Modeling and simulation of active hydro-pneumatic suspension system through bond graph/Aktyvios hidro-pneumatines pakabos sistemos modeliavimas ir imitavimas naudojant grafy teorija.
Emami, M.D. ; Mostafavi, S.A. ; Asadollahzadeh, P. 等
1. Introduction
The modeling of physical systems is of great importance within all
engineering fields because it allows us to understand the behavior of
the system without having to experiment on it. It also allows the
determination of certain characteristics of the system, and can give
important information on operating conditions with the use of relatively
simple and inexpensive procedures. Moreover, it is an essential tool for
the design of control strategies, which are very important at industrial
level [1].
Nearly all the systems that an engineer may work with are dynamic,
i.e., things are usually changing so that there is no status quo or
lasting steady state. To be sure, many of important facets of a design
may be based on steady-state consideration, but a new device or system
will fail if it cannot withstand transient peak loads, respond quickly
enough to a changing input, or operate without violent oscillations when
disturbed. Dynamic analysis can predict such problem before a system is
built; a system analysis which does not include the effects of
significant dynamic phenomena is very likely to be worthless [2, 3].
Therefore, it is of utmost importance to have models that are able to
predict the dynamic behavior of the system. One of these models, which
is the focus of the present paper, is the Bond Graph (BG) model.
The Bond graph model was first developed in 1961 at MIT, Boston, by
Paynter [4] and further by Karnopp, Rosenberg [5] and Thoma [6].
BG are a multidisciplinary and unified graphical modeling language
which provides, from this point of view, a convenient and useful tool
for model builder conception. BG is a modeling and simulation tool,
providing many possibilities. It allows both a causal and a behavioral
system analysis. From the behavioral point of view, BG tool allows to
deal with the enormous amount of equations describing the dynamic
behavior of different phenomena which occur in the system. It allows,
independent of the physical nature of the studied system, precisely by
its graphical nature, to display the exchange of power in a system,
including storage and transformation and the instrumentation diagram
(sensor location in the real process). Furthermore, BG is subject to
evolution, that is to say, the model can be refined by adding
graphically more elements like thermal losses or inertia and storage
effects, without having to start all over again. This property is very
attractive and is one of the main advantages of the BG method [7]. An
exhaustive review of the BG modeling and its characteristics may be
found in [8-11]. For a com prehensive review of the applications of BG,
the reader is referred to [12].
The aim of the present paper is to use BG method in modeling of
hydro-pneumatic suspension. A hydropneumatic spring consists of two
fluids acting upon each other, usually gas over oil. A compressible gas,
such as nitrogen is used as the springing medium, while a hydraulic
fluid is used to convert pressure to force. In a pneumatic or air
spring, the external force directly compresses the gas, whereas in a
hydro-pneumatic suspension, hydraulic fluid is used [13-17].This system
consists of mechanical, hydraulic, thermo-fluid and control subsystems.
Therefore, it is a multidomain system and BG is a good candidate to
model it.
The organization of this paper is as follows: section 2 deals with
a brief introduction to hydro-pneumatic suspension systems. In the third
section, the present system is described. In section 4, BG model of an
active hydro-pneumatic suspension system and its equations are
presented. In section 5, the simulation results and validation studies
are reported. Finally, the conclusions drawn from the current work is
presented.
2. Hydro-pneumatic suspension
2.1. Historical overview
Hydro-pneumatic suspensions have been introduced on battle tanks in
the 1950's. The first hydropneumatic struts were fitted to a
prototype tracked vehicle, as a result of research done by two German
companies; Frieseke and Hopfner from Erlangen, and Borgwald from Bremen,
into the use of compressible fluids in suspension systems [13].
This type of suspension system is popular due to its nonlinear
characteristic and versatility. The nonlinear characteristic causes the
spring rate to increase as the load is increased. It also reduces body
roll and pitching, results in more constant wheel loads and usually
eliminates the necessity for a sophisticated bumps top. Many
controllable suspension systems make use of hydro-pneumatic springs
because the hydraulic fluid can easily be channeled through ducts,
orifices and valves. By adding or removing hydraulic fluid, the vehicle
dynamics and the ride height can be altered [13].
The main objective of hydro-pneumatic suspension is to eliminate
the body roll of a car when cornering at a high speed. Additionally, the
hydraulic system can enable self-leveling, variable ride height, and
assisted jacking. Furthermore, it can provide the hydraulic power to
assist the braking systems and the power steering. In some car models,
the high-pressure hydraulic system also operates the clutch and gear
change. Generally speaking, the idea of using a hydraulic circuit as a
suspension instead of springs and dampers is quite effective, and has
been in use for half a century. The first automotive company to use this
kind of suspension was Citroen [14].
Main components are: reservoir, high-pressure pump, main
accumulator, the "load-bearing" shockabsorbers, the height
control valves.
2.2. The suspension subsystem
The suspension subsystem is fed directly from the main accumulator.
The fluid feed immediately splits into front and rear branches, each
passing through a height control valve (Fig. 1). When each valve is
activated, highpressure fluid inflates the pair of load-bearing shocks.
When the valve is in the neutral position, the pressure level remains
constant between the pair. When the valve is deactivated, the fluid in
the shock pair drains directly back to the reservoir.
[FIGURE 1 OMITTED]
Sharing the pressure between left and right shock absorbers proved
to have many benefits. The tendency to equalize pressure between them
accomplished horizontal self-leveling, even at high speeds. But it
proved more advantageous to have pressure separated fore and aft. This
was accomplished through the independent height control valves on the
front and rear axis. If the load on the rear of the car increased, the
rear valve would be activated and a greater volume of high-pressure
fluid would be allowed into that pair.
3. Setup of the intended system
The focus of this paper lays on an Active Hydro-Pneumatic
suspension (AHP). At the heart of each AHP suspension strut, there is a
force controller, which is responsible for tracking a certain desired
force (calculated by other, higher level or "outer loop"
controllers). The innermost control task is to set and track certain
pressure in the cylinder, which translates into the force then exerted
by the piston. The reference force (pressure) is the result of several
higher (outer) control loops that, for instance, compensate the cars
tendency to roll in corners or pitch with changing longitudinal
acceleration [15].
It is important to note that everything discussed here concerns a
single wheel only. In the car, however, (except for the pump) four of
these systems are required, one for each wheel.
3.1. Components
The setup of the hydro-pneumatic system is shown in Fig. 2.
[FIGURE 2 OMITTED]
Here, one suspension strut is made up of a: -cylinder, which is
connected directly to the wheel; -hydraulic capacitor or "gas
spring", consisting of two chambers, one connected to the oil
circuit, the other, separated by a membrane, contains gas;
-laminar resistance between the cylinder and the capacitor;
-hydraulic pump connected to the car's engine;
-4/3 servo valve which controls the in- and outflow of oil to and
from the system.
Pressures in the cylinder and capacitor (pz and ps), the system
pressure psys and the pressure in the reservoir pres are measured by
suitable sensors and are available for the use in controllers.
This could be the oil leaving the system through some worn out
fittings, especially in the valve, where it may be that a certain amount
of oil does not flow into the system, but directly into the reservoir
for instance. We then have the control current for the valve I, which,
for positive I, injects oil into the system, and for negative I allows
oil to leave it. The position of the plunger is xrel; it is zero at the
neutral (middle) position, positive if it is "above" that
position, negative when it is "below" it. We also allow for
some external force Fext(t), which could result from the car running
over a bump on the road for instance.
As mentioned above, a certain pressure in the cylinder translates
(via the effective surface of the plunger) into a force. This force,
diminished by some friction, will accelerate the body of the car sitting
on top of the cylinder.
3.2. Inner control loop
We shall now take a quick look at the existing controller in the
inner loop. By its structure it is a PI controller. Its input is the
difference between current and reference force exerted by the cylinder,
and its immediate output is a desired oil flow into (or out of) the
system. As the actuator is the valve, which takes a specific control
current and "translates" it into the wanted oil flow, an
inverse valve model is used to determine the necessary control current
needed.
3.3. Outer control loops
The cascaded control system of AHP suspension contains two
"layers". Several (parallel) components calculate the desired
force (or reference force) on the outer layer, which is then to be set
and tracked by the inner loop.
4. Bond Graph model
The modeling of the system has been performed through BG simulation
technique which is an effective tool for modeling and simulation of
physical systems. It facilitates the exchange, storage and dissipation
of energy among interacting physical elements efficiently. The bonds of
the model portray the paths of the exchange of power within the
constraint structure and atomic elements. It is to be noted that all
bond graphs including the present one are lumped element representation
[16].
BG model of the system shown in Fig. 3 is described as follow.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
The force exerted to the vehicle from the road is denoted by
[SE.sub.1] , which is considered an external force equal to [F sub.ext.]
[SE.sub.2] is the weight of the vehicle and [I.sub.4] denotes a quarter
of the vehicle inertia. Friction coefficient between the cylinder wall
and the piston is denoted by [R.sub.3]. Element [C.sub.5], which is an
activate bond, is an observer that monitors piston displacements.
Element TF is used to convert force to pressure, and velocity to
volume-flow-rate. In fact, this element transfers energy from mechanical
domain into the hydraulic domain. The volume flow rate of hydraulic oil,
[Q.sub.v] in Fig. 2, is considered as [SF.sub.8]. This parameter is
considered as an input to the system, and is calculated via Eq. 6.
[SF.sub.9] is considered as leakage flow rate, [Q.sub.1] in Fig. 2,
which is neglected in the present study. Oil compressibility is defined
by element [C.sub.10]. Element [R.sub.12] is the resistance due to
pressure drop in the orifice located between the cylinder and the
capacitor. The oil volume entering the capacitor is described by element
[C.sub.14] which is activated because the hydro-static pressure of oil
in the capacitor is neglected in the present study. The last portion of
the bond graph in Fig. 3 is the nitrogen tank, for which the ideal gas
law is considered to prevail. As seen in Fig. 3, the entire system
states are [q.sub.5], [p.sub.4], [q.sub.10], and [q.sub.14]. These are,
respectively, the piston displacement, the piston speed, the in-cylinder
oil volume, and the oil volume inside the capacitor.
In the aforementioned model, the control subsystem has not been
considered and the controller output (I) has been regarded as a model
input (Fig. 4), which motivates the valve. As mentioned earlier, for
positive I values, oil is injected into the system, and for negative I
values, oil is allowed to leave the system.
4.1. The governing equations of the system
The system equations derived from the BG model are described as
follows:
First equations
[q.sub.5 = [f.sub.5] = [f.sub.4] =[P.sub.4] / [I.sub.4] [right
arrow] [q.sub.5] = [P.sub.4] / [I.sub.4] (1)
Second equations
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Third equations
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)
Fourth equations
[q.sub.4] = [f.sub.14] = [f.sub.12]; [q.sub.14] = 1/[R.sub.d] [
1/[C.sub.10][q.sub.10] -[[P.sub.a]([V.sub.a] / [V.sub.M] -
[q.sub.14]).sup.5]](4)
In the above equations, friction force ([F.sub.fr]) and oil flow
rate from valve ([Q.sub.v]) are defined as in [15]
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)
Where Sat (I) is a limited function acts on input signal of the
valve. Saturating I at [+ or -][I.sub.s] as the opening fraction of the
valve is, of course, physically limited. In the present study input
signals as shown in Fig. 4 are under saturation limit.
4.2. System parameters
Oil pressure in the cylinder
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Therefore accumulator volume and initial volume of nitrogen:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Finally, system parameters are shown in Table 1.
4.3. Initial condition
The above set of equations is amenable to numerical methods after
defining a proper set of initial conditions. The vector form of the
initial conditions is presented in Eq. 9
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)
5. Results
Figs. 6, 7 and 8 are, respectively, the piston position
[x.sub.rel], the oil pressure [P.sub.z], and the nitrogen pressure
[P.sub.s] in the system versus time.
It is observed in these figures that during the first time
interval, pressure (in both cylinder and capacitor) builds up to a
certain level. Oil is forced into the system, but at the beginning (as
the piston is not moving) most of the oil has to go into the capacitor.
The increase of [P.sub.z] and [P.sub.s] starts to create a force which
is greater than that generated by the external force. Then the plunger
is accelerated outwards. When it starts moving, the available volume in
the cylinder increases, resulting in oil flow back from the capacitor,
and the subsequent decrease in pressure. After the plunger (and the
considered part of the car mass) starts moving, it comes to a stop due
to stoppage of the oil flow. In the second time interval, associated to
the second pulse, the opposite of the above description is relevant
(i.e. oil is taken from the capacitor resulting in a lack of pressure,
which in turn results in deceleration, as the external force of the car
is bigger than that generated by the piston).
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
For the sinusoidal part above descriptions are true but there is
not too much to discuss for the sinusoidal part, because oil is entered
into or is left from the system frequently in low periods.
[FIGURE 8 OMITTED]
Moreover, by solving the state equations, the piston speed and the
oil volume inside the cylinder may be calculated as shown in Figs. 9 and
10. As mentioned before, the downward movement of the piston increases
the cylinder volume, affecting both oil volume and reservoir pressure to
decrease, as is seen in these figures. For upward movement of the
piston, the opposite of the above description is relevant again.
6. Conclusions
An introduction is made to the main principles of active
hydro-pneumatic suspension system and how the combination of fluid, gas
and several accumulators in a suspension can provide both a good level
of comfort and good handling. Bond graphs have been introduced to model
physical systems in a domain independent way. Domain independency stems
from the fact that physical concepts are analogous for the different
physical domains. As a typical hydro-pneumatic suspension system
includes various phenomena in the field of hydraulic, thermo-fluid,
mechanical and control, the BG approach may render itself a convenient
choice for analyzing such systems. In this study a submodel for one
wheel is created. The model uses adiabatic ideal gas compression
assumption for spring force and dynamic friction submodel for the
friction force. The simulation results have been in accordance with
others results. Overall, the present results show that the bond graph is
indeed a suitable method for such applications.
Received December 14, 2010
Accepted May 27, 2011
References
[1.] Heny, C.; Simanca, D.; Delgado, M. 2000. Pseudobond graph
model and simulation of a continuous stirred tank reactor, J. Franklin
Inst. 337: 21-42.
[2.] Lucinskis, R.; Mazeika, D.; Hemsel, T.; Bansevicius, R. 2010.
The experimental research of piezoelectric actuator with two vectors of
polarization direction, Mechanika 2(82): 50-57.
[3.] Rugaitis, A.; Juocas, K.; Volkovas, V.; Deksnys, V. 2010.
Modeling, experimental research and critical parameter analysis of
glider's dynamic characteristics, Mechanika 6(86): 38-42.
[4.] Paynter, H. M. 1961. Analysis and Design of Engineering
Systems. MIT Press: 20-90.
[5.] Karnopp, D.C.; Margolis, R.L.; Rosenberg, R.C. 2006. System
Dynamics. John Wiley 30-210.
[6.] Thoma, J.; Ould Bouamama, B. 2000. Modeling and Simulation in
Thermal and Chemical Engineering. Springer Press: 50-170.
[7.] Ould Bouamama, B. 2003. Bond graph approach as analysis tool
in thermo-fluid model library conception, J. Franklin Inst. 340: 1-23.
[8.] Mukherjee, A.; Karmakar, R. 2000. Modeling and Simulation of
Engineering Systems Through Bond Graphs. Alpha Science International Ltd
Press: 30-250.
[9.] Breedveld, P.C. 2003. Modeling and simulation of dynamic
systems using bond graphs, Control systems Robotics and Automation from
Encyclopedia of Life Support Systems (EOLSS), Developed under the
Auspices of the UNECO, Eolss Publishers, Oxford, UK:200-230.
[10.] Broenink, J.F. 1999. Introduction to physical systems
modeling with bond graphs, Technical report, University of Twente,
Enschede, Netherlands, 1999. Available online at: Proceeding of
International Modelica conference. Available online at:
http://www.ce.utwente.nl/bnk/papers/BondGraphsV2.pdf.
[11.] Borutzky, W. 2007. Bond graph modeling and simulation of
multidisciplinary systems--An introduction, Simulation modeling practice
theory 17: 3-21.
[12.] WWW.ece.arizona.edu/cellier.
[13.] Giliomee, C. L. 2005. Historical and literature overview.
Thesis, Chapter2, University of Pretoria etd Giliomee, CL.
[14.] Westfall, C. 2006. The Citroen Hydraulic System, accessed
26th.http://www.actwin.com/toaph/citroen/work/work.html.
[15.] Kroneld, P.; Liedes, Nevala, T. K.; Marjanen, Y. 2006.
Modeling a selective hydro-pneumatic suspension element. The Thirteenth
International Congress on Sound and Vibration, Vienna, Austria.
[16.] Knorn, F. 2006. Modeling and control of active hydropneumatic
suspension. DAIMLER CHRYSLER Research & Technology and
Otto-von-Guericke-Universitat Magdeburg.
[17.] Dasgupta, K.; Watton, J.; Pan, S. 2006. Open-loop dynamic
performance of a servo-valve controlled motor transmission system with
pump loading using steady-state characteristics, Mechanism and Machine
Theory 41: 262-282.
M. D. Emami, Department of Mechanical Engineering, Isfahan
University of Technology, Iran, E-mail: mohsen@cc.iut.ac.ir
S. A. Mostafavi, Department of Mechanical Engineering, Faculty of
Engineering, Arak University, Arak, 38156- 88349, Iran, E-mail:
a-mostafavi@araku.ac.ir
P. Asadollahzadeh, Department of Mechanical Engineering, Iran
University of Science and Technology, Iran, E-mail:
p_asadollahzadeh@mecheng.iust.ac.ir
Table 1
System parameters
M = [I.sub.4] = 365 kg
[R.sub.d] = 2.08 x [10.sup.9] Pa/([m.sup.3]/s)
[P.sub.res] = [10.sup.5] Pa
[F.sub.c2] = 150 N
[V.sub.Z0] = 8.16 x [10.sup.-5] [m.sup.3]
K = 1.36
[Q.sub.1] = 0 [m.sup.3]/s
Va = 1.13 x [10.sup.-4] [m.sup.3]
[d.sub.v2] = 20 N/(m/s)
[V.sub.M] = 1.3057 x [10.sup.-4] [m.sup.3]
E = 2 x [10.sup.8] Pa
Kv = 5.9 x [10.sup.-7] [m.sup.3]/s/[Pa.sup.1/2]A
Pa = 42.6 x [10.sup.5] Pa
[F.sub.m2] = 100 N
[C.sub.10] = 4.08 x [10.sup.-13] [m.sup.3]/Pa
[A.sub.z] = 10.2 x [10.sup.-4] [m.sup.2]
Psys = 180 x [10.sup.5] Pa
[K.sub.1] = 1937 s/m
[K.sub.2] = -50 s/m
[F.sub.ext] = 7158.4 N