and analysis of the height adjustment of the vehicle
body because of existing unbalanced load in stationary state. And establishing a good control effect of the
fuzzy controller in the case of only a high degree of sensor so that we can carry on the research of the
height adjustment and the actual vehicle test in order to achieve the control requirement.
1. Air suspension system calculation
1.1 Characteristic analysis of air spring
The electronic control air suspension system uses ECU to manipulation data. The gasbag achieves
charging and discharging through controlling the electromagnetic valve to achieve the purpose of the
adjustment of the air suspension. In the working process, the gasbag is affected by the load. The height
and force area of the gasbag and the volume and pressure of the air spring change over time. Without
considering operation of gasbag charging and discharging, the gaseous mass in the gasbag keep constant.
It's assumed that working process of the gas heat exchanging is considered as a polytropic processes. The
air changes from state 1 to state 2, according to the ideal gas state equation we can get:
m m
PV
1 1 PV
0 0
Where, P1 is absolute air pressure for a moment, V1 is instantaneous air bag volume, P0 is absolute pressure
of gas in static equilibrium position, V0 is air bag volume in static equilibrium position, m is a variable
exponent and is determined by actual situation.
The force generated by the compression of the air spring due to compression should be balanced with
the additional load, we can get:
V
F P1 Pa A [ P0 ( 0 )m Pa ] A (2)
V
Spring displacement at balance position is 0, the direction of spring compression is positive direction,
dV
there is A , at the balance position, s 0 ǃ V V0 ǃ P0 P1 , spring stiffness can be expressed as:
ds
dF dA dP dA A2 m( P )
K0 P1 Pa A 1 ( P0 Pa ) 0 0 (3)
ds ds ds ds V0
Where, A0 is effective area at the static equilibrium position.
dA
The effective area of the static equilibrium position is basically unchanged, there is 0 , the
ds
pressure of the air bag is much greater than atmosphere, so P0 Pa | P0 , gasbag natural frequency can be
expressed as:
1 gdA A0 gm P0 1 A0 gm P0 1 gm
f = (4)
2S A0 ds P0 Pa V0 2S P0 Pa V0 2S h0
Where, h0 is theoretical height of the air spring.
1.2 Characteristic analysis of vehicle dynamics
Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659 2655
Fig. 1. (a) two degrees of freedom 1/4 vehicle body model; (b) seven degrees of freedom vehicle body model
When studying the vehicle ride comfort, it is necessary to consider the motion of the vehicle in the
vertical direction. The vehicle with air suspension system is simplified to two degrees of freedom 1/4 car
body model. In this case, the body and chassis are treated as a rigid body with the only mass but no elastic
elasticity. And damping is neglected in the process of tire deformation so that tire is similar to a single
degree of freedom spring with a certain stiffness.
1/4 vehicle body model dynamic differential equation is as follows:
m2 z 2 c z2 z1 k z2 z1 0
°
® (5)
¯m1 z1 c z1 z2 k z1 z2 kt z1 q 0
°
And seven degrees of freedom vehicle model dynamic differential equation are as follows:
mb zb CsA zwA zbA ksA zwA zbA CsB zwB zbB ksB zwB zbB
(6)
CsC zwC zbC ksC zwC zbC CsD zwD zbD ksD zwD zbD
I PT b ¬ªCsC zwC zbC ksC zwC zbC CsD zwD zbD ksD zwD zbD ¼º
(7)
a ª¬CsA zwA zbA ksA zwA zbA CsB zwB zbB ksB zwB zbB º¼
I rI ª¬CsA zwA zbA ksA zwA zbA CsC zwC zbC ksC zwC zbC º¼ B fl
(8)
¬ªCsB zwB zbB ksB zwB zbB CsD zwD zbD ksD zwD zbD ¼º B fr
mwA zwA ktA z gA zwA ksA zbA zwA CsA zbA z wA
°
°mwB zwB ktB z gB zwB k sB zbB z wB CsB zbB z wB
® (9)
°mwC zwC ktC z gC zwC k sC zbC z wC CsC zbC z wC
°mwD zwD ktD z gD zwD k sD zbD zwD CsD zbD z wD
¯
In fact, Single axis distance is large while the body roll and pitch angle is very small, as follows:
zbA zb aT 0.5B f I
° zbB zb aT 0.5B f I
®z (10)
° bC zb bT 0.5BrI
¯ zbD zb bT 0.5BrI
2. Modeling and Simulation of air suspension system
2.1 Air spring model in AMESim
2656 Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659
To study the influence of vertical vibration on vehicle ride comfort, it can be simplified as the linear
vibration model of 1/4 body with two degrees of freedom. 1/4 body simulation model setting parameters
and gasbag simulation parameters are shown in table 1.
Table 1. Model parameter
Damping Road
coefficient Tire Piston Initial
Simulation setting Sprung Unsprung Temperature surface
of shock stiffness kt diameter pressure
parameters mass (kg) mass(kg) excitation
absorber (K)
( kN/m) (mm) (MPa)
c(kNs/m) (m)
Air spring 927.5 87.5 3.5 500 116 0 293.15 0
According to the working mechanism of the object, the model of the vehicle suspension system and the
simulation model of air spring are established. The structures are as shown in fig2.
Fig. 2. (a) air spring simulation model; (b) 1/4 car body model with air spring
2.2 Vehicle model in Simulink
Selecting the grade B road. In the case of full load, the vehicle dynamic model is shown in fig3.
Fig. 3. (a)two degrees of freedom 1/4 car body model with air bag;(b) seven degrees of freedom vehicle model
2.3 Fuzzy controller model
There are four methods to study the control strategy of air suspension system: optimum control,
predictive control, fuzzy control and neural network control. Many researchers focus on the optimal
control research, but its practical application is limited. Predictive control has cost and reliability
Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659 2657
problems so it’s still in research. Fuzzy control contains a large number of people's control experience
and knowledge accumulation in control process, and people's intelligent behavior is similar to it. So it has
an obvious advantage in the application. Neural network control with learning and massive parallelism, so
it has wide application prospect. In the design, the selection of fuzzy control theory is studied.
We choose a fuzzy controller with three inputs and one output to solve the problem of obvious
difference between left and right in the process of height adjustment, as shown in fig4.
Fig. 4. Three input one output fuzzy controller
3. Simulation analysis results
Fig. 5. (a) height adjustment with no fuzzy control;(b) height adjustment with fuzzy controller
As shown in fig5. In the actual situation, left and right loads are uneven. In the process of regulation,
reducing the height difference between left and right sides, but it still has obvious difference. We choose a
fuzzy controller with three inputs and one output to solve the problem of obvious difference between left
and right in the process of height adjustment. The greater the compression degree of air spring, the
stiffness increases more rapidly. Under the driving condition of sharp turning, rapid acceleration and
braking, the air spring stiffness increases rapidly, which limits the movement of the vehicle body and
improves the operation stability.
Fig.6. (a) Comparison of the effect of the controller;(b) Controller to adjust the height difference
2658 Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659
The semi-vehicle model with fuzzy controller is compared with the semi-car model without fuzzy
controller, as shown in fig6. In the process of vehicle body height adjustment, established three inputs and
one output fuzzy controller is good to restrain the uneven vehicle body caused by uneven load. Compared
with no control and two input fuzzy controller, the three input fuzzy controller is able to control the height
of the vehicle model in a more efficient way.
4. Test validation
There is error between the simulation and the actual results, therefore need to carry on the validation in
actual vehicle. JMC V348 rear suspension is changed into the air suspension in the process of the
experiment and vehicle height adjustment test is carried out. Regulation is divided into the process of the
rise and the fall. The overall control strategy is divided into the power on self-test, automatic height
adjustment, manual adjustment and fault detection. Changing from 2 to 3 and 2 to 1, the curves are as
shown in fig7. The speed adjusting gradually slow down with fuzzy control and there is no obvious sense
of shock. The ride comfort is improved. It‘s showed the effectiveness of the fuzzy controller. In general,
the phenomenon of "over regulation" is avoided. Furthermore, this demonstrates the effectiveness of the
fuzzy controller.
Fig. 7. (a) Change from 2 to 3; (b) Change from 2 to 1
5. Conclusions
a) According to the mechanical equation and gas state equation, using AMESim to analyze the
characteristics of air spring and establish the dynamic differential equations of 1/4 vehicle model and
seven degrees of freedom vehicle model with air suspension system. Then, establishing corresponding
body model in Simulink. The results show that the ECAS system can improve the vehicle ride comfort.
b) Aimed at the "Suspension overshoot" phenomenon appeared in the process of simulation, the fuzzy
control theory is applied into the process of air suspension control. Height difference and its change rate
and the left and the right height difference as input and the PWM signal controlling solenoid valve open
and close as the output. Comparing with the simulation results, the fuzzy controller can solve the problem
of "overshoot" and improve the control effect.
Gao Zepeng et al. / Energy Procedia 105 (2017) 2653 – 2659 2659
References
[1] Peter Holen, Boris Thorvald, Aspects on Roll and Bounce Damping for Heavy Vehicles. SAE Paper 2002-01-3060
[2] Ashley T . Dudding and William Wilson . Dewlopment of a New Front Air Suspension and Steer Axle System for On
Highway Commercial Vehicles. SAE Paper 2000—01—3449
[3] Woodrooffe John. Heavy Truck Suspension Dynamics:Methods for Evaluating Suspension Road Friendliness and Ride
Quality[J]. SAE Paper 962152
[4] Jianwen Zhang, Yang Xinglong, Lin Yi et al. Performance simulation research on bus with air suspension [C]. New York:
SAE Inc, No. 2002-01-3093
[5] Jianwen Zhang, Ersheng Guo, Zhiguo Huang et al. Simulation and analysis of ride comfort of air suspension bus [J].
Automobile Technology, 2005(4):1-5.
[6] Inseob Jang Hyunsup Kim Hyeongchul Lee and Sangsoo Han.Height Control and Failsafe Algorithm for Closed Loop Air
Suspension Control System.International Conference on Control,Automation and Systems.2007, 10:373-378.
Biography
Nan Jinrui served as deputy head in 2010, 863 electric vehicle research and development
projects in the electric air-conditioning system. He is an associate professor in Beijing
Institute of Technology. His research interests include battery management system,
vehicle control and vehicle bus.
