2011 International Conference on Electrical Engineering and Informatics

17-19 July 2011, Bandung, Indonesia

Design of Electronic Load Controller for a Self
Excited Induction Generator Using Fuzzy Logic
Method Based Microcontroller
Yahya Sofian
#1*
, Munawar Iyas
*2
#
Department of Electrical Engineering,Bandung State Polytechnic, Bandung, Indonesia
*
School of Electrical Engineering & Informatics, Bandung Institute of Technology, Bandung, Indonesia
1
sofianyahya@polban.ac.id
2
iyas@lskk.ee.itb.ac.id
Abstract The self-excited induction generators (SEIGs) are
considered to be well suited for generating electricity by means of
conventional energy sources and for supplying electrical energy
in remote and rural areas. Induction generators have many
advantages such as cost, reduced maintenance, rugged, and
simple construction, brushless rotor (squirrel cage). A three-
phase induction generator can be operated on a C-2C connection
for supplying single phase loads. The main disadvantage of SEIG
has is that it poor voltage regulation, and its value depends on
the prime mover speed, capacitance, load current and power
factor of the load. The electronic load controller (ELC) can be
used for maintaining constant voltage and frequency of SEIG
with variable consumer load driven by constant primer mover.
This paper presents the design and implementation of a digitally
controlled ELC using fuzzy logic method based microcontroller
for an SEIG feeding single-phase load. The ELC consist of a
rectifier, MOSFET as a chopper switch, ATMega32 micro-
controller, voltage sensor, opto-coupler, and resistive dump load
in which power consumption was varied through the duty cycle
of the chopper. However an ELC consist of electronics system, in
general, has complex nonlinear model with parameter variation
problem, and the control need to be very fast. The fuzzy logic
based controller gives nonlinear control with fast response and
virtually no overshoot. The proposed ELC has been tested by
step change in the consumer load. The simulated perfomance of
the controller is supplmented by experimental results.

Keywords Electronic load controller (ELC), Self excited
induction generator (SEIG), Fuzzy logic, Microcontroller
I. INTRODUCTION
Recently, most of the electricity is generated making use of
fossil fuels (coal, oil, and natural gas). These fossil fuels are
limited and will run out in the future. Fossil fuels are a non-
renewable energy source and they have continuously degraded
the environmental conditions. In such a situation researchers
are forced to make effort to make use of new and renewable
energy sources in an economical and safe way. Among
electrical generation using the renewable sources, micro-
hydro generation systems are an attractive alternative for
remote locations where an electrical grid is not available.
Furthermore, it is important to that such systems must be
consider economical, robust and require minimal maintenance
and are likely to be managed by un-skilled operators, because
they are usually installed remote from any maintenance
facilities.
The squirrel cage induction machine with capacitive self-
excitation, known as self-excited induction generators (SEIGs)
are considered as an alternative to the well-developed
synchronous generators. Induction generators are widely used
for micro-hydro powered electric generation, especially in
remote and isolated areas, because they do not need an
external power supply to produce the excitation magnetic field.
Furthermore, induction generators have more advantages such
as low cost, reduced maintenance, rugged and simple
construction, brushless rotor (squirrel cage), good over-speed
capability, and inherent protrction against short circuit [4].
Keeping the voltage and frequency of SEIG constant in spite
of the change in load, can be done by regulating the
capacitance value or by controlling the speed of the prime
movers. One alternative for supplying single phase loads
widely used in remote and rural areas is using three-phase
induction generator. It is used to supply the single phase loads
by using the connection C-2C. If the induction generator is
used to supply single phase with a constant load, the ELC
could be applied to maintain constant power output SEIG, for
this purpose the dump load must be connected in parallel with
the consumer load so that the total power load is generated by
the SEIG is constant. The amount of power that flowing into
the dump load is controlled by the electronic load controller
(ELC) [1-6].
Various controllers for SEIG have been reported in literature.
T.Chandra Sekhar et al. [9] proposed different voltage
regulation schemes such as power electronic controller,
electronic load controller and magnetic amplifier (saturable
core reactor). Bhim Singh et al [7] has been developed an
ELC for two winding single phase SEIG. Juan M.Ramirez et
al [3] proposed an ELC with antiparallel insulated-gate bipolar
transistor (IGBT) switches are used to control dump load
connection and disconnection. D.K.Palwalia et al [2] present
design and implementation of Digital Signal Processor (DSP)
based induction generator controller (IGC) for single phase
F6 - 2
978-1-4577-0751-3/11/$26.00 2011 IEEE


SEIG. Sarsing Gao et al [1] present analysis and the design of
a microcontroller (PIC 18F252) based SEIG-ELC. A PI
controller is used to provide proper control without steady
state errors or instability.
This paper present the design and implementation of
Electronic Load Controller (ELC) for a Self Excited Induction
Generator (SEIG). The SEIG using three phase induction
generator as a single phase SEIG feeding single phase loads.
The control technique to be used fuzzy based ATMega32
microcontroller. The mains reasons of using fuzzy logic
controller to gives nonlinear control with fast response and
virtually no overshoot. The proposed ELC will be tested under
single phase resistive load.
II. SYSTEM DESCRIPTION
The proposed as schematic diagram of ELC for SEIG using
fuzzy logic method based microcontroller is shown in Fig 1
and Fig 2 for supplying single phase loads. The ELC is
designed to operate an a delta connected three phase induction
machine with specification as follow 0.18 kW, 220 V, 1,67 A,
4 pole, 50 Hz, 1370 rpm. Two capacitors are connected C-2C
as shown in Fig 1, for facilities the single phase load of a three
phase induction generator. The ELC consists of a uncontrolled
rectifier with parallel-resonant filters, MOSFET, ATMega32
microcontroller, voltage sensor, opto-coupler, and a series
dump load (resistor). The proposed uncontrolled single phase
diode rectifier is shown in Fig 2, in which conversion of ac
power (from generator) to dc power. This rectifier with
parallel-resonant filter topology yields increased input power
factor and power density as compared with the conventional
diode rectifier [14]. A MOSFET is used as a chopper switch.
When gate pulse to MOSFET is high, the current flows
through the dump load and the power is consumed. A fuzzy
logic controller was applied to ELC, the controller regulates
the pulse width or duty cycle of chopper is decided by
difference of power generation to consumer load. The
ATMega32 microcontroller is used for generation of suitable
pulse width in accordance with consumer load. Since the
speed and hence the frequency is constant, the output power
remains constant when the voltage is maintained at the rated
value.
III. IMPLEMENTATION
The SEIG is connected in delta connection and excitation
capacitors are connected in a C-2C configuration. Excitation
capacitance has to provide required voltage on load at the
operating speed for the given induction machine operating as
a SEIG such that it generates rated terminal voltage at full
load. The output power of the SEIG is held constant at varying
consumers loads. In order to keep SEIG output power constant,
a dump load is connected in parallel with the consumer load
so the total generated power is held constant, that is

c d out
P P P + =

where Pout is the generated power of the generator (which
should be kept constant), Pc is the consumer load power, and
Pd is dump load power.

Fig. 1 Three-phase SEIG with ELC feeding single phase load


Fig.2 ELC consists of uncontrolled rectifier

The AC voltage from SEIG terminal rectified by means of an
uncontrolled rectifier with parallel-resonant filter. The dump
load is designed so that when the duty cycle of the switch is
unity, an operating power of 0.12 kW diverted to dump load.
The terminal voltage for feedback is sensed with voltage
transducer to achieve a DC value proportional to SEIG output
voltage. This analog voltage is given to ADC input of the
ATMega32 microcontroller. The sensed voltage is compared
with reference, which is taken as proportional to the rated
terminal voltage of the SEIG and may be altered when
required. The microcontroller is programmed in such a way
that the feedback voltage is compared with a reference value
of 220 V for every 50 milliseconds and an error signal is
generated. A fuzzy logic controller is used to gives nonlinear
control with fast response and virtually no overshoot. The
signal of PWM is fed to opto-coupler, which function as
insulation of the power circuit and the control circuit. The
signal drives the MOSFET switch an appropiate duty cycle.
The design aspects of single phase ELC are described in
subsequent sections.

A. Rectifier Circuit

The diode rectifier is designed using a single phase bridge
parallel-resonant filter, the rectifier which will be designed
has the following specifications:

Vrms 220 E
i
= Watt 150 P
r
=
Hz 50 Frek = % 5 Ripple =




Fig.3 Rectifier with parallel-resonant filter topology

A based on the calculation of the following component values
obtained [14]:

L
r
= 0,319 H ; C
r
= 3,52 uF ; C
0
= 58,39 uF


B. Specifications MOSFET and Dummy Load

The voltage rating of the uncontrolled rectifier and the
chopper switch (MOSFET) is calculated based on rms value
ac input voltage (VL) to the diode bridge and the average
value of dc voltage. The average value of the output dc
calculated based on the following equation:

V 198 220 x 9 , 0 V 9 , 0 V 2
2
V
L L dc
= = =

=


With estimates of 10% more voltage than the voltage rating
for the transient state, the rms voltage to 242 V (220 +22) and
the peak voltage is thus, calculated as :

V 2 , 342 242 x 2 V
dc
= =


Current rating of single phase uncontrolled rectifier and the
chopper switch is calculated as :

A 68 , 0
220
150
V
P
I
L
ac
= = =

A 51 , 1 9 , 0 / ) 2 x 68 , 0 ( 9 , 0 / ) 2 x I ( I
ac peak
= = =


From the calculation above, the maximum voltage is 342.2 V
and maximum current 1.51 A.

The rating of dump load resistance, R
d
is calculated as:

= = = 36 , 261
150
) 198 (
P
) V (
R
2 2
dc
D


C. Opto-coupler Circuit

The opto-coupler circuit that also functions as a gate driver
using the TLP 250.



Fig.4 Opto-coupler circuit with TLP 250

D. Fuzzy Logic Controller

A fuzzy logic controller consists of four main components as
fuzzification, rule base, inference mechanism and
defuzzification. The fuzzification converts its inputs into
fuzzy values with membership functions in the form of
triangle, trapezoid, bell or other appropriate forms expressed
by the fuzzy linguistic variables. The rule base contains the
experts linguistic descriptions expressed in the form of
logical implications such as IF x is positive THEN y is big.
The inference mechanism evaluates fuzzy information to
activate and apply control rules. The defuzzification that uses
methods such as centre of gravity, maximum and weighted
mean converts the inference mechanism into the crisp values
applied to the actual system.
The proposed schematic of an ELC is shown in Fig. 2.
There are two inputs and one output of the fuzzy controller.
The first input is the error between reference value that is
desired output value and generator output value. The second
input is the derivative of the error. The inputs are given by:

e(k) = r(k) y(k)
e(k) = e(k) e(k-1)

The design of the fuzzy controller depends on information
about the system behavior or experience of a human expert.
The fuzzification stage is determined by the choice of the
range, shape and number of the membership functions. The
input membership functions for the error and the delta error to
the fuzzy controller, the positioning universe was divided into
seven domains which are negative big (NB), negative medium
(NM), negative small (NS), zero (Z), positive small (PS),
positive medium (PM), and positive big (PB). The output
membership functions are chosen to be nonuniformly
distributed seven singletons functions. The output
membership function processed by the fuzzy logic algorithm
produces the PWM singletons taken as output assignments for
the control dump load.




Fig.5 Membership functions of error and the delta error


Fig.6 Membership functions output

The set of rules for fuzzy controller is shown in Table I.

Table I. Fuzzy Associative Memories (FAM)


E. Design of Fuzzy Logic Program

The software used to program ATMega32 microcontroller is
CodeVisionAVR by using C language. The program has
designed based on the flow chart as shown in the figure 7.
IV. EXPERIMENTAL RESULTS
Experimentation has carried out on a three-phase SEIG-ELC
system feeding a single-phase load with ATMega32
microcontroller has been developed and tested in laboratory
under various operating conditions. A single phase 0.18 kW,
220 V, 1.67 A, 4 pole, 50 Hz, 1370 rpm squirrel cage
induction machine is used as a single phase self-excited
induction generator. The SEIG is driven by 220 V, 1.5 kW,
1500 rpm shunt wound DC machine used as a prime mover.
To generate 0.12 kW at 220 V and rated speed, 20F and
10F capacitors of 400 V are connected as C
1
and C
2
across
the delta winding terminals of the SEIG.



Fig. 7 Flow chart of fuzzy logic program

The controller is designed to keep the operating point at
these conditions. The dump load is set as 0.12 kW in order to
consume the entire consumer load when the chopper duty
cycle is 100% i.e., when the consumer load is zero.
First test is experiment open loop transient response of the
generator without load. The results are shown in Fig 8, the
generator is rotated by prime movers with a constant speed
1510 rpm. The second test is still in open loop condition by
giving the load directly to the generator at a certain power
value, with initial voltage generator always at 231 volts.




Fig 8. Open loop transient response of generator
Error
dError
NB NM NS Z PS PM PB
NB PB PB PM PS PS Z NS
NM PB PM PM PS Z NS NM
NS PB PM PS Z Z NS NM
Z PB PM PS Z NS NM NB
PS PM PS Z Z NS NM NB
PM PM PS Z NS NM NM NB
PB PS Z NS NS NM NB NB



Fig 9. Graph of V = f(Pload) direct load open loop
Fig 9 is shows the graph based on the results of experimental
by giving direct loads. Fig 10 is shows the open loop transient
response when a load generator 80 watt (66.66%) is given
directly. The graph and open loop transient response of the
test generator with direct loading shows that the direct voltage
drops quickly when the load is given.


Fig 10. Open loop transient response when a load generator 80
watt (66.66%) is given directly
Closed loop response experiment is conducted to test the
performance of a induction genenerator controlled using
fuzzy logic method based ATMega 32 microcontroller. Closed
loop response testing with different loading conditions, i.e the
direct loading and gradual loading. Set point on a closed loop
testing is 228 volts, 3.63% of the planned reference voltage of
220 volts.

Fig 11. The generator transient response, when tested with a
load of 112 watts (93.3%).
Fig 11 shows the generator transient response, when tested
with a load of 112 watts (93.3%). The generator is only able to
survive on the voltage 228 volts for 4 seconds, after which it
shows an unstable response that resulted in loss of voltage
generator.
The test of load adding gradually has performed in the range
of 0 to 122 watt of power. Fig 12 shows the voltage response
when given a load of 122 watts (101.6%), the settling time is
5.5 seconds with 212 volt voltage value, -7.01% steady state
error, and error -3.63% against the reference voltage.


Fig 12. Close loop transient response of load 122 watt
(101.6%) adding gradually

V. CONCLUSIONS
Based on the design and testing electronic load controller for a
self excited induction generator using fuzzy logic method
based microcontroller has obtained conclusions. When testing
the generator without controller with the single phase loads
directly, the generator is only able to loaded 67 watt with 147
volt terminal voltage and frequency of 42.7 Hz. While with
load gradual, the same generator could only load 67 watt with
167 volt terminal voltage and frequency of 42.9 Hz. The ELC
of self excited induction generator can supply the single phase
resistive loads up to 122 watts or 66.77% of the rating as the
motor, with 3.63% voltage regulation, frequency error of -
2.6% to -3.6%, voltage THD average of 2.28%, and unity
power factor.

REFERENCES

[1] Sarsing Gao, S. S. Murthy, G. Bhuvaneswari, and M. Sree Lalitha
Gayathri, Design of a Microcontroller Based Electronic Load
Controller for a Self Excited Induction GeneratorSupplying Single-
Phase Loads, Journal of Power Electronics, Vol. 10, No. 4,pp.444-
449, 2010.
[2] Palwalia, D.K.; Singh, S.P.; , Design and implementation of induction
generator controller for single phase self excited induction
generator, Industrial Electronics and Applications, 2008. ICIEA 2008.
3rd IEEE Conference on , vol., no., pp.400-404, 3-5 June 2008.
[3] Juan M. Ramirez and Emmanuel Torres M., An electronic load
controller for self excited induction generators in IEEE PES General
Meeting, June 24-28, 2007.
[4] R. C. Bansal, Three-Phase Self-Excited Induction Generators: An
Overview, IEEE Transactions on Energy Conversion, Vol. 20, No. 2,
pp. 292-299 June 2005.
[5] B. Singh, S.S.Murthy and Sushma Gupta, Analysis and implement-
tation of an electronic load controller for a self excited induction


generator IEE Proc.-Gener. Transm. Distrib., vol. 151, no. 1, pp. 51-
60, January 2004.






[6] Bhim Singh, S.S. Murthy & Sushma Gupta, "Transient Analysis of
Self-Excited Induction Generator with Electronic Load Controller
(ELC) Supplying Static and Dynamic Loads", IEEE Transactions on
Industry Applications , pp .1194-1204,Sept-Oct. 2005.
[7] Bhim Singh, S.S. Murthy & Sushma Gupta ,An Electronic Voltage
and Frequency Controller for Single-Phase Self-Excited Induction
Generators for Pico Hydro Applications,IEEE PEDS, pp 240-
245,2005.
[8] B. Singh, S. S. Murthy and S. Gupta, Analysis and Implementation of
an Electronic Load Controller for a Self- Excited Induction Generator,
IEE Proceedings Generation Transmission and Distribution, Vol. 151,
No.1, pp. 51-60,January 2004.
[9] Chandra, T.S., Bishnu, P.M., Voltage Regulators for Self Excited
Induction Generator, IEEE Transactions on Energy Conversion,
Vol.20, no. 4, , pp 460463,April 2004.
[10] J.B. Ekanayake, Induction generators for small hydro schemes,
Power Engineering Journal, Vol. 16, Issue 2, pp.61 67,2002.
[11] T. F. Chan, Single phase operation of a three phase induction
generator with the Smith connection, EEE Trans. on EC, vol.17, No.1,
pp 47-54,March. 2002.
[12] Li Wang, Chang-Min Cheng, Excitation capacitance required for an
isolated three-phase induction generator supplying a single-phase load,
in Proc. IEEE Power Engineering Society Winter Meeting, Vol. 1, pp.
299-303, 2000.
[13] Alolah, A.I., Alkanhal, M.A. Excitation requirements of three phase
selfexcited induction generator under single phase loading with
minimum unbalance. u: Power Engineering Society winter meeting,
IEEE, vol. 1,pp. 257-259,2000.
[14] A.R. Prasad, P.D. Ziogas and S.N. Manias, A Novel Passive Wave
shaping method for single phase Diode Rectifiers, IEEE Transactions
on Industrial Electronics, Vol. IE-37, No.6, pp. 521-530,Dec. 1990.
[15] Simoes,M.Godoy dan Farret,Felix.A.,Alternative Energy System :
design and Analysis with Induction Generator,, 2
nd
ed, CRC
Press,2008.
[16] Wekhande, Shashank. dan Agarwal, Viviek., A New Varibale Speed
Constant Voltage Controller for Self-excited Induction Generator,
Electric Power Systems Research, Elseiver, 59, 157-164,2001.
[17] L. Zadeh, Outline of a new approach to the analysis of complex
systems and decision processes, IEEE Trans. Syst., Man Cybem., vol.
28, pp.28-44, 1978.