Scribd
Upload
22 views

02. a Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics From Distorted Mains

The document presents a hybrid active power filter (HAPF) controlled by artificial neural networks for harmonic correction in power systems. It discusses the limitations of traditional passive filters and highlights the advantages of using active power filters for improved power quality. The proposed system was validated through MATLAB/SIMULINK simulations, demonstrating its effectiveness in reducing harmonics and enhancing power quality in various load conditions.

Uploaded by

editor.jets
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
22 views

02. a Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics From Distorted Mains

The document presents a hybrid active power filter (HAPF) controlled by artificial neural networks for harmonic correction in power systems. It discusses the limitations of traditional passive filters and highlights the advantages of using active power filters for improved power quality. The proposed system was validated through MATLAB/SIMULINK simulations, demonstrating its effectiveness in reducing harmonics and enhancing power quality in various load conditions.

Uploaded by

editor.jets
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Journal of Engineering Technology and Sciences

Volume 02, Issue 03, March 2025 ISSN: 3048-913X


www.jets.org.in DOI: https://doi.org/10.5281/zenodo.14976374

A Hybrid Active Power Filter Using Artificial Neural Networks to


Remove Harmonics from Distorted Mains
Srikanth
Department of Electrical and Electronics Engineering, VR Siddhartha, Vijayawada
Corresponding Author: [email protected]
To Cite this Article
Srikanth, “A Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics from Distorted
Mains”, Journal of Engineering Technology and Sciences, Vol. 02, Issue 03, March 2025, pp: 04-07
_____________________________________________________________________________________________
Abstract: A description of the dynamic neural network-controlled hybrid active power filter (HAPF) is provided for
harmonic correction during source voltage and load current fluctuations. A neural network-based controller operates
to extract current essentials from non-sinusoidal and unbalanced currents of the examined supply system. The
presented compensation system merges shunt active and shunt passive filter mechanisms to achieve superior
performance. The analysis of a three-phase hybrid active power filter occurs within MATLAB/SIMULINK both for
balanced and unbalanced loads and unbalanced and distorted sources. The methodology validation took place
through the complete simulation level findings presentation.

Keywords: Artificial Neural Network, THD, PI controller, hybrid active power filter
_____________________________________________________________________________________
I. Introduction
The use of power electronics-based nonlinear loads has caused a continuous deterioration of power quality
during the last few years. The international standards define power quality standards through the maintenance of pure
sinusoidal waves between the current and voltage waveforms that stay perfectly sinusoidal. Power quality issues
emerge whenever sinusoidal waveform distortions occur in voltage or current or any frequency deviation occurs.
Commercial and industrial applications and utility interfaces drives frequent occurrences of power quality problems.
Such appliances generate both harmonics along with reactive power.
The IEEE 519 harmonic standard emphasizes dominant harmonic correction which requires THD levels
below 5% [1]. The reduction of harmonic distortion receives support from active and passive filters as two separate
filtering solutions. The basic operational approach to troubleshoot harmonic distortion and enhance power factor at
power system utilities uses passive filtering. The passive filter presents multiple fundamental limitations because its
size is cumbersome and it faces tuning problems related to tolerance while needing fixed L & C values for specific
compensation settings. Furthermore, its construction remains simple but expensive. Active Power Filters (APF) serve
as the solution to resolve this problem.

Fig 1: Hybrid filter configuration


Published by: Surya Publishers www.jets.org.in 4 | Page
A Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics from Distorted Mains

Within power systems active power filters represent an excellent harmonic current reduction solution due to
their small dimensions together with automatic operation and no need for tuning adjustments. A harmonic current
source functions through active power filters for successful compensation of reactive power and harmonic currents. It
generates harmonic current with equal magnitude while being out of phase with the load to enter the ac system [2].
The hybrid active power filter dominates harmonics solutions in figure 1 because its complex system and economical
parameter management makes it the preferred selection.

II. Proposed Topology


A prototype design of three-phase three-wire hybrid power filter appears in Figure 2. The shunt active power
filter generates compensatory current that operates in an opposite phase. A power circuit of IGBT-based three-phase
voltage source inverter with DC storage capacitor has been proposed for APF applications to enhance non-linear and
balanced and unbalanced load compensation. Reference current generation through the neutral network controller
works together with DC voltage management implemented by the PI controller in active power filter control methods.
An artificial neural network uses three adaptive linear neurons to extract necessary elements from three phase
voltages present in non-sinusoidal supplies. Active capacitors serve to maintain the dc voltage ripple between 1 and
4% of its predetermined range. Such capacitors should possess strength to endure maximum voltage requirements.
The active filter operates across different modes to provide compensation current for eliminating or reducing harmonic
currents in order to improve power quality. The power quality receives enhancement when the compensatory current
is injected. The precision of active power filters depends heavily on the method used for reference current computation.

Fig 2: Shunt active power filter proposed topology

III. Control Strategy


The control scheme performs multiple functions as shown in Figure 3 where it keeps up the supply current's
sinusoidal waveform then detects harmonics before controlling the DC voltage and supplying compensating power to
the system while also giving out harmonic currents to the three-phase non-linear load. Operation of the HSAPF control
scheme becomes necessary. The correct response of APF depends on extracting current fundamental components from
non-sinusoidal input signals together with reference current generation and DC voltage regulation and the supply of
compensatory currents [10]. Different regulating strategies are necessary to complete these duties successfully.

5 | Page
A Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics from Distorted Mains

Fig 3: Control strategy for shunt active power filter schematic diagram

IV. Neural Network Controller


The problem of deriving vital information from transformed current or voltage waveforms represents a
critical assessment method that determines power quality performance. Producing high-quality power supply
electricity depends on precise estimation or extraction of both fundamental time-varying components and their phase
angle magnitudes. Active power filters serve to decrease harmonic elements in power systems. The methodology
depends on an ANN controller to realize harmonic estimation. The authors choose ANN due to its demonstrated
superiority thus implementing this model for their research. The previously proposed actual conditions make
traditional fundamental component separation through low pass filters non-effective.

Fig 4: Internal blocks of proposed Neural Network

V. Conclusion
The estimation of the size and cost reduction of the active filtering unit with the addition of passive tuned
filters for partial compensation has been done in this article. Variable VAR compensation of the passive filters has
been estimated. A new strategy based on clever brain processes has been put forth. MATLAB simulation simulations
were used to confirm the suggested ANN's performance. The suggested controller provides precise supply. The
estimation confirms that the ANN control scheme-based shunt active filtering system may be made to be very cost-
effective in real-world applications for improving power quality in retrofit systems.

6 | Page
A Hybrid Active Power Filter Using Artificial Neural Networks to Remove Harmonics from Distorted Mains

References
[1] Roger C.Dugan, Mark F. McGranaghan, Surya Santoso and H.Wayne Beaty, “Electrical power system quality”, Mcgraw-Hill.
[2] H. Akagi, “Trends in active power line conditioners,” IEEE Trans. Power Electronics, vol. 9, no. 3, pp. 263 May 1994.
[3] K. Sangsun, and P. N. Enjeti, "A new hybrid active power filter (APF) topology", IEEE Transactions on Power Electronics, vol.17, no. 1, pp.
48
[4] S. Rahmani, K. Al-Haddad and F. Fnaiech, “A Series Voltages”, IEEE Industrial Electronics Conference IECON 2002, 649.
[5] F. Z. Peng, "Application issues of activ 30, 1998.

7 | Page

You might also like