147 337 1 SM PDF
147 337 1 SM PDF
147 337 1 SM PDF
Abstract
This paper propose about using PID control system based on Kp, Ki,
and Kd parameter determination with scheduling process from fuzzy
logic. Control system is used to arrange speed of three phase
induction motor using IFOC method. This method can be minimized
the main problem from speed control of induction motor which is a
transient condition. The robustness validation from this system use
testing process of dynamic speed which is compared with the other
control system to know the system performance in transient
condition such as (rise time, overshoot, undershoot and settling
time). The result shows using the proposed system has better
performance responses which is requiring 0.001 seconds time in
transient condition up to steady state condition without overshoot
and undershoot problem.
1. INTRODUCTION
Nowadays, like a consequence of the important progress in the power
electronics and of microcomputing, the control of the AC electric machines
known a considerable development and a possibility of the real time
implantation applications. It is widely recognized that the induction motor is
going to be the main actuator for industrial purposes [1], [2]. Indeed, as
compared to the DC machine, it provides a better power or mass ratio, a
simpler maintenance and relatively lower cost. However, it is traditionally for
a long time, used in industrial applications that do not require high
performances, this because its control is a more complex problem, its high
nonlinearity and its high coupled structure. Furthermore, the motor
parameters are time-varying during the normal operation and most of the
state variables are not measurable. On the other hand, the direct current
(D.C) machine was largely used in the field of the variable speed applications,
where torque and flux are naturally decoupled and can be controlled
independently by the torque producing current and the flux producing
current. Since Blashke and Hasse have developed the new technique known
as vector control [1-4], the use of the induction machine becomes more and
more frequent. This control strategy can provide the same performance as
achieved from a separately excited DC machine, and is proven to be well
adapted to all type of electrical drives associated with induction
machines[5].The vector control technique has been widely used when high
performance rotary machine drive is required, especially the Indirect Field
Oriented Control (IFOC) that is the most effective vector control of three
phase induction motor due to the simplicity of designing and implementation
[6]. Decoupled torque and flux control in IFOC of induction machines permits
higy dynamic response.While, induction motor control cann’t apart from its
previous condition. The real nonlinear characteristic from motor and the
modification parameter from control caused to the modification value in
control process when the time of trancient condition was toward to the
steady state in set point achievement. Using appropriate control system can
be minimized its problem.
The most widely used controller in the industrial applications is the PID-
type controllers because of their simple structures and good performances in
a wide range of operating conditions [7]. In the literature, the PID controllers
can be divided into two main parts: In the first part, the controller
parameters are fixed during control operation. These parameters are
selected in an optimal way by known methods such as the Zeigler and
Nichols, poles assignment…etc. The PID controllers of this part are simple
but cannot always effectively control systems with changing parameters or
have a strong nonlinearity; and may need frequent on-line retuning [8]. In the
second part, the controllers have an identical structure to PID controllers but
their parameters are tuned on-line based on parameters estimation of the
process. Such controllers are known as adaptive PID controllers [2].
The application of knowledge-based systems in process control is growing,
especially in the field of fuzzy control [9-12]. In fuzzy control, linguistic
descriptions of human expertise in controlling a process are represented as
fuzzy rules or relations. This knowledge base is used by an inference
mechanism, in conjunction with some knowledge of the states of the process
(say, of measured response variables) in order to determine con trol actions.
Although they do not have an apparent structure of PID controllers, fuzzy
logic controllers may be considered nonlinear PID controllers whose
parameters can be determined on-line based on the error signal and their
time derivative or difference [11].
This paper proposedthe application FGS-PID for speed control of three
phase induction motor based on IFOC. The new scheme utilizes fuzzy rules
and reasoning to determine the controller parameters, and the PID controller
generates the control signal for the process on IFOC systems.It is
demonstrated in this paper that human expertise on PID gain scheduling can
be represented in fuzzy rules. Furthermore, better control performance can
be expected in the proposed method than that of the PID controllers with
fixed parameters on dynamic conditions. The investigation of transient
condition (involve rise time, overshoot, undershoot, and settling time) in
speed of dynamic is presented section 2. The comparation between FGS-PID
and other control is presented section 3.
2. RELATED WORKS
The theory of Fuzzy Gain Scheduling-PIDwas firstly developed by Zhen-
Yu Zhao, Masayoshi Tomizuka, and Satoru Isaka in 1993.UsingFuzzy gain
scheduling could be used to determine gain parameters of PID control.PID is
one of popular control in industrial because it has simple design,robust
control andalso easy to be implemented. However, parameter of gain PID is
difficult to be determined which needs re-tuning to get a good result. One of
developed tuning methodsis Ziegler Nichols that is simple indetermining
ofKp gain, Ki, and Kd of PID parameters [7].
Fuzzy logic was presented by Prof. L.Zadeh in 1965 from California
University. Fuzzy logic was founded after crips logic method. The value of
crips logic is true “1”or false “0”. Fuzzy logic has uncertain value between
true and false. Fuzzy logic allows membership value between 0 and 1 and
also several variables that are expressed in linguistic language such as
positive, zero, and negative.The principle of FGS-PID is using two inputs
(error and Δ-error) with output fuzzy kp’, kd’, and α. Fuzzy kp’ is used to
schedule parameter gain (p), kd’ is used to schedule parameter gain (Td), and
α is used to schedule gain (Ti).By applying gain system scheduling of PID
parameter, PID tuning process is easier to do in stabil or unstabil condition.
In Zhao and friends’ presentation, also comparing the output response
system of FGS-PID control, PID-ZN, and Kitamori. Thecomparations are done
from output control system for second, third, and fourth order. The result
shows that FGS-PID has more optimal response because it can reduce
overshoot condition and osilationsystem[13].
Then, Bousserhane presents optimal fuzzy gains scheduling of pi
controller for induction motor speed control. To overcome the disadvantages
of PID controllers and FLC, we propose in this paper a combination between
them together. PID parameters controller can be tuned on-line by an adaptive
mechanism based on a fuzzy logic for induction machine speed control.
Design of an optimal fuzzy gain scheduling of PI controller combines the
merits of the sliding mode control and the fuzzy inference mechanism is
proposed. A fuzzy gain scheduling of conventional PI controller is
investigated, in which the fuzzy logic system is used on-line to generate the PI
controller parameters [2].
In this paper, it will be proposeddesign about speed control of three-
pahse induction motor based on IFOC with combining PID controller and
Fuzzy as gain scheduling parameter of kp, ki, and kd that can simplify the
3. ORIGINALITY
The constribution of this paper is to implement speed control of
induction motor based on PID with scheduling and reasoning from fuzzy logic
control to obtain parameter system which is suitable with determined
condition using IFOC method. Using the proposed method can be minimized
the problem of induction motor rotation in transient condition which has
non-linear characteristic. The principle of fuzzy gain scheduling is used to
arrange the PID gain that is appropriate with the declared condition. The
output of control system is control signal which is used for IFOC system in
control of induction motor system. The implementation of proposed IFOC
method is simply to apply which is using control system based on vector
technique. The vector technique is oriented to the field control which can be
controlled the torque unit and flux separately. This system is usually called
by decoupled system. the modification of flux and torque can be influenced to
the motor rotation which is represented as current and speed modification.
The parameter modification from control system can be decreased the
performance of control system which is showed from motor responses in
transient condition. Using FGS-PID is one of solution to minimize this
problem. The control design is arranged for 0 – 1000 Rpm speed which can
be given the better result in modification parameter condition. Verification
process and the robustness test of this sytem is used dynamic speed control
sytem which is observed in transient condition involve rise time, overshoot,
undershoot and time settling.
4. SYSTEM DESIGN
4.1 RESEARCH METHOD
4.1.1 PID Controller
PID control is a combination from three methods; those are
proportional control, integral, and derivative. Combination of these methods
can be used to minimize disadvantage of each controls. Proportional control
(P) has function to accelerate up to set point condition, but it can cause
overshoot. Disadvantage of control (P) can be solved by adding integral
control (I) that functions to reduce overshoot, but system will be slow. So it
needs derivative control (D) to accelerate system. But sometimes, it causes
osilation condition before getting steady state. The equation of PID control is
given to this following equation [13]:
= + + (1)
Type of Controller
The parameters Kp’, Kd’, α are determined by a set of fuzzy rules of the form.
The membership functions (MF) of these fuzzy sets for error [e (k)] and delta
error [Δe (k)] are shown in Fig. 3. In this figure, N represents negative, P
positive, ZO approximately zero, S small, M medium, B big. Thus NM stands
for negative-medium, PB for positive big, and so on.
S MS M B
Figure 5. Output of Membership function α
The rule base of Fuzzy gain scheduler to deremine PID parameter shown in
table 2, 3, and 4 [2].
Based on the result from Kp’, Kd’, α value, PID and FGPS parameter can be
calculated using this equation:
Kp2/(αKd) (7)
(Kp max, Kp min) and (Kd max, Kd min) is obtained from this following
equation:
+ + . 1∗
∗
() = , ∗ - ∗ . / ∗ 32∗ (10)
0 /
. 6
45 = 9:;
0 78
(11)
< /
.
=5 = >/ is the time constants of rotor. The number of direct-axis stator
/
current reference ∗
() is based on the flux reference input 4∗5 as follow:
3∗
∗
) = . (12)
0
?@ = AB + A)C (13)
∗
Slip speed is calculated from stator current reference () with this following
motor parameter:
. >
A)C = 0∗ ∗ / ∗ ∗
()
3 .
(14)
/ /
HNE? −E P?
F∗ M 9 √, 9 √, T ∗
I G ∗ J = L(− + HNE? + + E P?) ( E P? + HNE?) SU ∗V
L 9
+ +
S D
(15)
H∗ √, 9 √,
K(− + HNE? − + E P?) (+ E P? −
+
HNE?) R
Te* ∗
() Y
IFOC
W
AB
∗
FGS ∗
PID Flux* )
PID X
∗
Speed
∗ W ∗
Y X
Reference
3Phase
Theta Inverter
IM
3~
AB
Actual Speed
The system IFOC the transformation stationer frame into rotational frame
using Clark and Park Transformation. Clark transformation is used to modify
three-phase stationary (ia, ib, ic) to be two-phase stationary (iα, iβ). While,
Park Transformation modify two-phase stationary to the two phase
rotational (id, iq) that was illustrated in Fig. 10.
Settling Time
Figure 11.Simulation result speed controlwithIFOC on the speed motor 350 Rpm
The simulation result(Fig.11) can see the output response system. The rise
time value to get speed in 350 Rpm was 0.23 second. The maximum
overshoot is 28 %, and time settling 0.89 second before steady
speedcondition.
Figure 12.Simulation result speed control with IFOC on the speed motor 400 Rpm
While, the number ofrise time to get speed 400 Rpm was needed 0.256
second that has maximum overshoot 29 %, and time settling 0.74 second
before condition steady speed, depicted in Fig.12.
Figure 13.Simulation result speed control with IFOC on the speed motor 450 Rpm
In Fig.13 speed 450 Rpm was needed 0.23 second rise time, 28 % maximum
overshoot, and time settling 0.93 second before condition steady speed.
In Fig. 15, showsthe diagram block of speed control FGS-PID for three-phase
induction motor based on IFOC.
Figure 16.Simulation result FGS-PID at speed modification 300 Rpm – 600 Rpm
The performance result from FGS-PID control for three phase induction
motor of IFOC based is using set point variable as shown in Fig.20. The sytem
responses can be analyzed or observed with 0.1 seconds sampling time as
follows:
The next condition, we declared 400 Rpm setpoint and 0.05 seconds steady
state condition at the runned simulation. It will be reduced up to 200 Rpm
setpoint and 0.1 seconds in steady state condition. For the last, the setpoint is
set in 150 Rpm and 0.1 seconds steady state condition.
Figure 17.Simulation result FGS-PID at speed modification 400 Rpm – 200 Rpm
The result responses sytem from second condition of FGS-PID can see in
Fig.17 and it analyzed or observed show in table 7:
From the observation above is known that FGS-PID control has a better
response can be solve the problem in transient condition to get steady state
condition only takes less than 0.01 seconds without getting overshoot and
undershoot.
Figure 18. Comparison between FGS-PID, PID-ZN, and Fuzzy Backstepping at speed
of dynamic
From the comparison result of the control system in Fig. 18, can be observed
the transient condition result involved rise time, overshoot undershoot and
settling time. The result shows the better performance from the proposed
method using FGS-PID in modification parameter which is verifed by the
result from Table 7-10. The longest time for modification only needs 0.0014
seconds to achieve steady condition without overshoot and undershoot
condition. While using PID-ZN take 0.007 seconds to reach steady condition
and the Fuzzy-Backstepping need 0.0014 seconds. It proves that the
proposed method has better performance in modification parameter.
The results of the scheduling gain of PID parameters are presented below:
On the graph in Fig 25, Fig 26, and Fig 27,it can be observed that scheduling
gain process will be performed when control process parameter is modified,
so it still gets a good result. From the results above with a sampling time of 0-
1 second, it is observed on performed scheduling process. At time 0 – 0.25
second with setpoint 250 rpm, then obtained value gain of proportional (Kp)
is 8,256, time integral constant (Ti) is at 1.474, and time derivative constant
(Td) is 0.2455. When there is a modifiedsetpoint to be 500 rpm at 0.25 - 0.5
seconds,it is performed the scheduling again to get a suitable value.(Kp) is
8.2537, (Ti)is 1.486, and (Td)is 0.2437. And then, when setpointis100 Rpm at
time 0.5 - 0.75 second, the result of scheduling (Kp)is 8.2577, (Ti)is 1.478,
and (Td)is 0.2457. When the last condition, setpointis 300 Rpm at time 0.75 -
1 second, (Kp)is 8.255, (Ti)is 1.478, and (Td)is 0.24457.
6. CONCLUSION
Design of proposed FGS-PID in the three-phase induction motor speed
control using IFOC methods can provide good results. It is verified with
testing a reliability of control system on dynamic speed. With this method of
observation in transient conditions (involve rise time, overshoot, undershoot,
and settling time) it can be knownthat the performance of the control system
is in good condition although the parameters of process control is modified,
because it has a fast response below 0.0004 second to achieve a steady state
condition without getting overshoot and undershoot. The other advantage of
FGS-PID based on IFOC in induction motors is, it can have linearization
characteristic of the motor, so it can be easy to control as DC motor. Thus,
when getting linear conditions, FGS-PID control can more easily generate
optimum performance even though applied to condition of fluctuation.
ACKNOWLEDGEMENTS
The first I extend thanks to “ALLAH SWT” always give love and help me in
researcher, secondly my family is always pray for me and then i thanks
to“Kementerian Riset, Teknologi dan Pendidikan Tinggi Republik Indonesia”
which was approve the scholarship freshgraduate program in Study Program
of Applied Master’s Degree Electrical Engineering at Politeknik Elektronika
Negeri Surabaya and also the all friend in “Tanggul Islamic Centre” who have
helped this research better.
REFERENCE
[1] J.P. Hautier J.P. Caron, Modeling and Control of Induction Machine,
Technip ed., 1995.
[2] I.K. Bousserhane, A. Hazzab, M. Rahli, B. Mazari, and M. Kamli "OPTIMAL
FUZZY GAINS SCHEDULING OF PI CONTROLLER FOR INDUCTION
MOTOR SPEED CONTROL ," Acta Electrotechnica et Informatica , vol. 7,
no. 1, 2007.
[3] R.D.Lorenz and D.B. Lawson, "A Simplified Approach to Continuous On-
Line Tuning of Field-Oriented Induction Machine Drives," IEEE Trans. On
EMITTER International Journal of Engineering Technology, ISSN: 2443-1168
257 Volume 4, No. 2, December 2016