Real-Time Implementation of A SPLL For FACTs
Real-Time Implementation of A SPLL For FACTs
q
PI 1/s
A A
Voltage sag from 1 p.u. to 0.5 p.u. and phase-angle jump of -45 deg.
Vphase qerr w ,
qo.
Phase Detector
A
1
PI 1/s
A
0.8
Loop filter DCO
qo.
A
0.6
0.4
Amplitude (p.u.)
Fig. 2. One-phase SPLL equivalent scheme of classical PLL. 0.2
2391
3
2392
4
filter. However the gain of the loop filter for 2ω is not small H(z) = (1 − z − F req·Ts ) (4)
z−1
enough to cancel the influence of the second harmonic in the
This block can be optionally used in amplitude-loop for
dynamics. This influence is clearly visible in analog PLLs
avoiding the ripple in the amplitude measure (Fig.5).
where the phase measure has a ripple over the ideal ramp.
The phase detector can help to mitigate the second harmonic
generation. The first try of removing the problem has been ap- D. Coefficients choice and performance
plied in [2] moving the second harmonic to a higher frequency The rigorous study of the SPLL dynamics is quite complex,
which is much more attenuated in the low-pass filters.A less because it contains a lot of nonlinearities; however, the use
complex, but effective method, has been developed in the of software tools can help the designer in the stability and
PSB/SPLL [12]. The second harmonic is estimated from the performance analysis. The “Simulink response optimization
output phase. Then it is directly subtracted in the output of tool” (SRO) has been employed in this work [13]. The
2393
5
50
40
1
30
20
0.5
Phase-error (deg.)
Amplitude (p.u.)
10
0
0
-10
-20
-0.5
-30
Vspll
-40
-1 Vphase
-50
0 0.05 0.1 0.15 0.2 0.25 0.3
Time (sec)
0.1 0.15 0.2 0.25 0.3 0.35
Time (s)
Fig. 9. Fitting the phase-angle error during a 0.5 p.u. sag with a +45 deg.
phase-angle jump using the SRO tool.
Fig. 10. SPLL response to a 0.5 p.u. sag with a +45 deg. phase-angle jump.
1.1
Fig. 11. SPLL amplitude during the 0.5 p.u. sag with +45 deg. phase jump.
V. S IMULATION RESULTS
Simulations results have been obtained using Mat-
lab/Simulink. These results analyze the main SPLL features used to determine the time used by the SPLL for tracking
during a voltage sag with phase-angle jump of +45 deg, similar after a disturbance; determining when the system is tracked
to Fig. 3. The main reason of electing this disturbance is its depends on a designer criterion. For example, for a phase-
significance over the SPLL whole performance, because it has jump of 45 deg. the designer may establish that the system is
an amplitude depression and a big phase-angle jump. tracked when the phase-error is less than 2.5 deg. However,
→
− this is not an absolute measurement, because a system with
In Fig. 10 it is shown V phase with the disturbance (dashed)
→
− →
− high overshoot can achieve quickly a phase-error smaller than
and the SPLL response V spll (solid); V spll is obtained
multiplying the output amplitude by the normalized output the threshold value, but due to oscillation the system cannot
signal. It can be seen that in approximately a cycle the SPLL be considered tracked.
output is adapted to fault voltage and it spends the same time Comparing these results with the results of adaptive SPLL
to return to pre-fault situation. shown in [2], it can be seen that both systems have similar
The SPLL amplitude measured during the disturbance is performance under amplitude sags and phase-angle jumps.
shown in Fig. 11. It can be seen that the overshoot is greater
→
− VI. R EAL -T IME IMPLEMENTATION RESULTS
when V phase returns to pre-fault situation. It is important to
say that the adaption time depends on the deep of the sag, To analyze the feasibility of the SPLL in a real system it has
being especially critical for very low amplitude sags. Tuning been implemented using the fast prototyping platform dSpace
the SPLL for a faster response to deep sags, involves a high DS1103 which contains a microprocessor, a DSP and an I/O
gain in the amplitude-loop which generates large overshoot in interface. A programmable AC power source has been used
the outputs during the transients. to generate the disturbances. The inputs and outputs of the
In Fig. 12 it is shown the SPLL response for the disturbance prototype have been captured in a digital oscilloscope.
phase-angle steps. The difference between SPLL phase and In Fig. 13 they are shown the input signal from the ac
the real phase gives the phase-error. The phase-error is widely source, the SPLL output signal and the normalized measured
2394
6
60
50
40 Vphase
Phase-angle (deg.)
VSPLL
30
20
10
-10
0.1 0.15 0.2 0.25 0.3 0.35 0.4
Time (s)
Fig. 12. SPLL phase-angle response to a 0.5 p.u. sag with a +45 deg.
phase-angle jump.
Fig. 14. Real-Time implemented SPLL response to a 0.5 sag with phase-angle
Vphase jump of +45 deg. (Vphase dashed, Vspll solid).
ACKNOWLEDGEMENT
Vspll Phase
This work was supported by the Spanish Ministry of Science
and Technology under project number DPI2003-01513.
R EFERENCES
[1] N. G. Hingorani and L. Gyugyi., Understanding FACTS, R. J. Herrick.,
Ed. IEEE Press., 2000.
[2] D. Jovcic, “Phase locked loop system for FACTS,” IEEE Transactions
on Power Systems, vol. 18, no. 3, pp. 1116–1124, August 2003.
[3] A. H. Norouzi and A. M. Sharaf., “Two control schemes to enhance the
dynamic performance of the statcom and sssc,” IEEE Transactions on
power delivery., vol. 20, pp. 435–442, 2005.
Fig. 13. Input Vphase , output Vspll and SPLL measured phase-angle.
[4] J. G. Nielsen, F. Blaabjerg, and N. Mohan, “Control strategies for
dynamic voltage restorer compensating voltage sags with phase jump,”
Proceed. of APEC2001, vol. 2, 2001, pp. 1268-1273., vol. 2, pp. 1268–
phase-angle during a typical steady-state situation. It can be 1273, 2001.
[5] C. Zhang, C. Fitzer, V. K. Ramachandraramurthy, A. Arulampalam,
seen that the output signal is closely equal to reference. M. Barnes, and N. Jenkins, “Software phase-locked loop applied to
Moreover, the phase-angle ramp is straight and it has not dynamic voltage restores (DVR),” Proc. IEEE Power Eng. Soc. Winter
second harmonic, because it has been removed previously in Meeting. Conf., vol. 3, pp. 1033–1038, 2001.
[6] V. Kaura and V. Blasko., “Operation of a phase locked loop system
phase-detector. under distorted utility conditions.” IEEE IEEE Transactions on industry
In Fig. 14. the real-time response of the SPLL for the applications., vol. 33, pp. 58–63, 1997.
disturbance analyzed in simulation is shown. The input of [7] H. Awad, J. Svensson, and M. Bollen., “Phase-locked loop for static
series compensator.” in EPE, 2003.
SPLL with the disturbance and the synchronized oscilloscope [8] P. Rodriguez, J. Pou, J. Bergas, I. Candela, R. Burgos, and D. Boroyevic,
trigger signal have been generated in the AC power source. It “Double synchronous reference frame pll for power converters control,”
can be seen that the implementation results are very similar in Conference on Power Electronics Specialists, 2005.
[9] ABB, “Series voltage restorer-data sheet.” 2004, 3BHT 490 484 R001.
to the simulation results, which confirms the feasibility of a [10] R. E. Best, Phase Locked Loops. Design, Simulation and Applications.
discrete implementation of the model. 4th Edition. McGraw-Hill, 1999.
[11] M. H. J. Bollen., Understanding power quality problems. Voltage sags
and interrupions., R. J. Herrick., Ed. IEEE Press Editorial Board.,
VII. C ONCLUSIONS 2000.
[12] Mathwoks, Power System BlockSet, User Manual., 2000.
A SPLL for FACTS control has been proposed, and its [13] Mathworks., Simulink Response Optimization For Use with Simulink.
features have been shown with simulation and real-time im- User’s guide., 3rd ed., March. 2006.
plementation. Moreover, the used algorithms make the discrete
2395