Position Control of A Pneumatic Cylinder
Position Control of A Pneumatic Cylinder
Position Control of A Pneumatic Cylinder
Abstract
Pneumatic cylinders are used as actuators for modern automated
systems in many industries applications for their numerous advantages.
Basically, those applications are implemented using components such
as proportional and servo pneumatic control valves that are relatively
expensive because of their high accuracy even in applications that
Keywords: do not require such high accuracy. Recently, binary (on/off) solenoid
Dead-zone compensator, non-linearity, pneumatic directional control valves (DCVs) have been used to replace many of
double acting cylinder, solenoid DCV. the expensive components and consequently reduce the system cost
with achieving acceptable accuracy. This paper presents a control
technique for a positioning system of a pneumatic double acting cylinder
controlled by pair of on/off solenoid DCVs. The control technique has
been developed based on a parametric study presented previously by the
authors[1]. The control technique depends on operating the DCVs using
a pulse width modulation (PWM) algorithm that is modified to employ
a valve dead-zone compensator (DZC) along with using optimum
pneumatic parameters gained from the performed parametric study. An
experimental test bed for a pneumatic double acting cylinder positioning
Corresponding Author:
system that uses a pair of on/off solenoid DCVs has been installed to test
Mohamed Hussien Mabrouk, Military the accuracy of the presented control technique. A comparison between
Technical College, Cairo, Egypt, the performance of the positioning control technique using the modified
Tel: +201011345207. Email: mmabrouk@ PWM control algorithm and the traditional one has been carried out
mtc.edu.eg. to experimentally show the enhancement of the positioning accuracy
with acceptable maximum steady state error less than 0.7 mm when
using the modified PWM control technique due to reducing the system
nonlinearity associated with its dynamics.
Fig. 1: Experimental pneumatic circuit used to test the modified PWM controller 1-Air preparation unit, 2-Air distributing unit, 3, 4- 3/2
Solenoid DCV, 5-Double acting pneumatic cylinder, 6-Pressure transducer no.1, 7-Pressure transducer no.2, 8- Pressure transducer no.3,
9-LVDT, 10-Accumulator, 11-Pressure regulator.
122
Engineering Science and Military Technologies
Volume (4) - Issue (1) - Mar 2020
III. MODELING OF THE PNEUMATIC SYSTEM based PWM control technique is developed by modifying
In order to enhance the pneumatic systems performance, the traditional PWM control algorithm to be used with
a better understanding of the system behaviour through maintaining a constant cylinder outlet pressure by taking
mathematical modeling and simulation is needed. Matlab® advantage of employing a dead-zone compensator (DZC)
Simscape has been used by many researchers as it offers along with using the optimum pneumatic parameters as
advanced and easy solutions to model and simulate recommended by the parametric study performed to the
pneumatic systems[8-10]. system.
Adopting the approach of physical network modeling Pneumatic drives are usually controlled by proportional
technique followed by many researches[11-15], the dynamic or servo valves for high accuracy of position control.
equations for the pneumatic system in this research have In these valves, continuous varying of the valve flow
been modeled and simulated using Matlab® Simscape resistance leads to vary the airflow mass that enters into
toolboxes and discussed in details in[1]. The model has and out from the actuator[15]. On/off solenoid DCV is used
been validated by comparing the predicted dynamic here to control the varied flow discretely. The airflow
behaviour of the system to the experimentally measured enters into or out from the actuator in form of packets.
one. A parametric study has been performed to understand If those packets are delivered faster than the actuator
the dynamic system behavior, determine the major dynamics, the actuator responds negatively to the average
affecting parameters, and configure the optimum operating mass flow rate that enters into or out from it. The desired
PWM frequency suitable for supply pressure and the PWM signal needed to control the solenoid valves can
corresponding effective duty cycle range. be realized by comparing the continuous control signal
The parametric study presented in[1] has shown that and a high-frequency carrier wave[8]. The carrier wave is
throttling the air flow to the cylinder inlet by adding a small usually a high-frequency tooth wave with a period T. The
orifice (with area of 1e-6 m2 and discharge coefficient frequency and amplitude of the carrier wave must vary
of 0.8) decreases the nonlinear effect of the air faster than those of the continuous signal. Following the
compressibility on the piston side pressure by about 80%. approach adopted in[7], the mathematical description of
However, the velocity of the piston is increased by about the used PWM signal here can be given by the following
130% at the recommended working supply pressure, which equations:
ranges from 2.5 to 4 bar, and consequently the pneumatic
system looses the high speed utility. In addition, the
(1)
parametric study has shown that using an accumulator of
1-liter volume and pressure of 1 bar to the rod side chamber
tends to decrease the backpressure nonlinearity for no less
than 50%.
(2)
Using a constant dead weight at rod-side would have
been a simple choice to maintain a constant force at rod-
side, however, the benefit of using the accumulator is to where Upwm is the PWM signal, Up is the maximum
maintain a constant pressure difference between both operating voltage, Vc is the voltage of the carrier wave,
cylinder’s sides at constant load with no sudden pressure Vd is the command signal voltage, j is the jth modulation
change. In addition, using a pressure regulator facilitates period.
investigation of the effect of different accumulator For valves operated by PWM signal, if the PWM
pressures on the positioning accuracy, as recommended in frequency is very high, the switching-on time will be less
previous studies[1]. than the valve delay time, so the valve will not respond and
Studying the valve response to PWM input signal has stay closed, also if the PWM frequency is too small, the
showed that the optimum PWM signal frequency to operate valve will stay off[8].
the valve is 20Hz and the corresponding effective range of This means that the modulating frequency should be
duty cycle is from 10% to 60% at 4 bar supply pressure. selected to ensure the valve opening for certain minimum
duty cycle andthe range of the effective duty cycle defined
IV. SYSTEM CONTROL STRATEGY
to determine the boundary of the valve dead-zone.
Previously in[1], two control methods have been used Also, it can help in avoiding of nonlinearity[9, 10].
to perform piston positioning control for the system in this Figure 3 (left) represents the dead-zone characteristics of a
research. The first method is a simple on/off control method single 3/2 solenoid valve where Uc is the valve input and Uv
(bang-bang control) which has shown an undesirable is the valve output and UcM is the minimum effective valve
effect of the DCV delay time on the system response. input and Sv is the slope of output. While figure 3 (right)
The traditional PWM based control method has been shows the graphical representation of the lookup table
applied to the system to enhance the dynamic behaviour for (DZC). The slope of the compensator output Uvc(es)
of the system, make a stable positioning, and to waive the and Uvc(eMax) are obtained using simulation-based tuning.
undesired behaviour of the bang-bang control method. Equation (3) represents the mathematical description of the
In this paper, the optimum pneumatic parameters dead-zone characteristics:
123
ESMT, Youssry et al., 2020
124
Engineering Science and Military Technologies
Volume (4) - Issue (1) - Mar 2020
Fig. 7: Position error using modified PWM algorithm at Pac=2.5 VI. CONCLUSION
bar.
This paper presents a modified PWM positioning
control algorithm for a double acting pneumatic cylinder
with acceptable accuracy that can be used to aid in replacing
of expensive pneumatic constituents by commercially low-
cost constituents. The modified PWM control algorithm
depends on employing dead-zone compensator along
with using the optimum pneumatic parameters values
gained from parametric study previous performed on the
system[1] to reduce the system nonlinearity associated with
its dynamics. Using the modified PWM control algorithm
enhances the performance of the system, which shows
good positioning accuracy, where the steady state error is
decreased from ±2.5 mm using traditional PWM control
Fig. 8: Actuator response at Ps=4 bar and Pac=4 bar.
algorithm[16] to ±0.28 mm .
The results show that increasing the accumulator
pressure enhances the system performance increasing in
the system damping characteristics, by 70%, the mean
error to as of 27%, and . These results ensure the capability
of the presented control method to improve the pneumatic
positioning systems performance by reducing system's
nonlinearities with less control algorithms complications.
VII. REFERENCES
[1] Youssry,A., Elmayyah, W., Mabrouk, M., and Mahgoub, M.,
"Parametric Study Of A Low Cost Pneumatic System Controlled By
On/Off Solenoid Valves", IJRET: International Journal of Research in
Engineering and Technology, 2016. Vol-05(Iss-03): p. 184-193.
Fig. 9: Position error using modified PWM algorithm [2] Mabrouk, M., and Ahmed Y. Abdel-Hamid, "Control of Polar
at Pac= 4bar. Robot Manipulator Subjected to Harmonic Excitation Using Low Cost
Controller," Lecture Notes in Engineering and Computer Science:
Discussing the effect of the load value and direction Proceedings of The World Congress on Engineering 2019, 3-5 July, 2019,
London, U.K., pp485-489.
on the system, the piston acceleration is decreased with [3] AbdelHamid, A., AbdelKader, H., and Mabrouk M., "Load Tracking
decreasing the pressure difference among both sides And Anti-Sway Control For Telescopic Rotary Crane", Proceedings of
when using a constant load. Consequently, at the same the ASME 2017 International Mechanical Engineering Congress and
Exposition, IMECE2017, November 3-9, 2017, Tampa, Florida, USA
pressure deference the acceleration is decreased further by [4] Chavoshian, M., Taghizadeh, M. and Mazare, M., "Hybrid Dynamic
increasing the load, which results in reducing the error at Neural Network and PID Control of Pneumatic Artificial Muscle Using
the expense of speed. the PSO Algorithm", Int. J. Autom. Comput. 17, 428–438, 2020. https://
doi.org/10.1007/s11633-019-1196-5.
Regarding the load direction, it can be noticed from the [5] Nguyen, T., Leavitt J. ; Jabbari F. ; Bobrow J.E., "Accurate sliding-
results that moving the load against the rod side reduces mode control of pneumatic systems using low-cost solenoid valves",
the overshoot and achieves smooth motion with almost the Mechatronics, IEEE/ASME Transactions on, 2007. 12(2): p. 216-219.
[6] Kunt, C. and Singh, R., "A linear time varying model for on-off
same accuracy. However, this conclusion needs further valve controlled pneumatic actuators", Journal of dynamic systems,
investigation in the future to determine the effect of the measurement, and control, 1990. 112(4): p. 740-747.
value of the pressure deference between cylinder sides. [7] Šitum, Ž., Novaković, B., and Petrić J., "Pneumatic Servodrives
Control Using On/Off Solenoid Valves", 7th Int. Conf. on Prod. Eng.
The system nonlinearity has been investigated CIM 2001.
thoroughly with the conclusion that the nonlinearity appears [8] Sorli, M, Figliolini, G., and Almondo A., "A. Mechatronic model and
125
ESMT, Youssry et al., 2020
experimental validation of a pneumatic servo-solenoid valve", J Dyn Syst [13] Najjari, B., "Modelling and controller design
Meas Control 2010; 132(5): 054503. of electro-pneumatic actuator based on PWM",
[9] Shaalan, A., Badr, M., Al-Ameen, E.,"Motion control of Electro IAES International Journal of Robotics and Automation
pneumatic system on Directional Control Solenoid Valve", Internationa (IJRA), 2012. 1(3): p. 125-136.
Journal of Mechanical and Production Engineering Research and [14] Bone, G., "Position Control of Pneumatic Actuators Using Three-
development (IJMPERD), 2019. Mode Discrete-Valued Model Predictive Control" , Actuators 2019, 8, 56;
[10] Andrighetto, P.L., Valdiero, A., and Bavaresco, D.,"Dead-zone doi:10.3390/act8030056.
compensation in pneumatic servo systems", in ABCM Symposium Series [15] Zhang, X., Lu, Y., Li, Y. , Zhang, C., Wang, R., "Numerical
in Mechatronics. 2008. calculation and experimental study on response characteristics
[11] Salim, S., Rahmat, M., Faudzi, A., Ismail, Z., and Sunar, N., "Position of pneumatic solenoid valves", Measurement and Control 2019,
Control of Pneumatic Actuator Using Self-Regulation Nonlinear PID", Vol. 52(9-10) 1382–1393
Mathematical Problems in Engineering, 2014. 2014. [16] Messina, A., Giannoccaro, N., and Gentile, A., "Experimenting
[12] Kothapalli, G. and Hassan, M.Y., "Design of a neural network based and modelling the dynamics of pneumatic actuators controlled
intelligent PI controller for a pneumatic system", IAENG International by the pulse width modulation (PWM) technique", Mechatronics,
Journal of Computer Science, 2008. 35(2): p. 217-225. 2005. 15(7): p. 859-881.
126