Available online at www.sciencedirect.

com

Energy Procedia 18 (2012) 560 – 570

Fuzzy Logic Controller Based Three-Phase Shunt Active

Power Filter for Compensating Harmonics and Reactive Power
under Unbalanced Mains Voltages
R. Belaidia,b*, A. Haddouchea, H. Guendouza
a
Laboratoire des Systèmes Electromécaniques,Université Badji Mokhtar, 23000, Annaba, Algeria
b
Unité de Développement des Equipements Solaires (UDES), Route national N°11BP386, Bou-Ismail 42415, Algeria

Abstract

In this paper, a shunt Active Power Filter (APF) is proposed for the compensation of harmonic currents and reactive
power in polluted environment and under unbalanced mains voltage. For this purpose, a fuzzy logic controller is
developed to adjust the energy storage of the dc voltage. The reference current computation of the shunt APF is based
on the instantaneous reactive power (p-q) theory. We applied the system based on PLL (Phase Locked Loop) in order
to control the shunt APF under unbalanced mains voltage. Hysteresis Controllers is used to generate switching signals
of the voltage source inverter. MATLAB/SIMULINK power system toolbox is used to simulate the proposed system.
The results show the effectiveness of fuzzy logic control to optimize the energy storage of the DC capacitor, the
sinusoidal form of the current and the perfect of the reactive power compensation. The proposed system has achieved a
low Total Harmonic Distortion (THD) which demonstrates the effectiveness of the presented method.

© 2012
2010 Published
PublishedbybyElsevier
Elsevier Ltd.
Ltd. Selection and/or peer review under responsibility of The TerraGreen Society.
Open access under CC BY-NC-ND license.
Key words: Shunt active power filter, Harmonics, Fuzzy logic control, Reactive power, (p-q) theory, PLL, Hysteresis Controllers and
THD.

1. Introduction

The power quality (PQ) problems in power utility distribution systems are not new, but only recently
their effects have gained public awareness. Advances in semiconductor device technology have fuelled a
revolution in power electronics over the past decade, and there are indications that this trend will continue
[1]. However the power electronics based equipments which include adjustable-speed motor drives,
electronic power supplies, DC motor drives, battery chargers, electronic ballasts are responsible for the rise

* Corresponding author. Tel- fax: +213 24 41 01 33.

E-mail address: rachidi3434@yahoo.fr.

1876-6102 © 2012 Published by Elsevier Ltd. Selection and/or peer review under responsibility of The TerraGreen Society.
Open access under CC BY-NC-ND license. doi:10.1016/j.egypro.2012.05.068
 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570 561

in PQ related problems [2-3]. These nonlinear loads appear to be prime sources of harmonic distortion in a
power distribution system. Harmonic currents produced by nonlinear loads are injected back into power
distribution systems through the point of common coupling (PCC).These perturbations (harmonics) are the
origin of many problems and affect electrical equipments connected to the power supply.
These harmonics induce malfunctions in sensitive equipment, Voltage stresses, increased heating in the
conductors and harmonic voltage drop across the network impedance that affects power factor.
Traditionally passive filters have been used to compensate harmonics and reactive power; but passive
filters are large in size, aging and tuning problems exist and can resonate with the supply impedance.
Recently active power filters are designed for compensating the current-harmonics and reactive power
simultaneously. The shunt APF based on Voltage Source Inverter (VSI) structure (a DC energy storage
device in this case is capacitor) is an attractive solution to harmonic current problems. The shunt APF is
designed to be connected in parallel with the nonlinear load. It detects the harmonic current of nonlinear
load and injects into the system a compensating current, identical with the nonlinear load harmonic current
but in opposite phase. Therefore, the net current drawn from the distribution network at the point of
coupling of filter and the load will be a sinusoidal current of only fundamental frequency.
One of the important tasks in the shunt APF design is the maintenance of constant DC voltage across the
capacitor connected to the inverter. This is necessary because there is energy loss due to conduction and
switching power losses associated with the controllable switches of the inverter, which tend to reduce the
value of voltage across the DC capacitor. Generally, PI controller [6] is used to control the DC bus voltage.
The PI controller based approach requires precise linear mathematical model which is difficult to obtain.
Also, it fails to perform satisfactorily under parameter variations, non-linearity, and load disturbances [7].
Recently, fuzzy logic controller has generated a great deal of Interest in various applications and has been
introduced in the power electronics field [3-5]. The advantages of fuzzy logic controllers over the
conventional PI controller are that they do not need an accurate mathematical model, they can work with
imprecise inputs, can handle nonlinearity, and may be more robust than the conventional PI controller. In
the other hand, In APF design and control, p-q theory was often served as the basis for the calculation of
compensation current [8]. In this theory, the mains voltage was assumed to be an ideal source in the
calculation process. However, in most of time and most of industry power systems, mains voltage may be
unbalanced and/or distorted, in this case the control using the p-q theory does not provide good
performance [9-10].
This paper presents an analysis and simulation of a shunt APF topology that achieves simultaneously
harmonic current damping and reactive power compensation under unbalanced mains voltages. To
optimize the energy storage, a fuzzy logic controller is developed to adjust the energy storage of the dc
voltage to its reference and to attenuate harmonic frequencies resulting from power fluctuations. For the
reference current computation of the shunt APF, we used a new technique with p-q theory based on PLL as
a suitable method to unbalanced mains voltages and for the control of shunt APF. Hysteresis Controllers is
used to generate switching signals of the voltage source inverter.
Figure 1 shows the proposed system; the three phase shunt APF system is based on a three-phase inverter
with six controllable switches, each of the switches in the switching network is IGBTs with anti-parallel
diode to allow current flow in both directions. The shunt APF is designed to be connected in parallel with
the nonlinear load. It is connected to the distribution network in the PCC. The network is represented as an
unbalanced voltage source.
562 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570

Fig.1.General structure of the Shunt Active Power Filter

2. Current reference generation algorithm for shunt APF

There are different methods for generating the current reference for shunt APF which are classified as
frequency, time and time-frequency approaches. Fast Fourier Transformer (FFT) [11] and adaptive neural
network [12] in frequency domain, synchronous reference frame theory d-q-0 (SRF) [13] and P-Q theory
[9] in time domain and the other methods such as small wave technique and one-cycle control or separation
with suitable digital or analogue filters have wide applications. In this paper the current reference for active
power filter is generated using P-Q theory.

2.1 P-Q Theory

This theory (Akagi, Kanazawa and Nabae in1983-84) with the objective of applying it to the control of
APFs [14].This theory is based on time-domain, what makes it valid for operation in steady-state or
transitory regime, as well as for generic voltage and current power system waveforms, allowing to control
the APFs in real-time. Another important characteristic of this theory is the simplicity of the calculations,
which involves only algebraic calculation (exception done to the need of separating the mean and
alternated values of the calculated power components). [14-15]
P-Q theory is suitable for the shunt APF control, specifically for reference current calculation. It is based
on instantaneous voltage and current in three phase system (3 or 4 wire). It applies an algebraic
transformation (Clarke transformation) of three-phase system voltages and load currently in the a-b-c
coordinates to the – coordinates by the relations:
 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570 563

1 1
V 1 Va
2 2 2
. Vb
3
V 3 3 Vc
0
2 2 (1)
1 1
I 1 ia
2 2 2
. ib
3
I 3 3 ic
0
2 2 (2)

The instantaneous power for the three-phase system is as follows:

P V V I
q V V I
(3)

Where: P is the instantaneous real power.

q is the instantaneous imaginary power.
By observing the formulations of P and q , it is possible to put them in the following form:
~
P P P
q q q~ (4)

Where:
P : DC component related to fundamental active current conventional.
~
P : AC component of P , devoid of mean value and associated with harmonic caused by the AC
component of instantaneous real power.
q : DC component related to the reactive power generated by the components fundamental currents and
voltages.
q~ : AC component of q and related to harmonic currents caused by the components of AC instantaneous
reactive power.

I 1 V V P
2
I V V2 V V q
(5)

I ~
1V V P 1V V 0 1V V P
I V V 0 V V q V V q~
(6)

Reactive current Active current Harmonic current

2 2
V V
With:
564 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570

Three phase distorted currents representing identified currents (reference currents Iref), are calculated from
( - ) inverse transformation (Clarke transform) shown in the relation (7) presented below.

I 1 0
ref 1
2 1 3 I (7)
I ref 2
3 2 2 I
I ref 3 1 3
2 2
The voltage must be of good quality (sinusoidal and balanced); otherwise the method of the p-q theory
does not apply. Since the network voltage is often unbalanced and/or distorted, and to generalize the
application of the identification method, the PLL-based system is proposed to extract the fundamental
component of the direct voltage.

3. PLL operating principle

The PLL system used here can extract the phase of the direct component of voltage which is necessary
for the interference currents identification. Its operation is based on Park transformation P ( ) of the
Voltages V a-b-c, measured at the PCC of the shunt APC. The angle of this rotation results from the
integration of the pulse determined by the regulator. This can be achieved by selecting the Vd_ref value.
~
The PLL will be locked when the estimated angle will be equal to the forward voltage ( ˆd d ) [15].

Fig.2. Overall structure of PLL-based system

Finally, this algorithm (p-q theory) can be represented as shown in the block diagram of figure 3.

Fig.3. Calculations of the p-q theory

 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570 565

4. Fuzzy Logic Controller

The concept of Fuzzy Logic Controller (FLC) was proposed by Professor Lotfi Zadeh in 1965, at first as
a way of processing data by allowing partial set membership rather than crisp membership. Soon after, it
was proven to be an excellent choice for many control system applications.
Fuzzy control is based on a logical system called fuzzy logic. It is much closer in spirit to human thinking
and natural language than classical logical systems [16]. Nowadays, fuzzy logic controller is used in almost
all sectors of industry, power systems and science. One of them is the harmonic current and reactive power
compensation control [17].
The structure of a fuzzy logic control system shows in figure 3. This figure shows two inputs the error (E),
its variation ( E) and one output (the command D E).

Fig.4. Structure of fuzzy logic controller

4.1 Fuzzification

The fuzzification module converts the crisp values of the control inputs error signal E and its variation
E into fuzzy values. A fuzzy variable has values which are defined by linguistic variables (fuzzy sets or
subsets) such as low, Medium, high, big, slow . . . where each is defined by a gradually varying
membership function.

4.2 Rule Elevator

The basic fuzzy set operations needed for evaluation of fuzzy rules are AND, OR and NOT
AND -Intersection: A B min[ A ( X ), B ( X )]

OR-Union: A B max[ A ( X ), B ( X )]

NOT -Complement: A 1 A (X )

4.3 Defuzzification

The rules of fuzzy logic controller generate required output in a linguistic variable (Fuzzy Number),
according to real world requirements; linguistic variables have to be transformed to crisp output (Real
number). This selection of strategy is a compromise between accuracy and computational intensity.
566 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570

4.4 Database

The Database stores the definition of the triangular membership function required by fuzzifier and
defuzzifier. The determination of the membership functions depends on the designer experiences and
expert knowledge.

4.5 Rule Base

The Rule base stores the linguistic control rules required by rule evaluator (decision making logic The
formulation of its rule set plays a key role in improving the system performance [18-19-20].

5. DC Capacitor Voltage Control

Among the various available powers filter controllers PI, PID, RST hysteresis and fuzzy logic
controller. In this application, the fuzzy control algorithm is implemented to optimize the energy storage of
the DC capacitor voltage based on DC voltage error E(t) processing and its variation E(t) in order to
improve the dynamic performance of APF and reduce the total harmonic source current distortion [4].
Fuzzy logic uses linguistic variables instead of numerical variables. In a control system, error signal E, its
variation E and output signal D E can be assigned as negative Large: (NL); negative medium :( NM);
negative small :( NS); zero: (ZE); positive small: (PS); positive medium: (PM) and positive Large: (PL)
The triangular membership function is used for fuzzifications. The process of fuzzification convert
numerical variable (real number) to a linguistic variable (fuzzy number).

Table 1 .fuzzy control rule

E NL NM NS ZE PS PM PL
E
NL NL NL NL NL NM NS ZE
NM NL NL NL NM NS ZE PS
NS NL NL NM NS ZE PS PM
ZE NL NM NS ZE PS PM PL
PS NM NS ZE PS PM PL PL
PM NS ZE PS PM PL PL PL
PL NL NM NS ZE PS PM PL

Membership functions used for the inputs and output variables used here are shown in figure 5.

Fig.5.Membership functions for the inputs and output variable

 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570 567

6. Hysteresis Current Controller

The hysteresis band is used to control load currents and determine switching signals for inverters gates.
Suitable stability, fast response, high accuracy, simple operation, inherent current peak limitation and load
parameters variation independency make the current control methods of voltage source inverters.

Fig.7. Principle of hysteresis current control

In this approach the current error (difference between the reference current, and the current being injected
by the inverter) e(t)= Iref(t )- iInj(t) . When the error current exceeds the upper limit of the hysteresis
band, the upper switch of the inverter arm is turned OFF and the lower switch is turned ON. As a result, the
current start to decay that is shown in Fig 8. When the error current crosses the lower limit of the hysteresis
band (HB), the lower switch of the inverter arm is turned OFF and the upper switch is turned ON [18-23].
As a result, the current gets back into the hysteresis band. The switching performance as follows

0 if iInj(t) > Iref(t) + HB

S=
1 if iInj(t) < Iref(t) - HB

Fig.8. Diagram of hysteresis current control

7. Results and Discussions

The analysis of the three-phase system given in figure 1 has been done in SIMULINK/ MATLAB
environment.
The system parameters values are; Source voltage are Va=0.8*V, Vb= V, Vc=1.1*V with V=230 V;
frequency f= 50 Hz; source impedance Rs = 0.5 m , Ls = 15 H; filter impedance Rf=5 m , Lf=180 H;
DC voltage capacitor Vdc_ref=800 V,(Cdc=4.4 mF); nonlinear load Rch = 0.75 , LL = 55 H.
568 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570

7.1 Simulation results

The simulation results show a good filtering of harmonic currents and a perfect compensation of
reactive power.
Figure 9 shows the simulation results obtained for the mains voltage (V a-b-c) and the direct component of
the voltage (V* a-b-c), this figure confirms the accuracy of the extraction of direct component by using
PLL. Figure 10 shows the source current waveform deformed before filtering. The shunt APF controlled by
fuzzy logic controller is injected current (if) as shown in figure 12. The active filter has imposed a
sinusoidal source current waveform instantaneously as illustrated in figure 13. Figure 14 shows the
simulation results of the dc-side capacitor voltage which is nearly constant with small ripple. The obtained
current and voltage waveforms are in phase as illustrated in figure 15. The current THD is reduced from
20.18% to 1.89% as shown in the frequency current spectrum (figure 16).

(a)

(b)
Fig.9. Temporal analysis of the PLL with a voltage unbalanced power system (a) Vabc (V), (b) V*abc (V)

Fig.10. Source current iSa (A) waveform before filtering

Fig.11.Reference current Iref1 (A)

 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570 569

Fig.12. Injected current ifa (A)

Fig.13. Source current iSa (A) waveform after filtering

Fig.14. DC voltage control

Fig.15. Power factor correction (Vsa,isa)

Fig.16. (a).Source current spectrum without filter (b) Source current spectrum with filter.

8. Conclusions

In this paper a new technique with instantaneous power theory (p-q theory) based on PLL (Phase
Locked Loop) is used in order to control APF under unbalanced mains voltage.
570 R. Belaidi et al. / Energy Procedia 18 (2012) 560 – 570

Also, a fuzzy logic control of shunt APF based on this technique is proposed to identify reference currents
the proposed system show excellent shunt APF performances. These performances are related to the
current references quality. This method is very important because it allows harmonic currents and reactive
power compensation simultaneously. The obtained results show that the dc-side capacitor voltage is nearly
constant with small ripple and the source current waveform purely sinusoidal after filtering .Also, the
results show that the current obtained after filtering and the voltage waveforms are in phase. The current
THD is reduced from 20.18% to 1.89% which confirms the good filtering quality of harmonic currents and
a perfect compensation of reactive power which improve the power quality.

References
[1] H. Akagi, “New Trends in Active Filters for Power Conditioning,” IEEE Trans. on Industry Applications, vol. 32, no. 6, pp.
1312-1322, 1996.
[2] W. E. Kazibwe and M. H. Sendaula Electric Power Quality Control Techniques. Van Nostrand Reinhold, 1993, New York,
USA.
[3] R. C. Dugan, M. F. McGranaghan, S. Santoso and H. W. Beaty. Electrical Power Systems Quality 2nd. ed. McGraw-Hill, 2002,
USA.
[4] S. Saad, L. Zellouma “Fuzzy logic controller for three level shunt active filter compensating harmonics and reactive power”
Electric Power Systems Research, Elsevier, May-2009 pp.1337–1341
[5] Z Salam, Tan Perng Cheng and Awang Jusoh, “Harmonics Mitigation using Active Power Filter : A Technological Review”
Elekrika, Vol.8, No.2, 2006, pp.17-26.
[6] S. Buso, L. Malesani, P. Mattavelli, “ Comparison of current control Techniques for Active power Filter Applications”, IEEE
Transactions on Industrial Electronics, Vol.45, no.5, Oct 1998, pp.722-729.
[7] S.K.Jain, P.Agrawal and H.O.Gupta,”Fuzzy Logic controlled shunt active power filter for power quality improvement”,IEE
proceedings in Electrical Power Applications, Vol 149, No.5, September 2002.
[8]M. Kale, E.Ozdemir “Harmonic and reactive power compensation with shunt active power filter under non-ideal mains voltage”
Electric Power Systems Research Elsevier 74 (2005), pp. 363–370.
[9] Akagi,H., Kanazawa,Y., and Nabae,A., “Instantaneous reactive power compensators comprising switching devices without
energy storage components. IEEE Transactions on Industrial Applications, 1984, Vol.20, pp.625- 630
[10] Singh, B., Haddad K., Chandra, A., “A New Control Approach to Three-Phase Active Filter for Harmonics and Reactive
Power Compensation,IEEE Trans. on Power Systems, 1998, Vol.13, No.1, pp. 133-138.
[11] A.Ametani et all “Harmonic reduction in thyristor Converters by harmonic current injection,” IEEE Trans8.Power,
Appar.Syst.1976, 95, pp.441-449.
[12] M. Rukonuzzaman and M. Nakaoka, “An advanced active power filter with adaptive neural network based harmonic detection
scheme,” IEEE power Electronics Specialist cascade, PESC, Vancouver Canada, 2001, pp. 1602-1607.
[13] M. C. Benhabib and S. Saadate, “New Control approach for four wire active power filter based on the use of synchronous
reference frame,” Elsevier Electric power systems Research 73, 2005, pp. 353-362.
[14]H. Akagi, Y. Kanazawa and A. Nabae, "Generalized Theory of Instantaneous Reactive Power and Its Applications,"
Transactions of the lEE-Japan, Part B, vol. 103, no.7, 1983, pp. 483-490
[15] Alali A.M,"Contribution à l’Etude des Compensateurs Actifs des Réseaux Electriques Basse Tension», Thèse de doctorat de
l’Université Louis Pasteur – Strasbourg I, Septembre 2002.
[16] C.C. Lee, Fuzzy logic in control systems: fuzzy logic controller-part II, IEEE Trans. Syst. Man Cybern. 1990.
[17] S. Tesnjak, S. Mikus, O. Kuljaca, Load–frequency fuzzy control in power systems, in: Proceedings of the Fifth SONT,
Simpozijo Novim Tehnologijima, Poree, 1995, pp. 136–139.
[18] Karuppanan P and KamalaKanta Mahapatra “PLL with PI, PID and Fuzzy Logic Controllers based Shunt Active Power Line
Conditioners” IEEE PEDES- International Conference on Power Electronics, Drives and Energy Systems-Dec 21 o 23, 2010 at IIT-
Delhi.
[19] Karuppanan P and KamalaKanta Mahapatra “Fuzzy Logic Controlled Active Power Line Conditioners for Power quality
Improvements” International Conference on Advances in Energy Conversion Technologies (ICAECT2010), Jan- 2010 pp.177-181.
[20] Karuppanan P, Kamala Kanta Mahapatra “PI with Fuzzy Logic Controller based APLC for compensating harmonic and
reactive power” Proc. of Int. Conf. on Control, Communication and Power Engineering 2010 pp.45-49