Design of PLC- based PI Controller for the

Permanent Magnet DC Motor under Real Constraints

and Disturbances

Jasmin Velagić, Kerim Obarčanin, Enisa Kapetanović, Senad Huseinbegović, Nedim Osmić
Department for Automatic Control and Electronics
Faculty of Electrical Engineering Sarajevo
Sarajevo, Bosnia and Herzegovina
[email protected]

Abstract— The purpose of this paper is to analyse and implement this paper is to analyze and demonstrate velocity control for the
PI control for the permanent magnet DC motor. The control permanent magnet DC motor with amplifier. The velocity of
algorithm is realised using Siemens S7-200 Programmable Logic DC motor can be varied by controlling the field flux, the
Controller (PLC). The complex motor system is composed of DC armature resistance or the terminal voltage [7]. In our case,
motor, driver and tachogenerator. The main objective is to velocity control would be realised with terminal voltage
achieve a satisfactory time response of the system output under method. The aim is to obtain a fast and stable response
disturbances like death zone, nonlinearity, measurement noise nonsensitive to system's parameters changes and disturbances
and external load acting. The PI controller is designed in the acting. The controller is realised by using PI control algorithm
programming enviroment on a previously identified nonlinear for DC motor under different constraints such as nonlinearities,
motor system. Then the PI controller is embedded into the S7-200 death zone, measurement noise and external load. The complex
PLC. The effectiveness of this controller are tested in both
DC motor system is composed of the DC motor with
simulation mode and experiments.
permanent stator magnet, the linear amplifier, the
I. INTRODUCTION tachogenerator and the axel gear. The effectiveness of the
controller is tested in both simulation mode and experiments.
Being simple, robust, effective and applicable to a broad
class of system, PI and PID controllers have been the most The paper is organised as follows. The structure and
widely used and well known controllers in industry for over 50 the elements of the complex DC motor system and overall
years [1]. In process industries, more than 97% of the control system are described in the Section II. The Section III
regulatory controllers are of the PID or PI type [2]. Easy describes identification procedure of the complex motor
controlling and cheapness of the circuit drive of DC motors system needed for simulation purpose. The design of PI
comparing to AC motors has lead to be chosen by the controller in PLC-based software (Siemens Step7 Micro/Win)
consumers and industries [3]. The permanent magnet DC is presented in Section IV. The essence of this paper is
motor becoming more popular in many control systems comparison of simulation and experimental results, shown in
because of its high power density, large torque to inertia ratio, Section V. The conclusions are written in section VI.
small and high efficiency [4], [5]. In that manner, analysis and
implementation variations of PI DC motor control is of crucial II. SYSTEM DESCRIPTION
importance, especially when we take into account widespread
use of permanent magnet DC motors. This control systems The DC control system based on the Siemens PLC is
exhibit good control performance without influences of shown in Figure 1. The system contains two components, PLC
significant disturbances and variation of controlling parameters based PI controller with analog input/output modul and the
[6]. It is known that moment of inertia of the DC motor always complex motor system. The PI controller is designed in
changed and does not produces properly system response. mentioned software and downloaded to Siemens PLC S7-200
Also, the variation of the load torque can create some ripples in CPU 222.
the rotor speed (or voltage). Therefore the PI controller should
be designed which would not be sensitive to the system
parameters and acting of input and output disturbances as much
as possible.
Also, auto tuning methods for adopting PI parameters are Figure 1. Block diagram of the control system
interesting for engineering because those methods ensure
improvement in industry automation systems. The objective of

Figure 1. Block diagram of the control system

978-1-4244-4221-8/09/$25.00 ©2009 IEEE

This model of PLC CPU has not included analog input/output
pins so it was necessary to use EM235 as analog input/output
modul. Using EM235 analog modul is acquired a response
from DC motor as voltage signal in a range (0V, 10V). Also,
using EM235 is generated control variable for the DC motor
system control.

A. DC motor and linear amplifier

The DC motor is constructed of two main parts, armature

(rotor) and stator. The stator consists of permanent magnets Figure 2. Power amplifier structure
which create magnetic field [11]. The armature contains an
electromagnet created by the coil wound around an iron core.
The armature rotates due to the phenomenon of attracting and
opposing forces of two magnetic fields. A magnetic field is If R1 = R4 and R2 R3 then Uout =Uc .
generated by the armature by sending an electrical current
through the coil and the polarity is constantly changed by Altought, this amplifier represents a linear system which
alternating current through the coil cousing armature to rotate. introduces a segnificant nonlinearity in the whole system due
to own saturation. The output range of the amplifier Uout is
The electrical and mechanical characteristic of the
permanent magnet DC motor are described by equation: limited under 1.5 V , Uin -1.5[V] .
The tachogenerator establishes relation between an output
dIa
Ua =Ra Ia +La +e motor voltage E and rotary velocity of motor shaft as
dt (1) follows :
e=kv ω (2) E=kΦω (7)
dω
τm =Jm +Dω+τt where k is tachogenerator constant and is magnetic flux. It
dt (3)
is important to note that the magnetic flux of the DC motor
with permanent magnets has a constant value.
τm =kt Ia (4)

where Ua is voltage applied on the armature, Ia is the armature B. Siemens PLC s7-200 + EM235 analog I/O modul
current , Ra and La are the armature resistence and inductance,
respectively, is rotational velocity of the armature, τm is the As PI controller it is used Siemens PLC S7-200, CPU 222.
motor driving torque, τt is the mechanical torque load, Jm and It is expendable, compact and very fast Siemens PLC,
D are moment of intertia and dumping coefficient at the motor especially with respect to its real time performance [8].
shaft and Kv is the voltage constant and Kt is torque constant. Mentioned PLC does not have an integrated analog input /
In this paper we used the permanent magnet DC motor whose ouput and for that reason expansion modul for analog input /
parametars are unknown. This motor is considered as a part of output EM235 has been used. The EM235 has 1 analog output
the complex DC motor system ilustrated in Figure 1. ( ±10V, 0-20mA ) and 4 analog inputs with capability to select
an voltage input range ±10V e.g.0-10V, 0-5V, ±10V, ±5V .
The power amplifier represents a linear serial controller The PLC is programmed using Step7 Micro/Win v4.0
shown in Figure 2. The output voltage is calculated as follows: software. More about PLC and analog I/O modul can be found
on Siemens official web site (www. siemens.com).
R4 R2
Uout = 1+ Uc (5) III. IDENTIFICATION OF COMPLEX MOTOR SYSTEM
R1 +R4 R3
The complex DC motor system is considered as black box.
where is controled voltage ( from PLC S7-200/ MR253 Further, parametric identification procedure of this complex
modul ) and represents appropriate resistor. Transfer system is performed using dSPACE system for rapid control
function of the power amplifier is : prototyping (RCP) [11].
The input signal is voltage applied to the power amplifier
and output is voltage on the DC motor shaft which is mesured
R4 R
A= R 1+ R2 (6) by tachogenerator. The choice of input data is essential for the
1 +R4 3
process identification. In this paper the input signal is chosen
to be band-limited white noise. The response of the complex
system on the excitation is shown in Figure 3 and Figure 4. clear trade-off between robustness and aggresiveness of the
Identification process is perfomed using Box-Jenkins method controller [9]. The block diagram of the control system and
achiving matching results of 92.65%. Results of this method process is depicted in Figure 6.
are shown on Figure 5. Due to constraints of the dSPACE
system, input and output values are scaled from [0,10] into
[0,1] and from [0, 2.8] into [0,1], respectively. The transfer
function of the complex motor system is :

2.105
G s = 0.40337s+1 (8)

Figure 6. IMC based controller

In the Figure 6. denotes a model of the process, is an

aproximate inverse of and is and low-pass filter [10]. If
we assume that all disturbances present in the the process can
Figure 3. Input data of complex motor system
be reduced to an equivalent disturbance d at the process output
according to Figure 6,then the controller transfer function is :

-1
Gf Gm
Gc = -1
.
1-Gf Gm Gm (9)

If the model matches the process, e.g. , the signal e is

Figure 4. Output data of the complex motor system equal to the disturbance d for all control signals u.The filter
is introduced to obtain a system that is less sensitive to
modeling error. From these reasons the following transfer
Figure 2.
function of the filter is choosen:

Figure 3. 1
Gf = 1+λs .
(10)
Figure 4.

Figure 5. To obtaing the PI controller from the IMC principle,consider

process with transfer function :

Kp
Gp = 1+Ts =Gm .
(11)

An approximate inverse of (11) is given by :

1+Ts
G-1
m= Kp
.
Figure 5. Measured and simulated motor system output (12)

IV. PI CONTROLLER DESIGN The controller, in accordance to previously written

relations, become :

In the simulation mode the parameters of the PI controller 1+sT

are obtained under feedback motor control using an Internal Gc s = K sλ .
p
Model Control (IMC) based tuning where controller setings are (13)
derived from a full-order controller. Simulations are done in
Matlab / Simulink simulation environment (Section V, part A). The transfer function (13) can be rewritten in the form of an
ideal PI controller :
The IMC methodology has been widely adopted for the
purpose of PID control tuning. The algorithm have the
adventage of only using a single tuning parameter to achive a
 tunning algorithm, described in section IV. The following
T 1 1
Gc s = 1+ =kc 1+ . 14 parameters values are obtained:
Kp λ sT sT

P=2.4285, I=6.0204.
The dominant closed-loop time constants is set to be λ=0.079.
For the experimental validation of the proposed PI controller The time response of the control system to step and pulse
the Siemens PLC with an appropriate software (Siemens Step7 signals with PI controller in simulation mode are depicted in
Micro/Win) is used. The PI parameters achieved in the Figures 9. and 10.
simulation mode are embedded into the PLC. The ladder
diagram of this controller is shown in Figure 7.
This ladder diagram which represents PI algorithm,
contains two lines. In the first line (Network 1) a scalling blok
with purpose of converting integer type of variable into real
variable type is used. The second (Network 2) implements the
PID0_INT subroutine using PID Wizard.

Figure 9. Simulation model in the programming

environment

Figure 9. Simulation in the programming environment

Figure 7. Step 7 Micro/Win ladder diagram

In the next section the simulation and experimental results

obtained using proposed system will be presented

V. SIMULATION AND EXPERIMENTAL RESULTS

A. Simulation results

The simulations of the DC motor control system has been

done in Matlab / Simulink simulation environment using
model shown in Figure 8.

Figure 10. The time response to pulse signal – simulation mode

Simulation results obtained demonstrate a satisfacory

behavior of the DC motor control system.
Figure 8. Simulation model in programming enironment
B. Experimental results

The PI controller is realised using Compensator design It is very hard to control this nonlinear system with PI or
tool as a part of Simulink simulation environment [12]. The PID controller especially when it is implemented as PLC-
parameters are determined using Internal Model Control based controller [9]. Also, there are measurement noise and
very large death zone (5V) what makes control real challenge. VI. CONCLUSIONS
As can be seen in Figure 11., PLC is connected to DC motor
system via analog I/O modul. Also, digital multimeter is used
for control of a set point variable from the power unit. In this paper the classical and robust PI controller has been
developed for permament magnet DC motor control. The
complex motor system contained of the tachogenerator, the
linear amplifier and the permanent magnet DC motor, is
nonlinear and very challenging to control, especially using
classical PI controller. For purpose of the system control, it
was identified first. Then the performance of designed PI
controller is tested in simulation and experimental modes. In
experiments the PLC-based PI controller was used. The
simulation and experiment results demonstrated effectiveness
of the proposed PI controller under feedback control of the
complex nonlinear DC motor system under different types of
internal and external disturbances.

REFERENCES

[1] Seun-Ming Beak, Tea-Yong Kuc, “An adaptive

Figure 11. Overall control and DC motor system
PI learning control of DC motor“, Computational Cybernetics and
Simulation, 1997, IEEE International Conference
[2] Cheng-Ching Yu, „Autotuning of PID Controllers. A relay feedback
approach“, 2nd edition, Springer-Verlag London Limited, 2006.
The power unit (Tektronix CPS 250) is equiped with
[3] H.R. Azevedo et al. “ A fuzzy logic controller for DC motor position
analog volmeter and potentiometer for setting voltage level. control”, Technical reports no 0-7803-0985-5,pp 18.-25. 1993.
Manually altered value of set point introduces an additional [4] G. R. Slemon, Electric machines and drives, New York: Addison-
noise on the actual output response (Figure 12.) Also, there is Wesley, 1992, pp. 503-510.
a load torque on a motor shaft that has a significant influence [5] M. A. Rahman and P. Zhou, “Field circuit analysis of brushless
on the control performance and stability. Despite all permanent magnet synchronous motors,“ IEEE Transactions on
constraints and disturbance, achieved system response using Industrial Electronics, vol. 43, no. 2, Apr. 1996, pp 256-267.
PLC as PI controller, is shown on Figure 12. The parameters [6] P .H. Lin , S. Hwang , J. Chou, “Comparison on fuzzy logic and PID
of PI controller are manually adjusted. controls for a DC motor position controller,“ Technical Report no. 0-
7803-1993-1, pp. 193-01935, 1994.
As it can be seen in Figure 12., green line that represents [7] S. Ayasun, G. Karbeyaz, „DC motor speed contol, Methods using
actual value of the system response, satisfactory tracks the set Matlab / simulink and their integration into undergraduate electric
point (blue line) under mentioned real time considerations and machinery course“, Nigde University, Turkey, 2007
disturbance. [8] Siemens S7-200 Programmable controller System manual, 09/2007
[9] A.Visioli, “Practical PID control”,Springer-Verlag London Limited,
2006.
[10] K.Astrom, T.Hagglund, “PID controllers :Theory,Design and
Tuning”,2nd edition, Instrument Society of America, 1995.
[11] J.Velagić, A. Galijašević, „Design of Fuzzy Logic Control of Permanent
magnet DC motor under real constraints and disturbance“, 3. IEEE
Multi-conference on Systems and Control(MSC 2009), Saint Petesburg,
Russia (in press), July 8-10, 2009
[12] The MathWorks Inc. , Matlab & Simulink R2008a. Control System
Toolbox , Product help

Figure 12. The time response of complex motor system