Journal of Fuzzy Systems and Control, Vol.

1, No 1, 2023
ISSN: 2986-6537, DOI: 10.59247/jfsc.v1i1.28 25

PID Control of Magnetic Levitation (Maglev)

System
Muhammad Arif Seto 1, Alfian Ma’arif 2*
1, 2
Department of Electrical Engineering, Universitas Ahmad Dahlan, Yogyakarta, Indonesia
Email: 2 [email protected]
*Corresponding Author

Abstract— Magnetic levitation is a process of drifting or electromagnetic suspension and electrodynamic suspension
shifting an object with a certain reference, by creating a are used in the train rate process as well as in the train braking
repulsive force from the gravity of an object caused by an process [6].
electromagnetic field without noise. In modern times like today, The use of magnetic levitation is as a reduction in the
with the density of community activities, especially those living
in urban areas, it is necessary to be able to carry out activities
friction force of components that are mechanically operating
efficiently as well as quickly. Therefore, the reason for many which causes noise to become noisy [7][8]. So that it can
people to be able to find an innovative solution to make it easier further optimize the work of the system itself because the
for humans to carry out their activities. One of the innovations friction force generated has been reduced. Some examples
that can be a solution to the limited space and time is in the are, the presence of magnetic bearings on the turbine shaft of
development of magnetic levitation technology. The system that a power plant, and magnetic suspension.
will be used in controlling this maglev is to use a PID controller. This research discusses a magnetic levitation system,
Using this PID control will be tested with several 𝑲𝒑, 𝑲𝒊, and which will require many experiments. Maglev control will be
𝑲𝒅 values with a setpoint of 950. At the initial tunning with the decisive in conducting research related to this system [9][10].
values 𝑲𝒑 =2, 𝑲𝒊 =0.01, and 𝑲𝒅 =2. Then in the second test using
In this system, a position control system of an elevated object
the values 𝑲𝒑 =3, 𝑲𝒊 =0.1, and 𝑲𝒅 =0.3. Software testing is also
carried out using matlab. The exact PID value for controlling an will be designed at a certain height under an electromagnetic
elevated object can be found by changing the value of the 𝑲𝒑, field generated from a voltage-fed coil [11]. The problem in
𝑲𝒊, and 𝑲𝒅 values that can be searched through a matlab the maglev system is controlling the object so that it can float
software in the PID tunner feature, which can then be seen the in a certain position without hitting a clumber or falling. This
PID tunning results by analyzing the graph that will be problem can be solved by using the PID controller.
displayed by the PID tunner.
II. METHOD
Keywords— Sensor; Controller; Actuators; Robot.
A. Magnetic Levitation System Modeling
I. INTRODUCTION The model of the magnetic levitation system in the form
In this modern era, in the midst of dense community of a mobile function can be written as [12]
activities, especially in urban areas, it requires people to
move and do activities quickly and effectively. The demands 𝑥̇ = 𝐀𝑥 + 𝐁𝑢 (1)
of work, family responsibilities, and other human activities
make people seem to be hunted by time. Time and space seem B. PID Controller
to be a barrier in carrying out the density of activities in this
This study designed a system used to control the handling
super-fast era. This is the reason for many people to find new
of an object controlled by a system called magnetic levitation.
technological innovations that aim to make it easier for
This system uses infrared sensors and photodiode sensors
humans to carry out their activities. Tools in the industrial
which have a function as a detector for an object [13][14]. If
field and also transformation have also used magnetic
an object is detected by a sensor at a certain height, the sensor
levitation as a solution to make it easier for a system to
will send analog data to be processed whether at that height
operate. The definition of magnetic levitation itself is the
it meets the set setpoint and the process can be shown on Fig.
process of lifting or shifting an object at a certain reference,
1.
without any noise, by creating a gravitational repulsive force
of the object by using a magnetic field as an antidote i r an

[1][2][3].
Magnetic levitation systems are modern technologies that n r ii n

can float an object by means of electromagnetic force [4].

This system has many applications such as maglev trains that
can reach speeds of 600 km/h. The research was conducted
by examining the mechanism and working of levitation ii n n r
magnets on superfast trains that are being developed in
several countries in the world, especially in Japan. This study
explains how a levitation magnetic system can be applied to Fig. 1. The proposed method
trains to accelerate the pace of these trains [5]. Also how

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 Journal of Fuzzy Systems and Control, Vol. 1, No 1, 2023 26

The PID controller equation can be written as [15] TABLE I. FULL TEST RESULTS
PID value System Response
𝑡
𝑑 No KP KI KD Rise Settling Over
Steady
𝑢(𝑡) = 𝐾𝑝 𝑒(𝑡) + 𝐾𝑖 ∫ 𝑒(𝑡). 𝑑𝑡 + 𝐾𝑑 𝑒(𝑡) (2) Value Value Value Time Time shoot
State
0 𝑑𝑡 Error
0.158 2.761 8.8017e-
1 30 50 1 0.5631
2 2 06
Where: 0.122 1.658 1.3443e-
2 40 50 1 0.1850
𝑢(𝑡) : Control signal 4 6 05
𝐾𝑃 : Proporsional constant 0.159 11.39 1.5518e-
3 25 75 1 0.9325
0 75 10
𝐾𝑖 : Integral constant 0.144 17.11 3.8392e-
4 25 100 1 0.7997
𝐾𝑑 : Derivatif constant 3 53 13
0.182 4.868 2.3223e-
𝑒(𝑡) : Error signal 5 25 50 1
2
0.9450
5 06

B. Hardware Testing
C. System Block Diagram
In the implementation of this study, a tool of the magnetic
The system block diagram of this design is arranged from levitation system has been designed which can be seen in Fig
several parts. This part consists of input, processing, and also 4.
output parts. In this input section, it uses a position sensor in
the form of an infrared sensor and also a photodiode sensor
[16]. In the process section, arduino uno and also motor
drivers are used in charge of processing voltage and also the
control program created [17]. In the output section used is an
inductor/coil that functions as an electromagnetic field
generator to float a kite object used [18]. The block diagram
of this system can be seen on Fig. 2.

i
Ar in n n r ri r n r
n r

Fig. 2. System Block Diagram Fig. 4. Maglev Tool Design

III. RESULTS AND DISCUSSION In testing this sensor, several test values of 5 photodiode
A. Matlab Simulation Testing sensors and an IR sensor will be carried out. The photodiode
sensor will be divided into sensor 1, sensor 2, sensor 3, sensor
In the software testing process of this maglev system, one 4 and sensor 5. Sensor Test results are shown in Table II.
formula will be used as a model of the transfer function [19].
Here is the model formula of the transfer function PID of the TABLE II. SENSOR TEST RESULTS
maglev system:
Name Value Information
Sensor 1 660 Detect
Sensor 2 683 Detect
9.359 Sensor 3 772 Detect
(3)
𝑠 2 + 21.36𝑠 + 39.21 Sensor 4 773 Detect
Sensor 5 736 Detect
All sensor 1022 Detect
Equation (3) model is used as a transfer function model to All sensor 632 Detect
look up values from KP, KI, and KD [20]. From the results
that appear will produce a graph. The resulting graph will be In this design the sensor is placed on both sides. The right
analyzed to see the value of rise time, settling time, over side is an infrared sensor that has a total of 1 piece. Then on
shoot, and also steady state error that appears. Matlab the opposite side there are 5 photodiode sensors facing each
simulation test results are shown on Fig. 3. The full test other with infrared sensors.
results are shown in Table 1. In this inductor, a wire winding with a magnitude of 0.7
mm is carried out which is twisted on a cylindrical bobbin
Reference Time
with a length of 7.5 cm and has a diameter of 6 cm. The
manufacture of this inductor is carried out by winding the
Position
wire up to 2510 coil.

The angular velocity µ0 𝑁 2 𝐴

𝐿= (1)
𝑙

Fig. 3. System Block Diagram Note: N = 2510 Coil

l = 7.5 cm = 7.5 × 10-2
A = 6 cm2 = 6 × 10-4
µ0 = 4π × 10-7

Muhammad Arif Seto, PID Control of Magnetic Levitation (Maglev) System

 Journal of Fuzzy Systems and Control, Vol. 1, No 1, 2023 27

Question: L ….? [6] Q. Ge et al., "3-D Electromagnetic Force Characteristics and

Modeling of Double-Sided Air-Cored Superconducting Linear
Answer:
Synchronous Motor for EDS Train," in IEEE Transactions on
µ0 𝑁 2 𝐴 (4𝜋 × 10−7 )(2510)2 (6 × 10−4 ) Transportation Electrification, vol. 8, no. 3, pp. 3411-3426, 2022,
𝐿= =
𝑙 7,5 × 10−2 https://doi.org/10.1109/TTE.2022.3157027.
(4𝜋)(252004)(6)(10−8 ) [7] J. Ding, et al., “An active geophone with an adjustable
= electromagnetic negative stiffness for low-frequency vibration
3 measurement,” Mechanical Systems and Signal Processing, vol.
= 8𝜋 (252004)(10−8 ) 178, p. 109207, 2022,
= 2016032 𝜋 × 10−8 https://doi.org/10.1016/j.ymssp.2022.109207.
[8] R. Krüger, B. Bergmann, and B. Denkena, “Electromagnetic
= 6330340.48 × 10−8 ultra-precision linear guide,” Ultra-precision High Performance
= 0.0633 𝐻 Cutting: Report of DFG Research Unit FOR 1845, pp. 75-106,
= 63 𝑚𝐻 2022, https://doi.org/10.1007/978-3-030-83765-5_4.
At this stage of testing motor drivers, a voltage value test [9] Y. Feng, C. Zhao, D. Wu, H. Xie, and L. Tong, “Effect of
levitation gap feedback time delay on the EMS maglev vehicle
is carried out that arises from the predetermined PWM values, system dynamic response,” Nonlinear Dynamics, pp. 1-20, 2023,
namely 50, 100, 150, 200, and 250. This test is carried out https://doi.org/10.1007/s11071-022-08225-5.
using a program on the Arduino IDE software that will be [10] F. Guo, F. Hu, S. Wu, F. He, J. Liu, and X. Wu, “System dynamics
run. Then after that the maglev system will be measured using in structural strength and vibration fatigue life assessment of the
wing ar f r ig ‐ p magl rain,” International Journal of
a multimeter. Motor Driver and PWM Test Results are shown Mechanical System Dynamics, vol. 2, no. 2, pp. 178-189, 2022,
in Table III. https://doi.org/10.1002/msd2.12045.
[11] J. Paredes, B. Prieto, M. Satrústegui, I. Elósegui and P. González,
TABLE III. MOTOR DRIVER AND PWM TEST RESULT "Improving the Performance of a 1-MW Induction Machine by
Optimally Shifting From a Three-Phase to a Six-Phase Machine
PWM Value Voltage Value (V) Design by Rearranging the Coil Connections," in IEEE
0 0V Transactions on Industrial Electronics, vol. 68, no. 2, pp. 1035-
50 4.77 1045, Feb. 2021, https://doi.org/10.1109/TIE.2020.2969099.
100 9.59 V [12] M. Urbanczyk, A. Zbinden, S. L. Layland, G. Duffy, and K.
Schenke-Layland, “Controlled heterotypic pseudo-islet assembly
150 14.26 V
f man β-cells and human umbilical vein endothelial cells using
200 18.92 V magnetic levitation,” Tissue Engineering Part A, vol. 26, no. 7-8,
255 23.85 V pp. 387-399, 2020, https://doi.org/10.1089/ten.tea.2019.0158.
[13] wan , M. . Ma nawan, N. Ma arani Ra ar a an A. Ma’arif,
IV. CONCLUSION "Infusion Liquid Level Detection Tool Using IR Sensors and
Photodiode Based on Microcontroller," 2020 2nd International
This study contains the control of PID in the magnetic
Conference on Industrial Electrical and Electronics (ICIEE), pp.
levitation system. Based on the results of system simulation 70-73, 2020, https://doi.org/10.1109/ICIEE49813.2020.9277363.
testing, the best PID control values were 𝐾𝑝=40, 𝐾𝑖=50, [14] X. Tang, M. M. Ackerman, M. Chen, and P. Guyot-Sionnest,
𝐾𝑑=1 with system response. In the hardware test, the results “Dual-band infrared imaging using stacked colloidal quantum dot
photodiodes,” Nature photonics, vol. 13, no. 4, pp. 277-282,
of the analysis were obtained with rise time = 0.1224, settling 2019, https://doi.org/10.1038/s41566-019-0362-1.
time = 0.1850, overshoot = 1.6586, and steady state error = [15] J. Zhang and L. Guo, "Theory and Design of PID Controller for
1.3443e-05. Nonlinear Uncertain Systems," in IEEE Control Systems Letters,
vol. 3, no. 3, pp. 643-648, 2019,
REFERENCES https://doi.org/10.1109/LCSYS.2019.2915306 .
[16] D. Laurijssen, W. Saeys, S. Truijen, W. Daems and J. Steckel,
[1] G. Pittiglio et al., "Magnetic Levitation for Soft-Tethered Capsule "Synchronous Wireless Body Sensor Network Enabling Human
Colonoscopy Actuated with a Single Permanent Magnet: A Body Pose Estimation," in IEEE Access, vol. 7, pp. 49341-49351,
Dynamic Control Approach," in IEEE Robotics and Automation 2019, https://doi.org/10.1109/ACCESS.2019.2910636.
Letters, vol. 4, no. 2, pp. 1224-1231, 2019, [17] K. Alsammarraie and T. Inan, "Car Control by using brain waves
https://doi.org/10.1109/LRA.2019.2894907. and Arduino based Mind wave Mobile," 2022 International
[2] P. Carneiro, M. P. S. dos Santos, A. Rodrigues, J. A. Ferreira, J. Congress on Human-Computer Interaction, Optimization and
A. Simões, A. T. Marques, and A. L. KholKin, “Electromagnetic Robotic Applications (HORA), pp. 1-6, 2022,
energy harvesting using magnetic levitation architectures: A https://doi.org/10.1109/HORA55278.2022.9799911.
review,” Applied Energy, vol. 260, p. 114191, 2020, [18] S. Watson, et al., “Future emerging technologies in the wind
https://doi.org/10.1016/j.apenergy.2019.114191. power sector: A European perspective,” Renewable and
[3] J. Wang and L. Yu, "Adaptive Resonant-EIDO-Based Optimized sustainable energy reviews, vol. 113, p. 109270, 2019,
Position Precision Control for Magnetic Levitation System," in https://doi.org/10.1016/j.rser.2019.109270.
IEEE Transactions on Industrial Electronics, vol. 70, no. 5, pp. [19] M. ElAnsary, et al., “Bidirectional peripheral nerve interface with
5013-5023, 2023, https://doi.org/10.1109/TIE.2022.3186348. 64 second-order opamp-l ΔΣ A C an f lly in gra
[4] F. Lamablawa, and S. Aritonang, “Application of Magnetic Force wireless power/data transmission,” IEEE Journal of Solid-State
to Means of Transportation Magnetic Levitation High-speed Circuits, vol. 56, no. 11, pp. 3247-3262, 2021,
Railway,” AJIRSS: Asian Journal of Innovative Research in https://doi.org/10.1109/JSSC.2021.3113354.
Social Science, vol. 1, no. 2, pp. 68-73, 2022, [20] C. Chen, J. Xu, W. Ji, R. Chao, and Y. Sun, “Adaptive levitation
https://doi.org/10.53866/ajirss.v1i2.94. control for characteristic model of low speed maglev vehicle,”
[5] Z. Qadir, A. Munir, T. Ashfaq, H. S. Munawar, M. A. Khan, and Proceedings of the Institution of Mechanical Engineers, Part C:
K. Le, “A prototype of an energy-efficient MAGLEV train: A step Journal of Mechanical Engineering Science, vol. 234, no. 7, pp.
towards cleaner train transport,” Cleaner Engineering and 1456-1467, 2020, https://doi.org/10.1177/0954406219899677.
Technology, vol. 4, p. 100217, 2021,
https://doi.org/10.1016/j.clet.2021.100217.

Muhammad Arif Seto, PID Control of Magnetic Levitation (Maglev) System