Getmanskiy 2016

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

2016 2nd International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)

Coupled Model of Hydraulic Circuit System and


Multibody Dynamics System of Vehicle
Getmanskiy V.V., Nasonov A.A., Novokshcenov A.A.
High Mathematics Department
Volgograd State Technical University
Volgograd, Russia
vm@vstu.ru

Abstract—The article describes development of complex automobile model and the solver usually lacks in performance
model of vehicle with controllable hydraulic suspension bracket. and efficiency. The most efficient way to create and simulate
This model includes a hydraulic circuit system model developed complex models with control algorithm is using high-
using the MATLAB software, a truck multibody dynamics performance programming language such as C++ for control
system model developed with the FRUND software and a control program, CAE software for model development and universal
unit and control panel modules intended for user interaction with simulation software for specific features.
simulation process by sending commands to execute of control
algorithms of a suspension bracket. Such a complex model allows One of real life cases of developing such technology for
simulating a control system and controllable object. The result of simulation of controllable hydraulic actuators in complex
the work is the software tool, which allows controlling the automobile suspension system is considered in present work.
suspension bracket almost in a real-time mode. The technology The aim of the research is development of complex
developed is used for testing control algorithms and programs as mathematical model of automobile with 8x8 wheel formula
elements of a suspension bracket control unit, and over the long with electro-hydraulics control subsystem for suspension. The
term other controllable systems of the vehicle. model includes:
Keywords—hydraulic circuit system; MATLAB; Simulink; • multibody system dynamics model of automobile;
automobile suspen-sion system; hydro-pneumatic spring;
mathematical model; multibody system dynamics; coupling • model of hydraulic-circuit system;
• model of digital control modules;
I. INTRODUCTION
• control algorithms.
The modern workflow of automotive suspension system
design with controllable characteristics includes complex
control algorithm. Creating, debugging and testing such II. COMPLEX MATHEMATICAL MODEL
algorithms on real hardware is a time consuming work of The complex mathematical model of automobile is
heavy costs. The development of technology that allow using developed using three simulation and programming tools:
mathematical models of real mechanical system to develop
control algorithm and simulate the actuators is an important • FRUND – multibody dynamics simulation package;
problem in modern industry. • MATLAB Simulink – universal simulation tool used
Specific feature of such technology is a requirement of for hydraulic-circuit system;
using different simulation tools and software codes. The most • Microsoft Visual Studio – software development tool
difficult and nontrivial part is a coupling of these tools. There for creating the control algorithms.
are a lot of multiphysical simulation [1,11] computer-aided
engineering software (CAE) and universal simulation software The complex model consists of car model, hydraulic model,
such as MATLAB Simulink [2]. State of the art approach for control unit and control panel. Blocks interactions are
vehicle simulation in CAE is multibody system dynamics presented at figure 1.
analyses [1,3-6]. As a matter of fact the simulation software of There are similar approaches in earlier researches of
high quality is targeted for solving single specific problem such Gorobtsov A.S. [12,13], Solodenkov S.V. [14], , Emami, M. D.
as multibody dynamics, stress-strain analysis, durability, [15], Guo K.[16], Wang W.[17], Feng J.[18] which also uses
hydrodynamics etc. Therefore, the CAE system is build based multibody simulation with special models for hydraulic system
on a heavily featured solver and other additional solvers have a circuit. In some of them MATLAB is also used but there is no
limited number of features. One of such software FRUND [7- analyses based on complex 3-dimensional model of vehicle
10] based on multibody dynamics is developing with coupled with hydraulics system circuit submodel.
participation of authors. On the other hand the universal
simulation software required a lot of efforts to develop full

The work was carried out with the financial support of RFBR (project #
16-47-340385).

978-1-5090-1322-7/16/$31.00 ©2016 IEEE


2016 2nd International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)

Fig. 1. General structure of complex model.

III. HYDRAULIC-CIRCUIT SYSTEM MODEL


Hydraulic-circuit system model is developed in MATLAB
R2014a Simulink. It consists of four similar blocks of
hydropneumatic springs [12-19], block of clearance control and Fig. 2. Single-wheel hydraulic-circuit model.
block of cylinder heads control. Each block of hydropneumatic
spring consist of elements for two wheels of adjacent axles.
Hydraulic-circuit system is controlled by valve states. Valve
states are changed by signals from control unit.
For developing of interaction interface between MATLAB
model and FRUND model the single-wheel simplified model is
developed. Hydraulic-circuit system model for it is shown at
figure 2.
This model can be used independently in case of correct
states of valves. It simulates normal hydropneumatic spring
and damper characteristics. For independent work this model is
complemented by additional mechanical spring and damper.
Additional spring gets the force from hydro cylinder in case of
zero force from automobile suspension.
All submodels, which include hydropneumatic springs,
Fig. 3. Axonometric view of automobile model.
have spring force as output parameter and force from
suspension as input parameter. Force from suspension is
Each 3-dimensional body has 6 degrees of freedom. Model
measured in special joint in multibody model of automobile.
consist of 52 bodies. All bodies are weighted (have inertia).
The states of control elements are described as inputs Impact from road profile is simulated by tire model, which has
S_R.1.VP, S_R.1.1.VP,S_P.1.2.VP,C_C1.P,A_C1.P. The three-component force in contact point and self-aligning
control signal is discrete and represents open states of valves in torque.
case of value 1 and close state in case of value 0.
Suspension bracket consist of upper and lower arms, slave
arm, hydropneumatic spring and wheel hub.
IV. AUTOMOBILE MODEL
Three-dimensional model of four-axle truck is developed in V. CONTROL UNIT
FRUND software. General design scheme is presented at figure
The submodel of one suspension bracket is included in
3. Dynamics model consist of rigid bodies connected by
multibody system model of the car by the way of linking
spring-damper links and kinematics rigid and flexible joints a
additional C-language subroutine to multibody solver
multibody system.
generated by FRUND software. Simulink hydropneumatic
circuit system model is exported as C-code and linked to solver
as well. Input parameters for control subroutine are the
variables received from multibody model and Simulink model.
2016 2nd International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)
And output parameters are the values of control parameters of Realization of the coupling interface includes three main
actuators in MATLAB model. functions:
This subroutine allows to change conditions of the • void ModelToControl (double t, double * inParams);
controllable elements (for example, operating the hydraulic
valves), but doesn't allow to vary internal parameters of • void RunControlAlgorithm (double t, double *
components of the hydraulic circuit (hydraulic resistance, controlParams);
charging pressure, etc.) therefore at each change of internal • void ControlToModel (double t, double * outParams);
parameters it is necessary to export MATLAB model sources
and rebuild the solver. The “ModelToControl” function fills internal variables of
the control unit with the necessary entrance data taken from
For the full vehicle model the description of parameters of model variables.
the subroutine is very large so the principle of coupling model
of a hydraulic system in MATLAB and model of the vehicle in The “RunControlAlgorithm” function carries out the
a FRUND is given by an example of the simplified hydraulic control algorithm for a current moment of time “t”. As an input
circuit system. parameter is array of control parameters, which allow choosing
particular control algorithm, depending on the transferred,
The simplified hydraulic circuit system is used for control carried-out action.
of wheel raising. The model is exported in the form of source
codes and build as a static C-language library. The library with The “ControlToModel” function transfers the output data
dependent components are linked together with source codes of from the control unit to model, modifying necessary variables.
multibody model. As the external interface, which is switched on in a code of
The force element for modeling of influence from a the settlement module, function is used the function:
hydraulic system for raising of a wheel is added to vehicle void Control (double t, double * inParams, double *
model in FRUND. Force value is simulated in the MATLAB outParams, double * controlParams);
model and on each iteration moves to a force element in the
FRUND model. The input parameters for MATLAB model are The subroutine parameters are the current time, input
force influence from the hinge of fastening of a hydraulic parameters, output parameters and a set of the operating
cylinder and states of the valves. parameters. Subroutine consist of consecutive calls of three
functions described above.
Any control algorithm for hydraulic system is the
subroutine, which changes states of the valves depending on The described scheme of interaction is applied to
the satisfied conditions. The states of the valves are modelled connection to model of dynamics of the car of auxiliary model
by values 0 (the valve is closed) and 1 (the valve is open). of the hydraulic scheme of suspension brackets of the car
General structures for all valve states is developed and used in created in MATLAB Simulink. The model of vehicle is
all subroutines. simulated faster than Simulink model. FRUND solver is
developed taking into account high performance computing
The FRUND software allows to generate the source code and the software is highly optimized [7-9, 20]. The
for compiling the solver. For integration with the Simulink consideration of time synchronization is in synchronization the
model exported in the form of C-language source code the time steps and waiting for slower MATALAB model at each
control unit file allows to insert in process of generating the step it is called. Also for MATLAB model the greater time step
included code fragments in the form of additional files by is used to reduce the number of calls for slow solver.
means of the directive “Include”.
For coordination of time steps (integration steps) of the
The connecting element with the elastic damping main and auxiliary models linear interpolation is used. The
characteristic is used in multibody model for transferring the integration step for auxiliary model is greater than for
force value from the Simulink model. Parameters of damping multibody model therefore value of a signal on each time
and stiffness of a connecting element are specified in Simulink interval equal to this step for the multibody model is
model, therefore connecting element is distinguished in the interpolated by linear function (fig. 4).
special parameter representing consecutive number of an
independent suspension bracket of the vehicle model. By this way the technology developed allow to simulate the
According to the number of suspension bracket, the address in coupled model of vehicle and hydraulic circuit scheme.
the buffer used to transfer the force value is calculated.
VI. CONTROL PANEL
In the control unit there is a call of the program simulating
the hydraulic circuit system. Values of output signals are A control panel was developed as a user interface of control
interpolated under the linear law, thus new iteration on time is unit. It allows sending commands to control the suspension
calculated only if it is necessary. bracket during simulation. Control panes sows the states of
hydraulic circuit valves. Control panel interface is presented at
The call of C-language control module is inserted in figure 5.
multibody dynamics model code in time before decomposition
of a force vector to components of the Cartesian coordinate Upper field contains the states indicators of 62 valves in
system, thus there is a replacement of the standard calculated hydraulic circuit. Colored rectangles correspond to open valves
force value by simulated one obtained from Simulink model. and white rectangles are for closed.
2016 2nd International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)
Interpolated Next computed value

60
Force (N) (10^3)

50

40

Fig. 6. Initial front view (wheel is lowered – a), final front view (wheel is
raised – b).
30
Still some conclusions:
1,000 1,002 1,004 1,006 1,008 1,010
Time (s • Speed of calculation of model of the car in the FRUND
program complex ensures functioning in real time
Fig. 4. Interpolated signal of calculated force values. together with a control system submodel.
• The simulation speed of full model of a hydraulic
system in the MATLAB is more than 50 times slower
than real time that complicates its use in the interactive
mode.
• The complex model allows modifying operating
programs written in language C ++.
• The developed structures of data for control unit can be
coupled with the hardware interface of executive part of
control system for adjusting the real hardware of a
control system.
The results of this work showed potential efficiency of an
Fig. 5. Control panel interface. interactive program complex for developing control algorithms
as for suspension bracket as also for other controllable systems
Bottom part of the panel contains the buttons to perform of vehicle.
control commands (for example rise the wheel with concrete
number). There are text field for information and reset button Computational experiments revealed sufficient computing
to return to default state. When user presses the button, the efficiency of multibody simulation systems for dynamic
control panel software writes to a file control command code. analysis of vehicles. Regarding programs for hydraulic circuit
This file is an input file for control unit. Depend on the control systems simulation lacks in performance. So more powerful
command the control unit runs the algorithm and modify the computing hardware is required to perform simulation in real
states of the valves. The states are written to a file, which the time. This specific circumstance shows the necessity of
control panel software is reading. After updating the states are development new solvers for hydraulic schemes simulation.
shown at the interface.
ACKNOWLEDGMENT
VII. THE SIMULATION RESULTS The work was carried out with the financial support of
General view of coupled model simulation is shown at RFBR (project # 16-47-340385).
figure 6.
REFERENCES
VIII. CONCLUSION [1] J.C. Samin, O. Brüls, J.F. Collard, L. Sass, P. Fisette, “Multiphysics
modeling and optimization of mechatronic multibody systems,”
The developed technology allows performing simulation in Multibody System Dynamics, vol. 18, no. 3, pp. 345, 2007.
close to real-time mode. Coupling hydraulics circuit with [2] MATLAB Documentation. [Online]. Avalaible:
multibody model of full vehicle and adding control unit allow http://www.mathworks.com/help/MATLAB/.
performing main control operations of vehicle suspension [3] T.D. Gillespie, Fundamentals of vehicle dynamics, Warrendale, PA: SAE
system. International, 1996.
[4] A.S. Gorobtsov, S.K. Kartsov, A.Ye. Pletnev, Yu.A. Polyakov, Computer
The created complex model allows testing the control methods of construction and study of mathematical models of the
algorithms in the interactive mode with visualization of the dynamics of vehicle designs: a monograph, Moscow: Mashinostroyeniye,
movement of the car. In addition, the model allows using 2011.
control panel of a suspension bracket.
2016 2nd International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)
[5] V.Ye. Tolskiy, A.S. Gorobtsov, S.M. Voyevodenko, “Modern car design [13] A.V. Andreychikov, O.N. Andreychikova, A.S. Gorobtsov, “Computer-
methods. Problems and solutions,” Automotive industry, no. 10, pp. 34- aided design of active suspension vehicles,” Engineering Bulletin, no. 2,
36, 2008. pp. 22-25, 2009.
[6] A.S. Gorobtsov, “Experience in the use of complex models of dynamics [14] S.V. Solodenkov, A.S. Gorobtsov, K.I. Lyutin, “Mathematical model of
of machines and promising fields of application,” in Proc. Problems and hydromechanical system constant speed,” Automotive industry, no. 8, pp.
prospects of precision mechanics and control in mechanical engineering, 21-24, 2008.
Saratov, 2007, pp. 191-200. [15] M.D. Mostafavi, S.A. Asadollahzadeh, “Modeling and simulation of
[7] V.V. Getmanski, A.S. Gorobtsov, E.S. Sergeev, “Concurrent simulation active hydro-pneumatic suspension system through bond graph Emami,”
of multibody systems coupled with stress–strain and heat transfer Mechanika, 3, pp. 312-317, 2011.
solvers,” J. of Computational Science, vol. 3(6), pp. 492-497, 2012. [16] Konghui Guo, Yuhang Chen, Yehai Yang, Ye Zhuang, Yanbo Jia,
[8] A.S. Gorobtsov, V.V. Getmanskiy, A.E. Andreev, “Duy Trung Doan “Modeling and simulation of a hydro-pneumatic spring based on internal
Simulation and Visualization Software for Vehicle Dynamics Analysis characteristics,” in Proc. Mechanic Automation and Control Engineering
Using Multibody System Approach,” in Proc. Creativity in Intelligent (MACE), 2011, pp. 5910-5915.
Technologies and Data Sci-ence. CIT&DS, Volgograd, September 2015, [17] Wenrui Wang, Bing Chen, Xiaoqiang Yan,; Liang Gu, “Design and
pp. 379-391. dynamic simulation on adjustable stiffness hydro-pneumatic suspension
[9] V.V. Getmanskiy, A.S. Gorobtsov, T.D. Ismailov, A.E. Andreev, vibration characters of tracked vehicle,” in Proc. Electric Information and
“Heuristic method of dynamic stress analysis in multibody simulation Control Engineering (ICEICE), pp. 4056-4059, 2011.
using HPC,” in Proc. 11th World Congress on Computational Mechanics [18] Jinzhi Feng, Christian Matthews, Songlin Zheng, Fan Yu, Dawei Gao,
(WCCM 2014). 5th European Conference on Computational Mechanics “Hierarchical Control Strategy for Active Hy-dropneumatic Suspension
(ECCM 2014) and 6th European Conference on Computational Fluid Vehicles Based on Genetic Algorithms,” Advances in Mechanical
Dynamics (ECFD 2014), Barcelona, July 2014, pp. 3115-3124. Engineering, vol. 7, no. 2, February 2015.
[10] A.S. Gorobtsov, S.V. Solodenkov, “Raskhotnye problems of dynamics [19] A.S. Gorobtsov, D.A. Miroshnichenko, “The spatial oscillations of the
systems tvordyh and elastic bodies in the software package Frunda,” sprung mass of the car with a random perturbation,” Automotive industry,
Engineering and Engineering Education, no. 4, pp. 31-38, 2008. no. 7, pp. 20-21, 2010.
[11] Ye.G. Gromov, A.A. Nasonov, A.A. No-vokshchenov, V.O. Firsova, [20] Ye.S. Sergeyev, V.V. Getmanskiy, A.S. Gorobtsov, “Transferring the
“Multifizicheskie problem in dynamics tvordyh systems and elastic system multibody dynamics for computing cluster,” Scientific and
bodies,” in Proc. XI All-Russia with "rides on the fundamental problems technical sheets of St. Petersburg State. Polytechnic. Univ, is. 101, pp. 93-
of theoretical and applied mechanics-term, Kazan, 2015, pp. 1080-1082. 99, 2010.
[12] A.S. Gorobtsov, A.V. Podzorov, “A mathematical model of the
frequency-dependent characteristics of the hydraulic shock absorber,”
Automotive industry, no. 7, pp. 18-20, 2009.

You might also like