Download An effective passive islanding detection method for PV

Available online at www.sciencedirect.com
ScienceDirect
Solar Energy 97 (2013) 155–167
www.elsevier.com/locate/solener
An effective passive islanding detection method for PV single-phase
grid-connected inverter
Ku Nurul Edhura Ku Ahmad ⇑, Nasrudin Abd Rahim, Jeyraj Selvaraj, Ahmad Rivai,
Krismadinata Chaniago
UM Power Energy Dedicated Advanced Centre (UMPEDAC), Level 4, Wisma R&D University of Malaya, Jalan Lembah Pantai, 59990 Kuala
Lumpur, Malaysia
Received 4 December 2012; received in revised form 31 July 2013; accepted 12 August 2013
Available online 12 September 2013
Communicated by: Associate Editor Nicola Romeo
Abstract
This paper presented a low cost and effective passive islanding detection method for single-phase photovoltaic grid-connected inverters. An analog circuit for over/under voltage protection is developed to ensure fast detection and no delay to system. An under/over
frequency circuit is also developed, for accurate and fast frequency detection with minimal external components. A new algorithm is
developed in a low-cost PIC18F4550. An improved disconnection time in the proposed method compared with that in the previously
developed method is an attractive solution for single phase grid connected inverters. The low cost, effective and minimal external component count are the advantages. A prototype is developed and tested to demonstrate the performance and feasibility of the proposed
method. The experiment results verified that the proposed islanding detection method able to detect islanding operation effectively under
various load types, inverter output powers and quality factors.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Passive islanding; Distributed generation; Inverter
1. Introduction
Development for renewable energy sources produce low
pollution compared to the fossil fuels and nuclear generation system (Yu et al., 2008). The new paradigm of distributed generation (DG) thus gains technical importance and
creates business opportunities (Chowdhury et al., 2009). In
principle distribution generation is a small scale generation
unit that need to be installed to the load and also connected
to the grid for selling or buying energy purposes. One of
the most important criteria that need to be considered is
the islanding issue (Yu et al., 2010). The islanding
condition as specified in Recommended Practice for Utility
Interface of Photovoltaic (PV) Systems (2000), occurs when
⇑ Corresponding author. Tel.: +60 3 22463246; fax: +60 3 22463257.
E-mail address: [email protected] (K.N.E.K. Ahmad).
0038-092X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.solener.2013.08.011
“a portion of utility system that contains both load and distributed resources remains energized while it is isolated
from the remainder of the utility system”. Such an undesirable event could potentially occur due to the circuit
tripping, accidental disconnection of the utility due to
equipment failure, human error, temporary disconnection
(for maintenance services) or uncommon network reconfiguration (Chowdhury et al., 2009; Yu et al., 2010). Integrating DG into utility is a major challenge to researchers. DG
could still be supplying local load demand while network is
already isolated from the main system. Existing methods
are still lacking, hence successful detection of islanding is
an ongoing challenge.
Two factors must be highlighted to better understand
the islanding phenomenon. The first one is the available
standards that have been established for the grid connected
system. These standards addressed the issue of islanding as
156
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
Nomenclature
PV
photovoltaic
DG
distributed generation
NDZ
none detection zone
PCC
point of common coupling
Qf
quality factor
LPS
load parameter space
PMS
power mismatch space
In,a
negative sequence current
OV/UV
over and under voltage
OF/UF
over and under frequency
PJD
phase jump detection
ROCOP
rate of change of power output
ROCOF rate of change of frequency
SMS
sliding mode frequency shift
PWM
pulse width modulation
MPPT
maximum power point tracking
R
resistor
L
inductance
well as procedure for testing and qualifying DG system
(Chiang et al., 2010). The second feature is associated with
the so called “Non-detection zone” (NDZ) which can be
defined as the zone for which an islanding detection
method would fail to operate in time. NDZ is an evaluating
criterion of islanding detection methods. In principle, islanding detection monitors changes in inverter output parameter or other system parameters that indicate islanding.
There are two types of anti islanding methods which are
the local and remote methods. The local methods can be
divided into passive and active. Passive islanding detection
detects changes in electrical parameters to determine the
occurrence of islanding (Chiang et al., 2010). The advantages of these passive techniques are their easy implementation (no additional controller), no degradation of PV
inverter power quality, and their inexpensiveness. The primary drawbacks are a relatively large NDZ and ineffectiveness in multi-PV inverter systems (Syamsuddin et al.,
2009). The most commonly used passive technique for
islanding detection are under/over voltage and under/over
frequency (OV/UV & OF/UF), phase jump detection
(PJD), voltage harmonic monitoring, current harmonic
monitoring, rate of change of power output (ROCOP)
and rate of change of frequency (ROCOF) Yu et al.,
2010; Llaria et al., 2011; De Mango et al., 2006; Freitas
et al., 2005; Redfern et al., 1993. Active techniques inject
a small disturbance at the PV inverter output to detect
islanding. The main advantage of these techniques is their
relatively smaller NDZ than that of passive methods. Their
main drawbacks are the possibly deteriorated output
power quality causing instability to the PV inverter and
normally require additional controllers which increased
the complexity of the method (Syamsuddin et al., 2009).
APS
SFS
AFD
RPEED
PLCC
T
R
SPD
SCADA
Vpcc
RMS
LED
IGBT
DFT
V
f
C
Qf
active phase shift
Sandia frequency shift
active frequency drift
reactive power export error detection
power line carrier communication
transmitter
receiver
signal produced by disconnect
supervisory control and data acquisition
voltage at point of common coupling
root mean square
light-emitting diode
insulated gate bipolar transistor
digital Fourier transformation
grid voltage
grid frequency
capacitance
quality factor
Active techniques developed include impedance measurement (IM), sliding mode frequency shift (SMS) or active
phase shift (APS), sandia frequency shift (SFS) or active
frequency drift with positive feedback, and reactive power
export error detection (RPEED) Chowdhury et al., 2009;
Mohamad et al., 2011; Lopes and Huili, 2006. AFD
method varies the output current frequency through positive feedback. This method will inject a current with
slightly distorted in frequency into the PCC. Upon grid disconnection, the phase error between the PCC voltage and
the inverter current will be detected by the inverter, which
then tries to compensate by increasing the frequency of the
injected current until it exceeds the OF/UF limits. However, the performance of this conventional AFD method
is inefficient and researchers are facing difficulty in choosing the suitable chopping fraction (cf) value to meet the
limit of harmonics. Hence, a novel AFD method (Jung
and Yu, 2007) with a periodic chopping fraction that deviates from the frequency in an instant way from nominal
was proposed. Remote islanding detection techniques are
based on communication between utilities and PV inverter
units (Syamsuddin et al., 2009). This technique do not have
the NDZ, do not degrade PV inverter power quality, are
effective in multi-PV-inverter systems, but are expensive
to implement (especially in small systems), and have a complicated communication technique for multi-PV inverter
systems. Common techniques that are communicationbased include power line carrier communication (PLCC),
signal produced by disconnect (SPD), and supervisory control and data acquisition (SCADA). In application, each
method has advantages and drawbacks. Among popular
reference standards for islanding are IEEE 929-2000, IEC
62116, IEE 1547, VDE 0126-1-1, and AS 4777.3-2005.
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
157
Table 1
Existing anti-islanding standards.
Quality factor, Qf
Required islanding
detection time, t
Normal frequency range,
f (nominal frequency f0)
Normal voltage range,
V (% of nominal voltage V0)
IEC 62116
1
t<2s
85% 6 V 6 115%
IEEE 1547
IEEE 929-2000
Japanese standard
1
2.5
0 (+rotating
machinery)
1
2
1
t<2s
t<2s
Passive: t < 0.5 s active:
0:5 s < t < 1 s
t < 0.5 s
t < 0.2 s
t<2s
ðf 0 1:5 HzÞ 6 f and
f 6 ðf 0 þ 1:5 HzÞ
59:3 Hz 6 f 6 60:5 Hz
59:3 Hz 6 f 6 60:5 Hz
Setting value
88% 6 V 6 110%
88% 6 V 6 110%
Setting value
59:3 Hz 6 f 6 60:5 Hz
47:5 Hz 6 f 6 50:2 Hz
Setting value
88% 6 V 6 110%
80% 6 V 6 115%
Setting value
Korean standard
VDE 0126-1-1
AS4777.3-2005
Table 1 shows all the standards having their own quality
factor (Q) value, the required islanding disconnection time,
and range of operational frequency and voltage (Yu et al.,
2010). According to IEEE929-2000 standard the quality
factor Q is equals to;
Q ¼ tanðarccosine½pfÞ
ð1Þ
The selected Q = 2.5 is equals to a power factor of 0.37.
As the power factor increases, Q decreases. Thus according
to IEEE929-2000, the test requirement that sets Q = 2.5
equals lines with uncorrected power factors from 0.37 to
unity and seems to cover all reasonable distribution line
configurations (Recommended Practice for Utility Interface of Photovoltaic (PV) Systems, 2000). Meanwhile, Japanese standard are proposing Q = 0 based on adding
rotating machinery while islanding test is being conducted
(Yu et al., 2010). Other than quality factor, disconnection
time is also important. The German VDE0126-1-1 standard has the strictest disconnection time limit which is less
than 0.2 s. Normal frequency and voltage range also are
crucial to determining islanding detection capability.
Australian standard AS4777.3-2005 requires nominal
frequency and voltage range to be set by manufacturer.
Steps to obtaining tripping values are: (1) determination
of under/over voltage and under/over frequency values
through gradual increase or decrease of voltage and
frequency until the device tested (the inverter) disconnects
from the variable ac supply, and (2) reading of the
over/under voltage or over/under frequency values at
which disconnection occurred. A criteria of acceptance
for over/under voltage is that the value at step (2) should
equal an under-voltage set point of ±5 V; for frequency,
the value should equal an under-frequency set point of
±0.1 Hz.
The primary objective of this paper is to propose a new
low cost and effective passive method for islanding detection of single-phase grid-connected PV inverter and to
compare between the proposed algorithm and existing
algorithm.
2. Research methodology
2.1. PV inverter system connected to grid
Fig. 1 is the block diagram of a typical single-phase gridconnected PV inverter; here, it has PV arrays, DC–DC
boost converter connected to two capacitors in series
(and functioning as maximum power point (MPP) tracker),
four switching devices connected in full bridge configuration (S1–S4) and presented as ideal switches, and a filter
inductor (Lf) filtering the current injected into the grid.
The power produced by the system is transferred directly
to the grid. Since grid voltage is uncontrollable, the simplest way of controlling system operation is by controlling
Fig. 1. Single-phase grid-connected PV inverter.
158
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
Fig. 2. The grid-connected PV inverter system with PIC 18F4550 microcontroller.
15V
R3
R1
LED
C1
+
R2
Grid
-
R1
Op
Amp1
LM311
R1
2W06
To Port
RD6
15V
R3
LED
R1
+
-
Op
Amp2
LM311
R1
To Port
RD7
Fig. 3. Under/over voltage circuit.
the current flowing into the grid. The injected current must
be sinusoidal and have low harmonic distortion (Rahim
et al., 2011; Rahim and Selvaraj, 2010).
2.2. Overall block diagram of the grid-connected PV inverter
with the proposed method
+
-
Fig. 2 is the block diagram of the grid-connected PV
inverter system with PIC 18F4550 microcontroller for
islanding detection integrated with the main controller
for the inverter system. In this paper, the proposed low cost
and effective passive method is developed using microcontroller PIC 18F4550. The controller handles the proposed
passive islanding of under/over voltage and under/over frequency detection before the signals are sent to the main
controller of the single-phase grid-connected inverter. The
proposed method takes the grid voltage to the under/over
Fig. 4. Zero-crossing detector circuit for under/over frequency detection.
voltage circuit and send the signals as input to PIC18F4550
for under/over voltage detection. For under/over frequency, the proposed method detects zero crossing
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
159
Table 2
Parameters of the single-phase grid-connected PV inverter prototype.
Fig. 5. Falling edge of a square wave signal with the triggered interrupt.
Parameter
Value
D1
S1–S4
Filter inductor (Lf)
Switching frequency
RHRP30120 VRR = 1200 V, I = 30 A
IGBT IRG4PH50UD VCE = 1200, Ic = 24 A
5 mH
20 kHz
The islanding detection algorithm output from PIC
18F4550 is input to the main inverter controller through
GPI0A11, the input then used as a signal to turn off the
relay, disconnecting the inverter from the grid (see Figs. 3
and 4).
2.3. The proposed under-voltage and over-voltage circuit
detection pulses and sends the signal as interrupt input to
PIC18F4550. These signals are used as conditional
detection in the islanding detection algorithm developed.
PIC 18F4550 was designed to read analog signals
between 0 V and 5 V, hence the under/over voltage circuit
Fig. 6. Flowchart of islanding detection.
160
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
Table 3
Load parameters of the simulation and experiment at Q = 2.5, and
P = 1 kW.
Table 4
Load parameters of the simulation and experiment at Q = 1, 2, 2.5, and
P = 1 kW.
Types
R (O)
L
C
Quality factor
R (O)
L (mH)
C (lF)
Pure resistor load
Resistor–inductor load
Resistor–capacitor load
Resistor–inductor-capacitor load
57.6
57.6
57.6
57.6
–
73.33 mH
–
73.33 mH
–
–
138.13 lF
138.13 lF
1
2
2.5
57.6
57.6
57.6
183.35
91.67
73.33
55.26
110.52
138.13
was designed to not exceed that range. In this work, a 240–
6 V transformer was connected to the utility source. A rectifier transforms the lower AC voltage into DC. Through
the addition of resistor R2 parallel with capacitor C1, this
circuit can convert a high voltage signal into a very stable
low DC voltage signal. The stable DC signal then act as
input to the under-voltage circuit which used the OpAmp1.
A 10 kO potentiometer is used as tuner for reference voltage Vref1. The DC voltage and reference voltage Vref1 are
compared, with OpAmp1 acting as comparator. OpAmp1
output is fed to RD6 as input signal to the microcontroller.
The same concept is applied to the over voltage circuit,
except that the location of the reference voltage is now at
the positive side of OpAmp2 and the stable DC voltage
now connects to the negative input of OpAmp2. This circuit was developed for accurate and fast detection of
under/over voltage without continuous calculation of the
grid voltage RMS as in the past method. Not only does
detection become very fast, there is no added delay.
forces the voltage down. Resistor pair R2 is a voltage divider, producing 2.5 V as reference voltage to Op Amp
LM311. The circuit output connects to RB0. Capacitor
C1 connects to the Op Amp output to act as noise filter
that gives the possibility to prevent noise in a simple and
cost effective way. It improves triggering of the interrupt
by reducing the noise in the low-voltage square-wave signals. As the square-wave signal is in phase with the grid
voltage, the falling edge will indicate very accurately where
the zero crossing is. This square wave signal will be the
input to interrupt routine INT0, making possible the
design of a zero-crossing-detection routine in an interrupt
routine and automatically making the detection fully interrupt-driven. Fig. 5 shows a square-wave input signal and
the triggered interrupt.
2.5. Islanding algorithm
Fig. 6 shows the control algorithm of the proposed passive islanding detection, applied in PIC18F4550, which
processes the algorithm through C programming language.
PIC18F4550 collects analog inputs data from the under/
over voltage circuit through RD7 and RD6. There is a filter
in the programming algorithm to meet AS 4777.3-2005
2.4. The proposed under-frequency and over-frequency
circuit
This circuit is to ensure accurate and fast frequency
detection with minimal external components. Resistor R1
Voltage sensor
PM6000
Current sensor
LAN Area
LeCroy
Oscilloscope
PC for PV Array Simulator
R
C
L
Inverter
PC for PM6000
PV Array
Simulator
Programmable
AC source
GPIB and serial USB
(a)
Computer
PM6000
Oscilloscope
under and over
voltage circuit
under and over
frequency circuit
Inverter
RLC load
(b)
Fig. 7. (a) Islanding test and (b) the laboratory test equipment with the proposed passive islanding detection method.
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
161
Fig. 8. Simulation model of the islanding test circuit.
standard reconnection time. For frequency calculation, a
simple zero-crossing circuit is utilized. The square wave
produced by the zero-crossing detector has the same frequency as the grid. By feeding the produced signal into
the RB0 of the PIC18F4550, the microcontroller is programmed to be a very accurate zero-crossing detector with
an interrupt-driven code.
Fig. 7 shows the PV grid-connected inverter hardware
prototype controller interface with the islanding detection,
implemented on PIC 18F4550 to validate the performance
of the proposed passive islanding detection algorithm.
Table 2 lists the prototype parameters of the single phase
grid-connected PV inverter. The hardware was tested under
various load conditions. Equipment such as PV array
Vgrid
Vgrid
400
200
0
-200
-400
400
200
0
-200
-400
Vpcc
Vpcc
400
200
0
-200
-400
400
200
0
-200
-400
Iinv*10
Iinv*10
100
50
0
-50
-100
100
50
0
-50
-100
0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
0.6
Time (s)
Time (s)
(a)
(b)
Vgrid
400
200
0
-200
-400
Vpcc
400
200
0
-200
-400
Iinv*10
100
50
0
-50
-100
0
0.2
0.4
0.6
0.8
1
1.2
Time (s)
(c)
Fig. 9. Simulation result for islanding detection, at (a) Q = 1, (b) Q = 2, and (c) Q = 2.5.
0.8
1
1.2
162
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
simulator, resistor-inductor-capacitor (R–L–C) load, programmable AC source, PM6000, and LeCroy Oscilloscope
were used. The test was conducted from 500 W to 2 kW
power output. In the following experiments, the grid-connected inverter was controlled to generate balanced real
power with the load. Various load types were used to verify
the proposed islanding detection method. Table 3 lists the
load parameters of the experiments. The parameters of
the resistor–inductor–capacitor (R–L–C) load were as
specified by the IEEE Std. 929-2000 (Recommended Practice for Utility Interface of Photovoltaic (PV) Systems,
2000) for testing islanding operation performance. IEEE
Std. 929-2000 states that a non-islanding inverter must shut
down within 2 s of grid disconnection for loads with a quality factor (Qf) equals to or lower than 2.5. The values for L
and C are calculated through this formula:
P¼
V2
R
ð2Þ
L¼
R
Qf x
ð3Þ
C¼
Qf
R
2
L
ð4Þ
Fig. 8 is a diagram of the PSIM model of the islanding
circuit.
3. Results and discussion
3.1. Disconnection time for different method and algorithm
To prove that the proposed passive islanding detection
has advantages over the implemented algorithm in
TMS320F2812 DSP in terms of disconnection times, the
corresponding comparisons were made on both passive
islanding methods. Low cost PIC 18F4550, compared with
ATMEGA and TMS320F2812 DSP with under/over voltage and under/over frequency islanding detection algorithms, were compared. The circuit connections and
algorithms of the past methods are as in Syamsuddin
et al. (2009). The prototype of both methods monitored
grid voltage and grid frequency at PCC (see Table 4).
Vpcc
400
200
0
-200
-400
Iinv*10
100
50
0
-50
-100
0
0.2
0.4
0.6
0.8
1
1.2
0.8
1
1.2
Time (s)
(a)
Vpcc
400
200
0
-200
-400
Iinv*10
100
50
0
-50
-100
0
0.2
0.4
0.6
Time (s)
(b)
Fig. 10. Simulation result for (a) R–L load (b) R–C load under islanding operation with Q = 2.5 and P = 1 kW.
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
163
Table 5
Disconnection times of the two methods.
ATMEGA and DSP based
algorithm
PIC 18F4550 algorithm
Under voltage
disconnection time (s)
Over voltage
disconnection time (s)
Under frequency
disconnection time (s)
Over frequency
disconnection time (s)
Could not be determined
1.2
2.1
52.5
0.6
0.6
0.6
0.6
Fig. 11. Experiment results for R load under islanding operation, for (a) 25% (500 W), (b) 50% (1 kW), (c) 75% (1.5 kW), and (d) 100% (2 kW) inverter
output, with (M1) Vgrid, (M2) Vpcc, (M3) the inverter output current and (M4) load current.
The disconnection time of PIC18F4550 was much faster
than that of ATMEGA and TMS320F2812. The cost of
design and implementation for the proposed method were
also much less. The disconnection times were validated
according to AS 4777.3-2005 standard. Table 5 compares
the disconnection times.
As the table shows, the proposed method is much faster
and more stable than the past method (very unstable).
Although over-voltage disconnection time of the past
method passed the less-than-2-s standard, the proposed
method enabled much faster disconnection. From Syamsuddin et al. (2009), this method used TMS320F2812
DSP for reading the grid voltage at the sensor signal
through the ADC and calculates the RMS voltage value
for under/over voltage disconnection. This may add some
lagging to the systems when detecting under/over voltage.
The proposed circuit and algorithm aim to correct the
problem.
3.2. Simulation results
Simulations were performed using PSIM system. The
DC voltage source was 400 V, the grid voltage 240 V, the
grid frequency 50 Hz, and the output filter inductance
5mH. The inverter was designed for 1.4 kW power output.
The RMS current injected into the grid was 6 A. The main
parameters of the simulation were as those of the experiments. The inverter was loaded by a parallel R–L–C circuit
164
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
with quality factors 1, 2, and 2.5. Fig. 9 shows the point of
common coupling voltage Vpcc, the inverter current and
voltage, and grid current and voltage ideally when the
inverter is tripped off.
Fig. 10(a) shows the simulation result for R–L under
islanding operation with 1 kW inverter output power.
The inverter output current is almost the same for both
simulation and experiment results (see Fig. 10(a) and see
Fig. 12(a)). From the simulation result for R–L load under
islanding condition, the value of the maximum inverter
output current before inverter is tripped off is 7.08 A and
5.7 A from the experiment (see Fig. 12(a)). Fig. 10(b) shows
the simulation result for R–C under islanding operation
with 1 kW inverter output power. The Vpcc voltage is
almost the same for both simulation and experiment results
(see Fig. 10(b) and 12(b)). From the simulation result for
R–C load under islanding condition, the value of Vpcc voltage before inverter is tripped off is 160 V and 216 V from
the experiment (see Fig. 12(b)).
3.3. Experiment results
The grid system was first connected to the inverter and
then removed to create islanding. Fig. 11 gives the different
Fig. 12. Experiment results for (a) R–L load, and (b) R–C load under islanding operation, with (M1) Vgrid, (M2) Vpcc, (M3) the inverter output current
and (M4) load current.
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
165
Fig. 13. Experiment results for R–L–C load under islanding operation, at (a) Q = 1, (b) Q = 2, and (c) Q = 2.5, with (M1) Vgrid, (M2) Vpcc, (M3) the
inverter output current and (M4) load current.
Fig. 14. Experiment results for R–L–C load under islanding operation, at Q = 2.5 with (M2) the inverter output current, (M3) load current.
166
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
Inverter starts to deliver power
Watt
Inverter power
Load power
load
power
grid
power
inverter
power
Grid power
seconds
Fig. 15. The inverter comes “on-line” and gradually increases its output to 1 kW.
Table 6
Detection times for the R–L–C load with Q = 2, as taken by PM6000
Universal Power Analyzer.
Timestamp
20130130155137.30
20130130155137.30
20130130155137.30
20130130155137.60
20130130155137.60
20130130155137.60
B
C
D
B
C
D
Channel
Arms
Watt
Load
Grid
Inverter
Load
Grid
Inverter
5.00
2.51
5.35
0.00
0.00
0.00
1091.88
182.38
1283.13
0.00
0.00
0.00
Table 7
Detection times for the different quality factors.
Capacity inverter
output
Quality factor of the RLC
Load
Detection
time
1 kW
1 kW
1 kW
1
2
2.5
21.25 ms
24.8 ms
55.7 ms
Table 8
Response to abnormal voltages.
Voltage (at PCC)
Maximum trip time
V < 120 (V < 50%)
120 6 V < 211 (50% 6 V < 88%)
211 6 V 6 264 (88% 6 V 6 110%)
264 < V < 329 (110% < V < 137%)
329 6 V (137% 6 V)
6 cycles
120 cycles
Normal operation
120 cycles
2 cycles
inverter outputs: 25%, 50%, 75% and 100% with pure resistive load under islanding condition. The results allow conclusion that the ratio of the inverter’s output and load
consumption can affect the amplitude of the load voltage.
Fig. 12 shows the experiment results for the R–L and R–
C loads under islanding operation. In the following experiments, the grid-connected inverter will be tripped off when
islanding is detected.
Under islanding conditions, the magnitude and frequency of the voltage at PCC tend to slide from the rated
grid values. It has been proved that the larger the power
imbalance (DP and DQ in local generation and consumption in the islanded system prior to grid disconnection),
the larger the variations are in the frequency of the voltage
and voltage at the PCC. Therefore, standard under/over
frequency and under/over voltage protections are successful in preventing islanding in system with large enough
power imbalances. The islanding condition for the experiments conducted was made to have the smallest power
imbalance possible, i.e., the worst-case scenario. Fig. 13
shows the experiment results for R–L–C load under islanding, at Q = 1, Q = 2, and Q = 2.5. In the following experiments, the grid-connected inverter will be tripped off
21.25 ms (Q = 1), 24.8 ms (Q = 2) and 55.7 ms (Q = 2.5)
after the grid is disconnected. From all the following experiments (see Fig. 13), the inverter (Vpcc) briefly swings up to
slightly more than + and 400 volt peak after the grid is
disconnected, and then shuts down. The detection time of
islanding operation may be different under different loads
and different quality factors of RLC load. As Fig. 14
shows, (M2) and (M3), at inverter output 1 kW and R–
L–C load with Q = 2.5, the inverter output current was
almost identical with the load current. This indicates the
worst case for islanding detection, where the power-flow
between the grid-connected inverter and the local load is
balanced.
Table 6 lists the disconnection times for when the grid
was disconnected until the inverter was tripped off. As
shown in Table 6 the disconnection time is 0.3 s (24.8 ms)
for R–L–C load with Q = 2. PM6000 Universal Power
analyzer was set to the minimum time setting available
for data recording (every 0.3 s). The detection time satisfies
the IEEE 929-2000 control standard, verifying that the proposed islanding method is able to detect islanding effectively. As shown in Table 6, the inverter current value
immediately zeros out after the grid is disconnected (see
Table 7).
According to IEEE Std. 929-2000 standard, the relationship between the amplitude of the load voltage and the
K.N.E.K. Ahmad et al. / Solar Energy 97 (2013) 155–167
maximum trip time for determining the islanding condition
are shown in Table 8. All the experiment results show the
amplitudes of the local load (Vpcc) to have passed the following requirement.
The proposed method is shown to effectively and effortlessly detect islanding operation under different load types
and quality factors. As Fig. 15 shows, the inverter comes
“on-line” and gradually increases its output power to
1300 W. Fig. 15 shows the transition as the inverter power
increases and the grid transitions from delivering the load
power to negative. Note that the load power stays constant
during the transition period.
4. Conclusions
A low cost and effective passive method for islanding
detection of single-phase grid-connected inverter is proposed. An analog circuit for under/over voltage protection
has been developed, to ensure fast detection with no added
delay. An under/over frequency circuit has also been developed, for accurate and fast frequency detection and with
minimal external components. A new algorithm is developed in a low cost PIC18F4550. The circuit topology, control algorithm, and operational principle of the proposed
method have been presented. PIC18F4550 optimized the
method’s performance, which has improved the disconnection time of a past method. The inexpensiveness, effectiveness, and minimal external component count are the
proposed method’s advantages. Its performance and feasibility has been tested and proven by a purpose-developed
prototype. The main improvements are the disconnection
times, effectiveness in detecting islanding under various
load types, inverter output powers and quality factors
and also cost down of the end product.
Acknowledgments
This work was supported by UMPEDAC (University of
Malaya Power Energy Dedicated Advanced Centre), a flagship HICoE (Higher Institution Centre of Excellence) of the
Ministry of Higher Education, Malaysia and HIR Grant H16001-00-0000032 Campus Network Smart Grid System for
Energy Security. The authors thank all concerned.
167
References
Chiang, W.-J., Jou, H.-L., Wu, J.-C., Wu, K.-D., Feng, Y.-T., 2010.
Active islanding detection method for the grid-connected photovoltaic
generation system. Electric Power Systems Research 80, 372–379.
Chowdhury, S.P., Chowdhury, S., Crossley, P.A., 2009. Islanding
protection of active distribution networks with renewable distributed
generators: a comprehensive survey. Electric Power Systems Research
79, 984–992.
De Mango, F., Liserre, M., Aquila, A.D., Pigazo, A., 2006. Overview of
Anti-Islanding Algorithms for PV Systems. Part I: Passive Methods.
In: Power Electronics and Motion Control Conference. EPE-PEMC
2006. 12th International, pp. 1878–1883.
Freitas, W., Wilsun, X., Affonso, C.M., Zhenyu, H., 2005. Comparative
analysis between ROCOF and vector surge relays for distributed
generation applications. Power Delivery, IEEE Transactions 20, 1315–
1324.
Youngseok Jung, J.C., Yu, Gwonjong., 2007. A novel active anti-islanding
method for grid-connected photovoltaic inverter. Journal of Power
Electronics.
Llaria, A., Curea, O., Jiménez, J., Camblong, H., 2011. Survey on
microgrids: unplanned islanding and related inverter control techniques. Renewable Energy 36, 2052–2061.
Lopes, L.A.C., Huili, S., 2006. Performance assessment of active
frequency drifting islanding detection methods. Energy Conversion,
IEEE Transactions 21, 171–180.
Mohamad, H., Mokhlis, H., Bakar, A.H.A., Ping, H.W., 2011. A review
on islanding operation and control for distribution network connected
with small hydro power plant. Renewable and Sustainable Energy
Reviews 15, 3952–3962.
Rahim, N.A., Selvaraj, J., 2010. Multistring five-level inverter with novel
PWM control scheme for PV application. Industrial Electronics, IEEE
Transactions 57, 2111–2123.
Rahim, N.A., Chaniago, K., Selvaraj, J., 2011. Single-phase seven-level
grid-connected inverter for photovoltaic system. Industrial Electronics,
IEEE Transactions 58, 2435–2443.
IEEE Recommended Practice for Utility Interface of Photovoltaic (PV)
Systems, 2000. IEEE Std 929–2000, i.
Redfern, M.A., Usta, O., Fielding, G., 1993. Protection against loss of
utility grid supply for a dispersed storage and generation unit. Power
Delivery, IEEE Transactions 8, 948–954.
Syamsuddin, S., Rahim, N.A., Krismadinata, Selvaraj, J., 2009. Implementation of TMS320F2812 in islanding detection for photovoltaic
grid connected inverter. In: Technical Postgraduates (TECHPOS),
2009 International Conference for, pp. 1–5.
Yu, B., Matsui, M., So, J., Yu, G., 2008. A high power quality antiislanding method using effective power variation. Solar Energy 82,
368–378.
Yu, B., Matsui, M., Yu, G., 2010. A review of current anti-islanding
methods for photovoltaic power system. Solar Energy 84, 745–754.