# Wavelet based reference current calculation method for active compensation systems/ Vilneliu saltinio sroves apskaiciavimo metodo taikymas aktyviosioms kompensavimo sistemoms.

Introduction

Active compensation and filtering systems have been becoming more important day by day for ensuring high quality in power generation and reducing its cost. Passive compensation and filter circuits have been replaced by SCR-based circuits as a result of development in semi- conductor technology. Determining reference current for these circuits is quite important for filter and compensation circuits [1-12].

Reference current can be defined as total of all instantaneous active and reactive currents drawn due to unbalanced loads or harmonic distortions. An active compensation circuit is run in a way that it generates this reference current. Most frequently preferred methods for obtain reference current are p-q theory, Peng theory, Park Power theory etc. PWM technique is generally preferred in compensator current control [1-12]. p-q theory, which is most prevalent calculation method, was took reference in this study.

p-q Theory

p-q theory is based on finding power values by converting voltage and current from a, b, c phase plane into [alpha], [beta], 0 plane [7,13,14].

Voltage and current vectors are expressed as the following to show three-phase instantaneous voltage values [v.sub.a], [v.sub.b], [v.sub.c] and three-phase instantaneous current values [i.sub.a], [i.sub.b], [i.sub.c]

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (1)

This voltage's and current's a, p and 0 values are expressed as (2-3).

Instantaneous active power is found by scalar multiplication (*) of voltage and current vectors while reactive power is found by vector multiplication of them (x) (4).

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (2)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (3)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (4)

Instantaneous active and reactive power can be expressed as the following for the systems without 0 component

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (5)

In this equation, [bar.p] stands for mean instantaneous active power while [??] represents alternating active power, [bar.q] represents mean instantaneous reactive power and [??] represents alternating reactive power. All powers must be compensated except [bar.p] [1-14]. Fundamental current drawn from the source ([[??].sub.f]) and shunt-compensator's reference current are calculated by using [bar.p] mean instantaneous active power

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (6)

Discrete Wavelet Packet Transformation

Wavelet transformation is one of the most recently developed methods for signal analysis. It is prevalently used in many areas like mathematics, physics and engineering .

Wavelet packet transformation is an expression of digital signal with time scale. The signal resolution is changed by filtering process while its scale is replaced by down sampling process. Output of cascaded filter banks depends on the main wavelet and high- and low-frequency components of the system. The procedure is started by passing a N-length discrete signal s(n) through an impulse response high-pass filter h(n) and an impulse response low-pass filter g (n) . Responses of high-pass and low-pass filters constitute first-stage decomposition of the discrete signal and expressed as the following [17, 18]

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (7)

Extension functions ([[phi].sub.j,j]k](x) and [[psi].sub.jk](x)) are not seen in this equation. This allows simulations to be performed through digital filter banks on computer environment [17, 18]. [s.sup.1.sub.0](n) and [s.sup.1.sub.1](n) sequences are more decomposed at second stage of wavelet packet transformation:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (8)

The entire wavelet packet decomposition can be produced by following similar procedure;

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (9)

The decompositions from the sequence [s.sup.i.sub.n] (n) at stage i into the sequences [s.sup.i+1.sub.m0](n) and [s.sup.i+1.sub.m1] at stage i +1 using the two decomposition quadrature filters are given by [17, 18]:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (10)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (11)

Proposed Method

The proposed method is, first of all, based on calculation of instantaneous active and reactive powers according to Peng's generalized reactive power theory with no need for p-q transformation. Instantaneous active power is calculated as the following [6, 7]

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (12)

Instantaneous active power is a scalar magnitude according to the equation above. Instantaneous reactive power is calculated as the following [6,7]

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (13)

Reactive power is a vector magnitude and [absolute value of [??]] represents magnitude of instantaneous reactive power for three phases [6,7]. DWPT decompositions of instantaneous active power at third stage can be written as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (14)

DWPT decompositions of instantaneous reactive power for each phase at third stage can be written as;

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (15)

where

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (16)

where instantaneous active currents drawn for each decomposition of instantaneous active power:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]

Sum of these currents yields value of the drawn instantaneous active current

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (17)

where instantaneous reactive currents drawn for each decomposition of instantaneous reactive power:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]

Sum of these currents yields value of the drawn instantaneous reactive current

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (18)

Frequency ranges of current decompositions for sampling rate of 0.000625 seconds are given in Table 1 .

As seen in the table, currents are decomposed into 100 Hz equal frequency ranges. Therefore, currents of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] are active and reactive currents drawn at fundamental frequency of 50 Hz. These currents cause [bar.p] ve [bar.q] powers from which the load requires from the source and consequently, all currents must be compensated except [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] [19, 20, 21]. If is is the source current, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] is the load current and iC is the compensation current then

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (19)

Because the load current is equal to sum of active and reactive currents then

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (20)

Because we aim that only the basic frequency [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] active current is drawn from the source, all other current components account for shunt compensator's reference current ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]). Thus,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (21)

Working principle of DWPT-based reactive power and harmonic compensator is given in Fig. 1.

[FIGURE 1 OMITTED]

Simulation results

Calculations and simulations were carried out according to p-q theory, as first, and then, according to the proposed method. Then, the results were compared. db20 was used in all DWPT simulations.

Phase-neutral effective value of source voltage is 2400 V and frequency is 50 Hz. Fig. 2 shows configuration of the compensation system. It was assumed that the compensator can track the reference current at instantaneously at infinite speed. Fig. 3 shows the source voltage and non-linear load current.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

p and q powers produced through (1-5) based on p-q theory are given in Fig. 4.

[FIGURE 4 OMITTED]

To find [bar.p], p signal was made pass through a low-pass filter, whose cut-off frequency was 30 Hz. Fig. 5 shows fundamental frequency instantaneous active current, ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]) and compensator's reference current ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]) obtained by using (6).

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

Fig. 8 and 9 show DWPT decompositions of instantaneous active and reactive currents found by using (17-18) according to generalized reactive power theory.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

Fig. 10 shows compensator's reference current [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] values calculated according to the proposed method considering [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.].

[FIGURE 10 OMITTED]

Fig. 11 shows the compensation results obtained through both of the methods

[FIGURE 11 OMITTED]

The biggest difference value in of both compensation currents is around [+ or -]7 A (Fig. 12).

[FIGURE 12 OMITTED]

Conclusions

In this study, a DWPT-based method was proposed for calculating reference current for active compensation systems. Unlike classical methods, mean instantaneous active power [(bar.p]) was obtained directly through DWPT and moreover, instantaneous active power signal (p) does not require to be made pass through low-pass filter. Fundamental frequency instantaneous active current ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]), reference current ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]) and other currents can be calculated easily by using proposed method. In comparison of the obtained compensation current with that, which is obtained according to p-q theory, maximum variation in current is [+ or -] 7 A. This accounts for very small and negligible error as 0. 43. according to maximum compensation current, [+ or] 1600 A. Thus, the proposed method can be used in active harmonic and compensation systems.

References

[1.] Bayindir K. C., Ermis. M. Understanding the Modelling and Analysis of a Shunt Active Power Filter Using MATLAB/Simulink // International Journal of Electrical Engineering Education, 2006.--ol. 43 (3).--P. 185-205.

[2.] George M., Basu K. P. Three-Phase Shunt Active Power Filter // American Journal of Applied Sciences, 2008.--Vol. 5(8).--P. 909-916.

[3.] Syed M. K., Ram B. S. Instantaneous Power Theory Based Active Power Filter: a Matlab/ Simulink Approach // Journal of Theoretical and Applied Information Technology, 2008.--Vol. 4(6). --P. 536-541.

[4.] George M., Basu K. P. Modeling and Control of Three-Phase Shunt Active Power Filter // American Journal of Applied Sciences, 2008.--Vol. 5(8).--P. 1064-1070.

[5.] Afonso J. L., Freitas M. J. S., Martins J. S. p-q Theory Power Components Calculations // ISIE'2003--IEEE International Symposium on Industrial Electronics.--Rio de Janeiro, Brasil, 2003,--Vol. 1.-P. 385-390.

[6.] Peng F. Z., Lai, J. S. Reactive Power and Harmonic Compensation Based on Generalized Instantaneous Reactive Power Theory for Three-Phase Power Systems // 7th international conference on harmonic and quality of power.--Las Vegas, United States, 1996.--P. 1-7.

[7.] Tolbert L. M., Habetler T. M. Comparison of Time Based Non-Active Power Definitions for Active Filtering // Power Electronics Congress, VII IEEE International, 2000.--P. 73-79.

[8.] Xu L., Han Y., Pan J., Chen C., Yao G., Zhou L.D. Selective Compensation Strategies for the 3-Phase Cascaded Multilevel Active Power Filter using ANF-based Sequence Decoupling Scheme // Electronics and Electrical Engineering. --Kaunas: Technologija, 2010.--No. 2(98).--P. 15-20.

[9.] Han Y., Xu L., Yao G., Zhou L. D., Mansoor, Chen C. Operation Principles and Control Strategies of Cascaded H-bridge Multilevel Active Power Filter // Electronics and Electrical Engineering.--Kaunas: Technologija, 2009.--No. 3(91).--P. 71-76.

[10.] Han Y., Xu L., Yao G., Zhou L.D., Khan MM., Chen C. The Adaptive Signal Processing Scheme for Power Quality Conditioning Applications Based on Active Noise Control (ANC) // Electronics and Electrical Engineering.--Kaunas: Technologija, 2009.--No. 8(96).--P. 9-14.

[11.] Benslimane T. Open Switch Faults Detection and Localization Algorithm for Three Phases Shunt Active Power Filter based on Two Level Voltage Source Inverter // Electronics and Electrical Engineering.--Kaunas: Technologija, 2007.--No. 2(74).--P. 21-24.

[12.] Driesen J., Belmans R. Active Power Filter Control Algorithms using Wavelet-Based Power Definitions // 10th International Conference on Harmonics and Quality of Power (ICHQP).--Rio de Janeiro, Brazil, 2002.--Vol. 2.--P. 466-471.

[13.] Pou F., J. Modulation and Control of Three-Phase PWM Multilevel Converters // Technical University of Catalonia, PhD. Thesis.--Barcelona, 2002.--P. 193-200.

[14.] Tahri A., Draou A. Design of a Simple Measuring Technique of The Instantaneous Power in Three Phase System // Journal of Electrical Engineering, 2005.--Vol. 56. --No.7-8.--P. 221-224.

[15.] Janicek F., Mucha M., Ostrozlik M. A New Protection Relay Based on a Fault Transient Analysis Using Wavelet Transform // Journal of Electrical Engineering, 2007.--Vol. 58.--No. 5.--P. 271-278.

[16.] Khan M. A. S. K., Radwan T. S., Rahman M. A. Wavelet Packet Transform Based Protection of Disturbances Three-Phase Interior Permanent Magnet Motor Fed From Sinusoidal PWM Voltage Source Inverter // Electrical and Computer Engineering, CCECE'06.--Canadian Conference, 2006.--P. 178-181.

[17.] Deng H., Ling H. Fast Solution of Electromagnetic Integral Equations Using Adaptive Wavelet Packet Transform // IEEE Transaction on Antennas and Propagation, 1999.--Vol. 47.--P. 674-682.

[18.] Onat N., Kiyak 1, Gokmen G. Experimental Wavelet Packet Transient-State Analysis of Electrical Machines Directly Fed by Photovoltaic Cells // International Review of Electrical Engineering (I.R.E.E.), 2010.--Vol. 5.--No. 5.--Part 1.--P. 2081 -2087.

[19.] Barros J., Diego R. J. Analysis of Harmonics in Power Systems Using the Wavelet Packet Transform // Proceedings of the IEEE Instrumentation and Measurement Technology Conference, IMTC 2005.--Vol. 2.--P. 1484-1488.

[20.] Gligor A. Design and Simulation of a Shunt Active Filter in Application for Control of Harmonic Levels // Electrical and Mechanical Engineering, 2009.--Vol. 1.--P. 53-63.

[21.] Couto E. F., Martins J. S., Afonso J. L. Simulation Result of a Shunt Active Power Filter with Control Based on p-q Theory // International Conference on Renewable Energies and Power Quality, ICREPQ'03. - Vigo, 2003. - P. 1-6.

G. Gokmen Department of Electrical Education, Technical Education Faculty, Marmara University, Goztepe Campus, 34722 Kadikoy, Istanbul, Turkey, phone: +90 216 336 5 7 70 (265), e-mail:gokhang@marmara.edu.tr
```Table 1. Wavelet packet decompositions of currents and their
frequency ranges

Decomposition                             Frequency    Harmonic
Range (Hz)     Order

[i.sup.3.sub.p000], [i.sup.3.sub.p000]     700~800        15
[i.sup.3.sub.p001], [i.sup.3.sub.p001]     600~700        13
[i.sup.3.sub.p010], [i.sup.3.sub.p010]     500~600        11
[i.sup.3.sub.p011], [i.sup.3.sub.p011]     400~500         9
[i.sup.3.sub.p100], [i.sup.3.sub.p100]     300~400         7
[i.sup.3.sub.p101], [i.sup.3.sub.p101]     200~300         5
[i.sup.3.sub.p110], [i.sup.3.sub.p110]     100~200         3
[i.sup.3.sub.p111], [i.sup.3.sub.p111]      0~100          1
```
COPYRIGHT 2011 Kaunas University of Technology, Faculty of Telecommunications and Electronics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Title Annotation: Printer friendly Cite/link Email Feedback ELECTRICAL ENGINEERING/ELEKTROS INZINERIJA Gokmen, G. Elektronika ir Elektrotechnika Report 7TURK Feb 1, 2011 2288 Generalized mathematical model of controlled linear oscillating mechatronic device/Apibendrintasis valdomo tiesiaeigio svytuojamojo mechatroninio... Mitigation of magnetizing inrush current using sequential phase energization technique/ Imagnetinimo sroves sumazejimo tyrimas taikant nuosekluji... Electric current Electric currents Electric equipment Electrical equipment and supplies Electrical machinery Equipment performance Wavelet transforms