Synchronous Reference Frame Based Control of MLI ... - IEEE Xplore

Synchronous Reference Frame Based Control of MLI-STATCOM in Power Distribution Network Venkata Reddy Kota

Sudheer Vinnakoti

Department of Electrical and Electronics Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, Andhra Pradesh, India. E-mail: [email protected]

Department of Electrical and Electronics Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, Andhra Pradesh, India. E-mail: [email protected]

Abstract: Ever increasing nonlinear loads and supplies are changing dynamics of the electrical network. The levels of harmonics are on the rise and polluting more and more of the electrical network, posing new challenges before the designers and operators. To maintain the present power quality levels and to improve the performance of the electrical network, many control schemes and devices like Active Power Filter’s have been proposed, but still an improvement is required. This paper proposes a five-level diode clamped Multilevel Inverter based Distribution STATCOM (MLI-DSTATCOM) with Synchronous Reference Frame based control for harmonic mitigation. A threephase four wire system with nonlinear, balanced/unbalanced load is designed and simulated in Matlab/Simulink for performance analysis of proposed MLI-DSTATCOM. The results provide the evidence of restricting harmonic currents at load end and prevent from entering into source effectively. The performance of five-level, three-level and two-level DSTATCOM’s are compared under nonlinear, balanced/unbalanced load. Simulation result analysis show superior performance of five-level diode clamped MLI-DSTATCOM controlled using SRF. Keywords—Multilevel Inverter based Distributed STATCOM (MLI-DSTATCOM); Synchronous Reference Frame (SRF); Power Quality (PQ); Harmonic Mitigation; Active Power Filter (APF); Total Harmonic Distortion (THD).

I.

INTRODUCTION

The revolution in the digital world and power electronics has a considerable impact on the electricity network. The increasing levels of nonlinear loads and power supplies are polluting electric network more and more with days passing. Low Power factor leads to reduction in energy efficiency and power handling capacity, more losses in electrical machines and appliances, torque pulsations in motor and also creating perilous disturbances to connected devices in electric network [1]-[2]. In distribution system many techniques and equipment’s are used to mitigate power quality related issues. The most common method employed is a passive filter which suppresses harmonics thereby improving power factor and consequently reducing losses in the network and appliances, but the drawback is bulky size, resonance and limited compensation characteristics [3]. Custom Powered Devices (CPD) later introduced gave improved performance over the passive filters and are very useful for maintaining the present

978-1-5090-0261-0/16/$31.00 ©2016 IEEE

power quality levels [4]-[5]. To suppress harmonics and other power quality issues related to current, Distribution Static Compensator (DSTATCOM) connected across load is widely researched and proved to be effective. Other FACTS devices also have been researched widely to improve power quality, flexibility, controllability and stability. Reactive power control or compensation is of importance for stability, control and quality in power system. DSTATCOM also have a significant role to play for reactive power control in distribution network [6] - [8]. Performance of any CPDs is greatly depending on the gating pulses and the control algorithm to generate estimated reference currents. Few control algorithms mostly employed are feed forward training [6], SRF theory and instantaneous active and reactive power theory [9] –[12], Lyapunov-function control [13] and the non-linear control technique [14] etc. Performance of MLI-DSTATCOM is analyzed in this paper with nonlinear, balanced/unbalanced load. With the proposed control method, load currents, source currents and source voltages are measured. Total Harmonic Distortion (THD) of supply currents with conventional two-level DSTATCOM, three-level diode clamped and five-level diode clamped MLI-DSTATCOM is developed and analyzed in Matlab/Simulink software. This study has been expanded to active and reactive power flow analysis. This paper is arranged as following: Section II explains DSTATCOM operation. Section III explains SRF control for DSTATCOM. Section IV discusses observations made from results. Section V concludes work. II.

OPERATION OF DSTATCOM

Distribution Static Compensator (DSTATCOM) is connected in parallel like a STATCOM, which is connected at transmission level where as DSTATCOM at distribution level. Its main function is to inject harmonically distorted current in phase opposition to the load current thereby supressing harmonics in the supply current [7], in addition to this it also supplies required reactive power to the load. A typical block diagram of a DSTATCOM with unbalanced and nonlinear load is depicted in Fig. 1. DSTATCOM equivalent circuit is given in Fig. 2, where V Sabc is phase voltage of supply; V SHabc is a fundamental

component of phase output voltage of DSTATCOM and I SHabc is a fundamental component of shunt APF current. The operation of DSTATCOM is explained in the following modes.

regulating the magnitude of the V SHabc and thus DC Link capacitor voltage. III.

SYNCHRONOUS REFERENCE FRAME CONTROL

SRF control is one of the efficient controls to suppress voltage and current harmonics. It referred d-q technique, in which transformations and its inverse transformations of a-b-c to d-q-0 are used [15]. The basic SRF Control technique to generate reference currents from nonlinear balanced /unbalanced load is depicted in Fig. 4. The load currents of ab-c coordinates ( I Labc ) are transformed into d-q-0 coordinates with the help of modified PLL according to the equation (1). These d-q-0 coordinates comprises of an oscillatory ~ ~ component ( I oSd and I oSq ) and averaged component ( I ASd

Fig. 1 Block diagram of DSTATCOM.

and I ASq ) resulting to oscillatory in nature. In order to avoid oscillatory response and maintain only averaged components of d-q-0 coordinates, a 2nd ordered Butterworth LPF is used. These averaged components are stable in nature and are referred to as source current averaged component ( I SdL ). Vdc* Discrete Filter 1200

Fig. 2 Equivalent circuit of DSTATCOM

Mode 1: If V SHabc is in-phase and equal to V Sabc then DSTATCOM does not inject and absorb any reactive power Mode 2: If V SHabc is greater than V Sabc and I SHabc leads V Sabc by some angle, then DSTATCOM will supply the reactive power of the system. This mode can also be called as capacitive mode and its phasor representation is given in Fig. 3(a). Here DSTATCOM supply all required load reactive power thereby making supply currents free from reactive currents. Mode 3: If I SHabc lags V Sabc then DSTATCOM will take reactive power from the system. This mode can also be called as inductive mode and its phasor representation is shown in Fig. 3(b).

PI

0

Vdc

0 ILabc

abc sin_cos

vsab vscb

vsab Sin_Cos vscb

dq0_to_abc Transformation2

d

d

Add2

Discrete Filter

0.55 sin_cos

a

ISaref

b

ISbref

C

IScref

Fig. 4 Shunt Controller using SRF

I S0 I Sd I Sq

1

1 2 2 = 2 sin(wt) sin wt − 2π 3 3 cos(wt) cos wt − 2π 3

( (

) )

1

2 sin wt + 2π 3 cos wt + 2π 3

( (

) )

I Loa I Lob I Loc

(1)

For proper compensation, voltage of DC link capacitor must be kept constant at rated value (i.e. 1200V in this case). The PI controller is therefore used to compensate the loss component of active current ( I Dloss ). Using Ziegler-Nichols’

(a)

(b)

Fig. 3 Phasor representation of DSTATCOM

From the phasor diagrams it is clear that, DSTATCOM either generates or absorbs the reactive power by controlling the phase angle (α) between V SHabc and V Sabc thereby

method, proportional gain (Kp) and Integral gain (Ki) are estimated and are fine tuned to values of 0.003 and 0.0025. The d-axis component of supply current including the active power loss component for capacitor voltage balancing can be represented by ∗ I Sd = I Dloss

+ I

SdL

(2)

Considering these currents as d-axis component, d-q-0 coordinates are transformed into a-b-c coordinates with reference to equation (3) and are taken as reference supply currents.

1 sin(wt ) I refSa 2 1 sin wt − 2π I refSb = 2 3 3 2 I refSc 1 sin wt + 2π 3 2

( (

) )

cos(wt )

( ) 3 2 π ) cos(wt + 3 cos wt − 2π

IV.

(3)

IS0 I *Sd I Sq

SIMULATION RESULTS AND ANALYSIS

The proposed five level diode clamped MLI-DSTATCOM configuration with SRF control for a three phase four wire system is developed and analyzed in Matlab/Simulink model and is given in Fig. 6. System parameters for this study are specified in Table 1. The proposed system is analyzed in two different cases, i.e., case 1 is a three phase system with nonlinear balanced Load and case 2 is a three phase system with nonlinear unbalanced Load. In these two cases five level diode clamped MLI-DSTATCOM is connected to the system at 0.5 Sec. Load currents, supply currents and supply voltages without and with compensation can be seen in Fig. 7 for case 2. The supply voltage is shown in Fig. 7(a). The impact of nonlinear and unbalanced load makes the load current to be nonlinear in nature and additional load on phase B draws more current than phase A and phase C, which can be clearly observed in Fig. 7(b).

These reference currents ( I refSa , I refSb and I refSc ) are compared with load currents ( I La , I Lb and I Lc ) to generate DSTATCOM reference currents i Shabc_ref . The currents of the DSTATCOM are maintained at reference values using Hysteresis current controller as shown in Fig. 5. The hysteresis current controller is operated with a lower band 0.25A and higher band of 0.5A to generate switching pulses to a fivelevel diode clamped MLI-DSTATCOM.

Table 1: proposed DSTATCOM System Parameters

Source

Load

Parameters Voltage (phase-phase) Frequency Three phase AC line impedance Single phase AC line Inductance

Value 415 Vrms 50 Hz R= 3Ω L=6mH L= 5e-3mH

R=30Ω and L=11.5mH R=100Ω and C=75μF C1=C2= 47μF Rshabc=0.5 Ω, Lshabc =32 mH 4.5KVA

Nonlinear load

Capacitor

Fig. 5: Hysteresis current control scheme for five level MLI-DSTATCOM

single phase nonlinear load Two series capacitors Shunt Filter Shunt APF

VSabc

c

A c aB

Phase A Phase B

b

Phase C

c

ISabc

C

aA

A

A

A

bB

B

B

B

cC

THREE PHASE SUPPLY

C

C

+

LOAD -

C

B

C

A

v+

c

b

Phase A

A

+

B

-

Cc

Aa

L

Bb

com

Universal Bridge

a

TIMER

C1

LOAD1

Universal Bridge1 Phase B

C2

and

V-

Phase C

MLI DSTATCOM

Fig. 6: DSTATCOM model of Three Phase four Wire System in Matlab/Simulink

1200 Voltage (Volts)

1000 800 600 400 200 0

Supply Voltage

400

0

0.5

1

1.5 Time (sec)

2

2.5

3

Fig. 8: DC Link Capacitor voltage

200 0 -200 -400 0.45

0.5

0.55 0.6 Time (sec)

0.65

0.7

(a)

Load Current

20

Further, this study is also extended to analyze active and reactive powers in the system. Before compensation the system is supplying an active power of 8.88 kW as required by the load and proposed MLI-DSTATCOM is observed to be zero, as it is isolated. The initial values of the active power at supply, load and MLI-DSTATCOM before compensation, i.e., during the period 0 Sec to 0.5 Sec can be observed from Fig. 9(a), Fig. 9(b) and Fig. 9(c) respectively. After compensation i.e., from 0.5 Sec the MLI-DSTATCOM connects to the system and it consumes 0.82 kW of active power from the supply to compensate the current harmonics, thus increases the active power from 8.88 kW to 9.7 kW at the supply during the period 0.7 Sec to 3 Sec as shown in Fig. 9(a). System dynamic state interval is observed between 0.5 Sec to 0.7 Sec

10

4

0

2

-10 -20 -30 0.45

0.5

0.55 0.6 Time (sec)

0.65

0.7

(b)

Active Power (Watts)

Load Current (Amp)

30

Supply Current

60 Supply Current (Amps)

DC Link Capacitor Voltage

Active Power Supplied

x 10

1.5 1 0.5 0

0

0.5

1

40

2

2.5

3

2.5

3

Active Power at Load 10000

0

-20 -40 -60 0.45

1.5 Time (sec)

(a)

20

0.5

0.55 0.6 Time (sec)

0.65

(c)

Fig. 7 Before and after compensation of (a) supply voltage (b) load current (c) supply current

0.7

Active Power (Watts)

Supply Voltage (volts)

The proposed MLI-DSTATCOM injects currents into the system in phase opposition to the harmonic currents generated by nonlinear load and restricts nonlinear currents to the load end which is depicted in Fig. 7(c). This makes the system supply current to be sinusoidal i.e. free from harmonics. In order to compensate the supply current harmonics and unbalanced current on phase B, voltage of DC link capacitor should be same i.e., at 1200V, this can be done by proper tuning of PI controller. The voltage across the DC link capacitor is observed to be constant nearly at 1200V as shown in Fig. 8.

8000 6000 4000 2000 0

0

0.5

1

1.5 Time (sec)

(b)

2

Active Power at MLI-DSTATCOM Reactive Power (VAR)

6000 4000 2000 0 -2000

0.5

1

1.5 Time (sec)

2

2.5

4000 2000 0 0.5

The reactive power at supply, load and MLI-DSTATCOM can also be analyzed from Fig. 10. From Fig. 10(a), load draws 1.96 kVAR reactive power. Before compensation, all reactive power required by load was supplied from source during which MLI-DSTATCOM was not connected until 0.5 Sec. After connecting MLI-DSTATCOM at 0.5 Sec, all reactive power required by load is supplied by MLI-DSTATCOM itself. This relieves source from supplying reactive power and it only supply active power as depicted in Fig. 10(b). Reactive power supplied by the proposed MLI-DSTATCOM is depicted in Fig. 10(c). Reactive Power at load

2.5

3

%THD for supply currents before and after compensation is shown in Fig. 11. It is observed from Fig. 11(a) that the fundamental component of supply current under nonlinear and unbalanced load is originated to be 16.77A with %THD of 21.20%. After compensation, the fundamental component of the supply current is increased to 19.52A and its % THD is reduced to 3.46% as shown in Fig. 11(b). Along with this study, simulations were also carried out for nonlinear balanced loads of conventional two level DSTATCOM, three level diode clamped and five level MLI-DSTATCOM and their respective fundamental component, %THD, active and reactive power flows are compared and are tabulated in Table 2, Table 3, Table 4 and Table 5 respectively.

1000 500 0

0.5

1

1.5 Time (sec)

2

2.5

3

100 80 60 40 20 0

0

(a) Mag (% of Fundamental)

6000 4000 2000 0 1

1.5 Time (sec)

(b)

2

20

Fundamental (50Hz) = 19.52 , THD= 3.46%

8000

0.5

5 10 15 Harmonic order (Hz)

(a)

Supplied Reactive Power

10000 R eac tiv e P ow er (V A R )

2

Fundamental (50Hz) = 16.77 , THD= 21.20%

1500

0

1.5 Time (sec)

(c)

2000

-2000

1

Fig. 10: Reactive power at (a) load (b) supply and (c) proposed MLIDSTATCOM

Fig. 9: Active power (a) supply (b) load and (c) proposed MLI-DSTATCOM.

0

0

3

(c)

Reac tiv e P ower (V A R)

6000

-2000

0

Reactive Power at MLI-DSTATCOM

8000

Mag (% of Fundamental)

Active Power (Watts)

8000

2.5

3

100 80 60 40 20 0

0

5 10 15 Harmonic order (Hz)

(b)

Fig. 11: % THD of supply current (a) before compensation (b) after compensation

20

Table 2: Comparison of fundamental component in conventional and MLI-DSTATCOM

Fundamental Component of Current (A) Supply Case 1 Load Supply Case 2 Load

Before

After Three Level 17.58 16.61 18.97 16.77

Two Level 16.58 16.61 17.63 16.77

16.61 16.61 16.77 16.77

Five Level 18.22 16.61 19.52 16.77

clamped MLI-DSTATCOM. The %THD of the supply current is observed to be very small after connecting the proposed five-level DSTATCOM. Hence, it can be concluded that the five-level diode clamped MLI-DSTATCOM gives superior performance compared to two-level and three-level DSTATCOM. REFERENCES [1]

Table 3: Comparison of %THD in conventional and MLI-DSTATCOM

%THD of Current Case 1 Case 2

Before

[2]

After Three Level 3.82%

Five Level 3.47%

Supply

20.81%

Two Level 5.14%

Load

20.81%

20.81%

20.81%

20.81%

Supply

21.20%

5.03%

3.73%

3.46%

Load

21.20%

21.20%

21.20%

21.20%

[3] [4] [5]

[6] Table 4: Comparison of Active power in conventional and MLI-DSTATCOM

Active Power (kW)

Case 1

Case 2

Before

Supply Load DSTATCOM Supply Load DSTATCOM

8.18 8.18 0 8.88 8.88 0

Two Level 8.3 8.18 0.12 9.05 8.88 0.17

After Three Level 8.7 8.18 0.52 9.4 8.88 0.52

[7]

Five Level 9 8.18 0.82 9.7 8.88 0.82

Table 5: Comparison of Reactive power in conventional and multilevel based DSTATCOM

Reactive Power (kVAR) Case 1

Case 2

Before

Supply Load DSTATCOM Supply Load DSTATCOM

2.08 2.08 0 1.96 1.96 0

Two Level 0 2.08 -2.08 0 1.96 -1.96

After Three Level 0 2.08 -2.08 0 1.96 -1.96

Five Level 0 2.08 -2.08 0 1.96 1.96

V. CONCLUSION The performance of a proposed five level diode clamped MLI-DSTATCOM is analyzed using Synchronous Reference Frame based control scheme under nonlinear balanced /unbalanced load. The performance of the proposed MLIDSTATCOM is compared with conventional two-level and three-level diode clamped MLI-DSTATCOM. From simulation results, it can be observed that the proposed fivelevel MLI-DSTATCOM compensates supply harmonics more effectively compared to the two-level and three level diode

[8]

[9] [10] [11] [12]

[13] [14] [15]

Ucar Mehmet, Ozdemir Engin, “Control of a 3-phase 4 leg active power filter under non-ideal mains voltage condition,” Electr Power Syst Res 2008:58–73. Phipps JK, Nelson JP, Sen PK., “Power quality and harmonic distortion on distribution systems,” IEEE Trans Ind Appl 1994;30:476-84. Lee TL, Hu SH, “Discrete frequency-tuning active filter to suppress harmonic resonances of closed-loop distribution power systems,” IEEE Trans Power Electr 2011;26(1):137–48. Ghosh A, Ledwich. G., Power Quality Enhancement Using Custom Power Devices, London, U.K.: Kluwer;2002. Lee Tzung-Lin, hushang-Hung, chanyu-Hung., “DSTATCOM with positive- sequence admittance and negative-sequence conductance to mitigate voltage fluctuations in high-level penetration of distributed generation systems,” IEEE Trans. Ind. Electron. 2013; 60(4):1417–28. Jing X, Cheng L., “An optimal-PID control algorithm for training feedforward neural networks,” IEEE Trans. Ind. Electron2013;60(6):227383. P. Rao, M.L. Crow, Z. Yang, “STATCOM control for system voltage control applications,” IEEE Trans. Power Delivery 15 (4) (2000) 13111317. P.S. Sensarma, K.R. Padiyar, V. Ramanarayanan, “Analysis and performance evaluation of distribution STATCOM for compensating voltage fluctuations,” IEEE Trans. Power Delivery 16 (2) (2001) 259264. Benysek Grzegorz, Pasko Marian., Power theories for improved power quality, London: Springer-Verlag; 2012. Benhabib MC, Saadate S., “New control approach for four-wire active power filter based on the use of synchronous reference frame,” Electr. Power. Syst. Res 2005; 73(3):353-62. Singh B, Solanki J., “A comparison of control algorithms for DSTATCOM,” IEEE Trans. Ind. Electron 2009; 56(7):2738-45. Da Silva CH, Pereira RR, da Silva LEB, Lambert-Torres G, Bose BK, Ahn SU., “A digital PLL scheme for three-phase system using modified synchronous reference frame,” IEEE Trans. Ind. Electron 2010;57(11):3814-21. Rahmani Salem, Hamadi Abdel amid, Kamal Al-Haddad A., “Lyapunov- function-based control for a three-phase shunt hybrid active filter,” IEEE Trans Ind Electron 2012; 59(3):1418-29. Rahmani S, Mendalek N, Al-Haddad K., “Experimental design of a nonlinear control technique for three-phase shunt active power filter,” IEEE Trans Ind Electron 2010; 57(10):3364-75. Kota V.R, Vinnakoti S., “SRF-based control of unified power quality conditioner for power quality enhancement,” IEEE (EESCO), 2015, Pages: 1-6.