- Email: [email protected]
Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode
Accepted Manuscript Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode Suk Whan Ko, Young Chul Ju, Hye Mi Hwang, Jung Hun So, Young-Seok Jung, Hyung-Jun Song, Hee-eun Song, Soo-Hyun Kim, Gi Hwan Kang PII:
S0360-5442(17)30595-9
DOI:
10.1016/j.energy.2017.04.030
Reference:
EGY 10670
To appear in:
Energy
Received Date: 21 November 2016 Revised Date:
7 April 2017
Accepted Date: 8 April 2017
Please cite this article as: Ko SW, Ju YC, Hwang HM, So JH, Jung Y-S, Song H-J, Song H-e, Kim S-H, Kang GH, Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode, Energy (2017), doi: 10.1016/j.energy.2017.04.030. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 2 3
Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode
RI PT
4
Abstract:
6
In this paper, characteristics of PV (photovoltaic) modules under partial shading or with a
7
damaged bypass diode in the junction box were evaluated by comparing a theoretical model
8
and empirical data. For the electrical analysis of the current- voltage (I-V) curve of each
9
module, a PV module with one diode cell model was proposed, and the model closely
10
matched the empirical results. The damaged bypass diode was replaced with an element of
11
resistance in the simulation model. The calculation shows that the open circuit voltage of the
12
PV module with a damaged bypass diode was slightly higher than that of a PV module under
13
shading conditions while the PV system was operating. The I-V curve of each module
14
obtained with the solar simulator was similar to the results of the simulation. From the results
15
of field testing each PV module, when the PV system was operating in connection with the
16
power grid, the internal temperature of the junction box connected to the shaded PV module
17
was 5 °C higher than that of the PV module with the damaged bypass diode. Furthermore
18
when the PV system was not operating, the internal temperature of the junction box in the PV
19
module with the damaged bypass diode was extremely high. This condition caused a short-
20
circuit and the surface temperature of the damaged bypass diode reached 219 °C. In this
21
paper, we theoretically and empirically analyzed the characteristics of a shaded PV module
22
and a module with a damaged bypass diode.
M AN U
TE D
EP
AC C
23
SC
5
24
Keywords: partial shading, damaged bypass diode, photovoltaic module, open circuit voltage,
25
short circuit current
26 27
1. Introduction
28
A bypass diode is widely used in PV (photovoltaic) modules to prevent fire caused by hot
29
spots as well as to decrease energy losses during shading and mismatching. Bypass diodes,
ACCEPTED MANUSCRIPT connected in reverse bias between a solar cell strings positive and negative output
31
terminals, generally is used per small group of series cells. For instance, in the case of a
32
250 W PV module consisting of 60 solar cells, the bypass diodes are integrated into the
33
PV module in parallel for every 20 solar cells. Various studies on bypass diode
34
configurations, summarized in Table 1, have improved the efficiency of PV system under
35
mismatch condition induced by partially shading, abnormally working electrical
36
components and so on. The power loss in a partial shaded module is inevitable as
37
presented as proven in theoretically calculation [1] and [2] and empirically demonstrated
38
its output characteristics [3] and [4]. For overcoming this issue, the method for additional
39
bypass diode configuration of array with a series-connected PV module [5] was proposed.
40
Additionally, an optimal configuration to suppress line and mismatch losses was
41
formulated [6] and [7]. Moreover, various studies were conducted on the electrical
42
variation and impact on PV module and system. A method of reducing output loss using
43
MPPT (Maximum Power Point Tracking) algorithm and dc-dc converter has been studied
44
[8] and various MPPT techniques have been reviewed to mitigate partial shading effects
45
in PV systems [9]. Recently, new MPPT technique using Hybrid DESPO (Differential
46
Evolutionary algorithm and Particle Swarm Optimization) method is developed [10].
47
Among these techniques, the introduction of bypass diode to PV module is a simple yet
48
effective way to circumvent the deficiency in mismatched modules. Previous works about
49
cells and bypass diode in [11] have proved that its installation is an efficient way for
50
improving maximum power and reliability of PV module. Additionally, PV module using
51
a smart bypass was suggested in [12] to minimize the power loss. The influence of the
52
bypass diode can be observed with the change in shape of the I-V (current-voltage) and P-
53
V (power-voltage) curves of PV module. Previous researches, [13] and [14], investigated
54
the characterization of the bypass diode in PV systems or arrays under an influence of
55
mismatch. Normally, mismatch losses in a PV module is occurred by a partial shading or
56
a damaged bypass diode. When the PV is under partial shading, the attainable power from
57
PV cell array is determined by shaded cell. However, the normally working bypass diode
58
makes a new path, thereby resulting in reduced losses. On the other hand, if a bypass
59
diode of the PV module is damaged, the sub-circuit of the PV module will enter a short-
60
circuit state. As a result of such a mismatching in the system, a failure to detect the
61
damaged bypass diode will lead to significant system losses, as reported in [15]. For
62
instance, the voltage of PV module with a damaged bypass diode will drop by 33 %
AC C
EP
TE D
M AN U
SC
RI PT
30
ACCEPTED MANUSCRIPT compared to that of one employing a normally working diode. In the case of a large PV
64
system with a centralized inverter system, the system loss will be larger than that of single
65
PV module, and may even worse depending on the number of series and parallel
66
connections. Also, the failure of a bypass diode on a PV module is severely dangerous
67
because the surface temperature rises dramatically: enough to cause a fire. The Schottky
68
type diode, widely used in PV module, has a lower surge reliability, a lower thermal
69
resistance and high leakage current [16]. For this reason, researches about mechanism of
70
failure in Schottky diode by surge voltage have been conducted [17] and [18].
71
Nevertheless, bypass diode failures, that occasionally occur owing to its low reliability,
72
are difficult to detect because the PV system works normally even with bypass diode in
73
trouble. Until now, in the conventional diagnosis method, the damaged bypass diodes can
74
be only detected by measuring the terminal voltage of bypass diode in junction box
75
during regular maintenance at PV plants using diagnostic sensors. Unfortunately, the
76
failure of a PV system is only recognized, when PV modules show a severe mismatch that
77
induces power loss in the system. According to a PV system at AIST (Advanced
78
Industrial Science and Technology) in Japan [19], where 53 units of 4 kWp PV arrays
79
were operated, only few malfunction diodes were detected among hundreds of PV
80
module with them. The inspection of bypass diodes of 1272 PV modules with 180Wp
81
showed that 47 % of the modules had damaged ones. However, burn marks on sub-
82
modules was observed only 3% of defective PV modules. Moreover, the P-V curve of PV
83
module with damaged diode is similar to that of module under mismatching condition,
84
likes as partially shading. Thus, it is not easy to manage the power loss due to uncertainty
85
of its origin. The loss from shaded module is a temporary one, while the failure of bypass
86
diode causes a continuous loss. However, the systematically exploration of each condition,
87
PV module with shading or failure of bypass diode, has not been conducted so far.
88
Therefore, effective diagnosis method to find out the origin of mismatch is imperative in
89
PV system. In this paper, we demonstrate a new method to discriminate damaged bypass
90
diode in PV module. Electric circuit modeling and simulation were performed to
91
distinguish between a shaded PV module and a module with one damaged bypass diode.
92
Moreover, thermo-electric properties of bypass diode in various conditions were
93
empirically investigated. Through the analysis of each condition of the PV module using
94
experimental data, elements for a fault diagnosis of a bypass diode in a PV module were
95
proposed by monitoring its temperature under operating condition.
AC C
EP
TE D
M AN U
SC
RI PT
63
ACCEPTED MANUSCRIPT 96
2. Experiments on the thermal characteristics of a bypass diode
97
The current through a Schottky barrier diode under a forward bias ‘V’ is given by the
98
following relation [20].
99
=
102
− 1
(1)
where , the reverse saturation current, is given as follows. = ∗∗
∅
RI PT
101
(2)
Here, Ad is the diode area, A∗∗ is the effective Richardson constant (12 A cm−2 K−2), T is the
SC
100
104
temperature, k is the Boltzmann constant, q is the electronic charge, ∅ is the energy barrier
105
Fig. 1 shows the I-V and P-V characteristics of normal and damaged bypass diodes measured
106
using a source meter. The difference in power consumption between a normal and damaged
107
diode is 5.41 W with forward current of 10 A. The bypass diode blocks the current under the
108
normal working condition of a PV module. Meanwhile, in shaded PV modules, the direction
109
of current in the bypass diode is forward. In contrast, in the case with a damaged bypass
110
diode, the current flows in the reverse direction regardless of operation of the PV system.
111
Hence, the thermal characteristics of bypass diodes as a function of current are analyzed
112
under the forward and reverse bias. Here, the surface temperature and the resistance of the
113
bypass diode were measured under various currents by using thermal imaging and resistance
114
measuring equipment, when the temperature was saturated.
115
Measurement results are shown in Fig. 2 and summarized in Table 2. Fig. 2 represents the
116
surface temperature of normal and damaged diodes as a function of current. The surface
117
temperature of the normal diode increased linearly as the forward current increased. On the
118
other hand, the temperature of the damaged diode rose exponentially according to the reverse
119
current and eventually the temperature exceeded 200 °C at 10 A, being approximately 100 °C
120
higher than that of a normal bypass diode with the same current. If the surface temperature
121
exceeds 210 °C, the junction box in the PV module will melt, thereby resulting in a fire. The
122
different current dependence of resistance for the diodes reveals that the high surface
123
temperature of the damaged bypass diode is attributed to its boosted resistance. The
124
resistance of a PV module with a damaged bypass diode increased linearly, and the reverse
EP
TE D
M AN U
height (in eV), and ‘n’ is the ideality factor.
AC C
103
ACCEPTED MANUSCRIPT 125
current increased. But the resistance of the normal diode was constant up to 10 A. To estimate
126
the origin of the high surface temperature of the damaged diode, a PV module with a
127
damaged diode was investigated using a circuit model.
128
3. PV module modeling and simulations
130
3.1 Modeling of PV module
131
The model of the PV -cell consisted of a diode, two resistances, and a current source, as
132
shown in Fig. 3. The typical I-V characteristics of a PV cell are given by the following
133
equation [21], [22] and [23].
134
= − exp #
− 1,- −
SC
()* +
$%& ' %&.
M AN U
135
$%& '
RI PT
129
(3)
where and are the photovoltaic and saturation current, respectively, / is the electron
137
charge, k is the Boltzmann constant, T is the temperature of the p-n junction, 0 is a series
138
The typical I-V characteristics of a PV array, where PV cells are interconnected in a parallel-
139
series, are given by the following equation [24].
140
= 2 − 2 3 − 1
141
If there is considerable parallel and series resistance, I-V characteristics can be expressed as
142
marked in [25], [26] and [27].
143
144 145
TE D
resistance, 0 is a shunt resistance, and 1 is the diode ideality constant.
&
(4)
EP
AC C
136
= 2 − 2 4 5
7 6 $%& ' & 9 78
3&
− 1:; − 4
7
$%& ' &
78 78 %&. 7&
;
(5)
where 2 and 2 are the number of module strings in parallel and the number of modules in each series string. Fig. 4 and Fig. 5 show equivalent circuits of one PV module with 33 %
146
shading and one with a damaged bypass diode, respectively.
147
Assuming that there are no electrical elements except for thermal voltage, the equivalent
148
circuit under partial shading is designed by considering a thermal voltage drop, as illustrated
ACCEPTED MANUSCRIPT 149
in Fig. 4. The voltage of the PV module (V) is given by
150
= 221
152 153
'
= Gexp @
'F
(6)
@ H3 $'%& $ IH3A
AJ − 1
I.
RI PT
151
V = <= + < = 2@2< − 20 A −
(7)
where N is the number of series cells in parallel connected to one bypass diode, and
156
solving Eq. (6)
SC
155
As the terminal voltage of a PV module comprised of 2 cells is equal to the terminal voltage
of the 2 shunt resistance, 2< is equal to 20 . Hence, of Eq. (8) is obtained by
M AN U
154
H $'% $ IH & 3 3 %&.
=
158
The forward voltage drop of the bypass diode is given by the Schottky barrier diode equation.
159
160
ln ' + 1 '
F
TE D
157
Therefore, the output current at Node A is given as follows, ()*
162
V
EP
(9)
()*
^ H3$'%& $ + [ ] Z − 1] − %&. ] \ Y
TU $= VF W X 3
AC C
161
O S H3 $'%& $ + R N = L − Nexp Q I. N M P
(8)
V
TU $= VF W 3
(10)
163
Meanwhile, in the case of a PV module with a damaged bypass diode, an equivalent circuit
164
can be modified as represented in Fig. 5. The voltage of a PV module incorporating a
165
damaged diode can be expressed as follows.
166
< = 2N@n
F
(11)
ACCEPTED MANUSCRIPT 167
' % H3 $'%& ` `H3
= exp #
,- − 1
I.
(12)
169
Similar to the case of a partially shaded module, the terminal voltage (including 2 PV cells)
170
follows.
171
is equal to the terminal voltage of 2 shunt resistance. Therefore, Eq. (13) is represented as < = 22 0 − 220 + _ 0_
RI PT
168
H $'% '` %`H & 3 3 %&.
=
173
The current equations at nodes A and B in Fig. 5 are as follows.
174
_ = L −
SC
172
(13) (14)
(15)
= L − −
176
The output current of a PV module under one bypass diode failure is represented by solving
177
Eq. (12), (14), (15) and (16).
H $'% '` %`H & 3 exp # 3
, I.
M AN U
175
= L −
179
Eq. (17) show the I-V characteristic of a working PV module with a damaged bypass diode.
180
power grid. If the PV plant is not operated, the equation of output current is zero. The
183 184
(17)
The output current in PV plants only flows when the system is properly connected to a = L
EP
current of a damaged bypass diode will then be as follows.
AC C
182
TE D
178
181
− 1- −
H $'% @@V&abVAc` A & 3 d7 %&.
(16)
(18)
185
3.2 Simulation of shaded PV module and one with a damaged bypass diode
186
The characteristics of a PV module for the simulation are described in Table 3. In this
187
simulation, the PV module consists of 60 cells. In addition, the bypass diodes are connected
188
to every 20 cells in the reverse direction, which is the same as the equivalent circuit model, as
189
shown in Fig. 6.
ACCEPTED MANUSCRIPT In the partially shaded PV module, 20 cells are shaded (33% shading) and the other 40 cells
191
are unshaded, and have three properly working bypass diodes. On the other hand, in the PV
192
module with a damaged bypass diode, the PV module consists of 60 unshaded cells with a
193
short circuited bypass diode, as shown in Fig. 6. The resistance of the damaged bypass diode
194
varies by the diode current in the reverse direction as mentioned earlier. Assuming that the
195
maximum operating current of the shaded PV module equals that of the module with the
196
damaged bypass diode, the maximum operating current, 8.44 A, is derived from the shaded
197
condition. Hence, the calculated current that flows to the damaged bypass diode is 0.75 A.
201 202 203 204 205 206 207
(_ A is 0.75 A, the resistance of the damaged bypass diode is 0.05Ω. Therefore, the resistance (0_ A for the simulation, is defined as 0.05Ω.
SC
200
current is derived by interpolation of the measured data, as shown in Fig 7. When the current
M AN U
199
For precise calculation of the current in the damaged bypass diode, the 0_ for reverse
o = 1.5109 =i A, the Boltzmann constant j = 1.381 ∙ 10n Hp , the electron charge The value of each variable for the simulation is given as follows: the saturation current
/ = 1.602 ∙ 10=r @stuvtwxA, the temperature = 298 K, the Diode ideality constant 1 = 1, the shunt resistance of a cell 0 = 7.5 Ω, the series resistance of a cell 0 =
0.007Ω, the series number of a cell in parallel connected to one bypass diode 2 = 20, and
TE D
198
RI PT
190
the resistance of the damaged bypass diode 0_ = 0.05 Ω.
I-V curves of Eq. (10) and (17) are solved by the Lambert W-function, using MatLab tools
209
[28] and [29].
210
Fig. 8 shows the I-V as well as the P-V characteristics of the 33 % shaded module and the
211
module with the damaged bypass diode under the normal stand test condition (STC).
212
Comparing the 33 % shading condition to the case with the damaged bypass diode in Fig. 5,
213
their I-V characteristics are similar, except for a slight difference in the open circuit voltage.
214
This is attributed to the increased series resistance in the 33 % shaded module.
215
3.3 Measured results for I-V
216
Fig. 9 shows the I-V characteristics of each condition: 33 % shading and a damaged bypass
217
diode. Also, the performance of the PV module without shading and with a damaged bypass
218
diode was compared. The results for the PV module in the STC experiment are summarized
AC C
EP
208
ACCEPTED MANUSCRIPT 219
in Table 4. Here the module consisted of 60 silicon PV cells with three bypass diodes, one
220
connected to every 20 cells. The measured results of the I-V curve were consistent with the
221
simulation results, which indicate that the efficiency of the two cases is in the same range.
223
Here, the slightly decreased < L originated from the thermal voltage of the bypass diode
224
difficult to judge between shading and bypass diode failure based on the I-V characteristics.
225
Partial shading caused by clouds and trees is generated instantly or at a specified time.
226
However, the PV module with bypass failure always shows a voltage drop compared to a
227
normal case.
RI PT
during forward conduction. Through the simulation and empirical results, we found that it is
SC
222
228
Table 4 shows the data obtained by measurement for each condition (normal, 33% shading,
230
and one damaged bypass diode). The module voltage under conditions of shading and one
231
bypass diode failure dropped by 33 % compared to the normal condition. However, the open
232
circuit voltage for the PV module with one bypass diode failure was 0.5 V higher and this
233
difference in voltage is the forward voltage drop of the diode.
234
4. Analysis of characteristics PV modules under shading and with bypass diode failure
235
4.1 Diode current and output current of PV module on each conditions
236
Fig. 10 shows the current flowing sub-circuit in the PV module. Under partial shading, the
237
output current ( ) in the PV module flows through the bypass diode in the forward direction.
240 241 242
TE D
EP
239
On the other hand, the current flow in the PV module with a damaged bypass diode, marked
in Fig. 10(b), is different from that of the module under a shading condition. The values are
AC C
238
M AN U
229
similar for the two cases during PV module operation.
The current to bypass diode (_ ), as expressed in Eq. (15), is smaller than that of the bypass
diode under shading in the normally working PV system, as shown in Fig. 10. Because the
244
temperature of the diode (_{ | ) is proportional to _ , the damaged bypass diode is thus less
245
However, the closed electrical loop that occurs due to the damaged bypass diode enables
243
heated (5 °C lower) compared to a regular bypass diode of a partially shaded PV module.
247
continuous L flow in the PV array. In particular, this current may elevate the temperature of
248
Meanwhile, if the PV system is not operating, the damaged bypass diode emits a large
246
the PV cell (}|~~ ), which gives rise to degradation of specific cells, akin to hot spots.
ACCEPTED MANUSCRIPT
250
amount of heat because the L generated by the PV module flows in the reverse direction, as
251
damaged bypass diode may cause fire in the PV system in the case of high irradiance.
249
252
given in Eq. (18).
Because the L is dependent on the light intensity, overheating of the
Consequently, the }|~~ and _{ | should be different depending on the origin of failure.
RI PT
253
4.2 Thermal properties of PV module under outdoor condition
255
4.2.1 Characteristics of PV module with damaged bypass diode
256
For the experiment with the module with a damaged bypass diode, we established a test bed,
257
as seen in Fig. 11 (B). Here, the PV module consisted of 60 cells with three bypass diodes in
258
a junction box, and one of them was damaged. Moreover, the test bed consisted of 10
259
modules (260 Wp) connected in series and controlled by a 3 kW inverter. To analyze the
260
M AN U
SC
254
261
electrical and thermal properties of the PV module with the damaged bypass diode, the
262
was measured. Moreover, }|~~ of the PV modules was checked using a thermos graphic camera (FILER A300). For the measuring conditions, the intensity of solar light was 900 -
263
910 W/m2 and the ambient temperature was 15 - 20 °C. Specifications of the installed PV
265 266
module in the test bed are summarized in Table 3. While the PV system was operating, the _
TE D
264
of the bypass diode was approximately 1 A, which is the same as the value obtained by subtracting from L of the PV module.
269
bypass diode was analyzed after being taken with a thermos graphic camera (FILER A300).
270
shown in Fig. 13 and 14, respectively. In Fig. 13, the }|~~ connected to the damaged bypass
EP
268
Fig. 12 exhibits the }|~~ of the test bed module having a damaged bypass diode (left side of
267
figure) in working mode. For comparison, the thermal image of a PV module with a normal
272
diode was higher than that of other points. In particular, at a specific point having the highest
AC C
271
The }|~~ at each point in the PV module under solar irradiance as a function of time, is
274
0 , it was observed that }|~~ was elevated by 15 - 20 °C. We believe that this can be
275
the PV module with the damaged bypass diode will undergo accelerated degradation
276
compared to other modules.
273
attributed to the current flowing in a closed loop originating from the damaged diode. Hence,
277
Fig. 15 shows the current of the damaged bypass diode, and the output current of the PV
278
system. In the figure, the output and damaged bypass diode currents are approximately 7.5
ACCEPTED MANUSCRIPT and 1 A, respectively. In fact, the current through the bypass diode measured outdoors was
280
different from the simulations results. In the case of simulation, the current of damaged diode
281
0.75 A based on the Eq. (15) was calculated under standard test condition, where the module
282
was under normally incident light 1000 W/m2 at the temperature 25 °C. Moreover, the losses
283
from invertor system and non-ideal circuit including resistance were ignored in the simulation.
284
As a result, the experimentally derived current in the diode is a little bit higher than that of
285
theoretical expectation. Through the experiment, it was verified that the diode current during
286
operation of the PV system was the same that given by Eq. (15).
287
From the results of this experiment, it is possible to predict the temperature dependence of a
288
junction box with a damaged bypass diode. Because of the broken diode, the efficiency of the
289
PV module is lower than that of a module with normal working diodes, and is caused by 33 %
290
reduced output voltage. Furthermore, degradation of the PV will be accelerated due to the
291
elevated surface temperature.
292
M AN U
SC
RI PT
279
4.2.2 Characteristics of PV modules under partial shading and with damaged bypass diode
294
Fig. 16 shows a thermal image of PV modules under partial shading and with a damaged
295
bypass diode during operation of the PV system. The left module in the figure is partially
296
shaded, whereas the center module has a damaged diode. In both the junction box and at the
297
surface of the module, the temperature of the PV module with the damaged diode was higher
298
than that of the partially shaded PV module.
299
The inner temperature in the junction box ( ) was monitored for 2500 seconds to
301 302 303 304 305 306 307 308
EP
AC C
300
TE D
293
determine thermal-light intensity characteristics of the PV module under non- and normaloperating modes. Fig. 17 reveals that was almost constant at 35 – 40 (°C). In contrast,
in the case with the damaged bypass diode, the maximum temperature was 95 (°C) under irradiance of 550 (W/m2). Low variance of regarding the light intensity was observed
in the case of partial shading, while the case with a damaged bypass diode exhibited high variation of from 50 - 100 (°C). Here, it should be pointed out that such a high induced by a damaged diode will lead to a fire in the system.
On the other hand, the values of the two cases were similar regardless of the origin of
low performance, as shown in Fig. 18, when the PV module was operating. The value
ACCEPTED MANUSCRIPT 309 310
of the partially shaded module was 5 °C higher than the with a damaged bypass diode.
This originated from the low current flow into the damaged bypass diode during PV system
311
operation. It will thus be possible to ascertain the origin of faulty PV modules, even though
312
they exhibit similar performance. Furthermore, a possible advantage attainable from
314
monitoring and }|~~ is that we can determine the origin of low performance, whether
RI PT
313
the PV system is operating or not. 5. Conclusions
316
This study was carried out to investigate the characteristics of PV modules under partial
317
shading and with a damaged bypass diode, respectively, by employing simulation of the
318
equivalent circuit model and by empirically measuring the temperature of the PV modules. In
319
the simulation results, the I-V curves of the two cases were similar, and thus it was difficult to
320
clearly distinguish them. However, the thermal characteristics of a bypass diode in a non-
321
operating module allowed us to distinguish between the two cases, because the temperature
322
of the damaged bypass diode is very high, at approximately 200 °C, enough to cause a fire or
323
melt the junction box. Therefore, we believe that effective detection of fault in the bypass
324
diodes is very important. This study can provide a guideline for ensuring a stable PV system.
325
Through this study, we verified the differences in thermal characteristics of a bypass diode
326
under a shading condition (including dust and snow) and bypass diode failure. Below is a
327
summary of the findings of the study.
328
(1) The I-V curve and performance of a 33 % shaded PV module is similar to those of a
TE D
M AN U
SC
315
331
(2) While the PV inverter is operating, the inner temperature of the junction box of the PV
332
module under shading is slightly higher (5 °C) than that of the PV module with a
333
damaged bypass diode, as a result of similar current flowing through the bypass diode in
334
both cases.
EP
330
module incorporating a damaged bypass diode, of which the maximum power and < L is 33% lower than those of a normally operating module.
AC C
329
335
(3) On the other hand, when the PV module is not operating, the current through the damaged
336
bypass diode will short-circuits, thereby resulting in increased temperature inside the
337
junction box; to a level that may be high enough to melt the box and the system, cause a
338
fire, and accelerate degradation of the PV cell.
ACCEPTED MANUSCRIPT 339
(4) Therefore, it is necessary to detect failure of a bypass diode in PV modules, in order to
340
prevent fires. Moreover, it is possible to distinguish shading or bypass diode failure in a
341
PV module by checking the temperature characteristics of the module in operation and
342
non-operation modes. Acknowledgment
344
This work was supported by a grant from the Renewable Energy of Korea Institute of Energy
345
Technology Evaluation and Planning (KETEP) and the Standardization Certification of Korea
346
Energy Agency (KEA) funded by the Korean Ministry of Trade, Industry and Energy.
347
(Project No: 20153010011980, 20143010011820, 71000106)
348
References
349
[1] Silvestre S, Boronat A, Chouder A. Study of bypass diodes configuration on PV modules.
350
Applied Energy 2009;86:1632-1640
351
[2] Alberto Dolara, George Cristian Lazaroiu, Sonia Leva, Giampaolo Manzolini,
352
Experimental investigation of partial shading scenarios on PV(photovoltaic) modules, Energy
353
2013;55:466-475
354
[3] T. H. Jung, J. W. Go, G. H. Kwang and H. K. Ahn, Output Characteristics of PV module
355
Considering Partially Reverse Biased Conditions, Solar Energy 2103; 92:214-220
356
[4] Adria E. Brooks, Daniel Cormode, Alexander D. Cronin, Elsa Kam-Lum, PV System
357
Power Loss and Module Damage due to Partial Shade and Bypass Diode Failure Depend on
358
Cell Behavior in Reverse Bias, Photovoltaic Specialist Conference (PVSC), IEEE 2015; 42
359
[5] Koray Sener Parlak, PV array reconfiguration method under partial shading conditions,
360
Electrical Power and Energy Systems 2014;63:713-721
361
[6]
362
NagamaniChilakapati, Positioning of PV panels for reduction in line losses and mismatch
363
losses in PV array, Renewable Energy 2015;78:264-275
364
[7] Santolo Daliento, Fabio Di Napoli, Pierluigi Guerriero, Vincenzo d’Alessandro, A
365
modified bypass circuit for improved hot spot reliability of solar panels subject to partial
366
shading, Solar Energy 2016;134:211-218
AC C
EP
TE D
M AN U
SC
RI PT
343
Srinivasa
Rao
Potnuru,
Dinesh
Pattabiraman,
Saravana
Ilango
Ganesan,
ACCEPTED MANUSCRIPT [8] Tey Kok Soon, Saad Mekhilef, A Fast-Converging MPPT Technique for Photovoltaic
368
System under Fast-Varying Solar Irradiation and Load Resistance, IEEE 2015; 11:176-186
369
[9] M. Seyedmahmoudian, B. Horan, T. Kok soon, R. Rahmani, A. Muang Than Oo, S.
370
mekhilef, A. Stojcevski, State of the Art Artificial Intelligence-based MPPT Techniques for
371
Mitigating Partial Shading Effects on PV systems – A review, Renewable and Sustainable
372
Energy Reviews, 2016; 64:435-455
373
[10] Mohammadmehdi Seyedmahmoudian, Rasoul Rahmani, Saad Mekhilef, Amanullah
374
Maung Than Oo, Alex Stojcevski, Tey Kok Soon, Alireza Safdari Ghandhari, Simulation and
375
Hardware Implementation of New Maximum Power Point Tracking Technique for Partially
376
Shaded PV System Using Hybrid DEPSO Method, IEEE Transactions on Sustainable Energy,
377
2015;6:850-862
378
[11] Yaw-Juen Wang, Po-Chun Hsu, An investigation on partial shading of PV modules with
379
different connection configuration of PV cells, Energy 2011;35:3069-6078
380
[12] Pieter Bauwens, Jan Doutreloigne, Reducing partial shading power loss with an
381
intergrated Smart Bypass, Solar Energy 2014;103:134-142
382
[13] N.A. Al-Rawi, M.M. Al-Kaisi, D. Asfer, Reliability of photovoltaic modules II.
383
Interconnections and bypass diodes effects. Solar Energy Materials and Solar Cells
384
1994;31:469-480
385
[14] Alonso-Garcia MC, Ruiz JM, Analysis and modelling the reverse characteristic of
386
photovoltaic cells. Solar Energy Materials and Solar Cells 2006;90:1105-1120
387
[15] Daniel Gomez Lorente, Simone Pedrazzi, Gabriele Zini, Alberto Dalla Rosa, Paolo
388
Tartarini, Mismatch losses in PV power plants, Solar Energy 2014;100:42-49
389
[16] Gough Yumu Lui, Richard Corkish, Comparison of Different Bypass Diode
390
Technologies, Solar 2011, the 49th AuSES Annual Conference 2011:1-10
391
[17] Haeberiln. H, Damages at Bypass Diodes by Induced Voltages and Currents in PV
392
modules Caused by Nearby Lightning Currents. 22nd European Photovoltaic Solar Energy
393
Conference, Milano, Italy, 2007
394
[18] Dhere. N, Shiradkar, N. Schneller E. and Gade V., The Reliability of Bypass Diode in
395
PV modules, Proc. SPIE 8825, 2013
AC C
EP
TE D
M AN U
SC
RI PT
367
ACCEPTED MANUSCRIPT [19] K. Kato, “PVRessQ!”: A Research Activity on Reliability of PV System from an user’s
397
viewpoint in Japan, Proc. Optics + Photonics 8112(SPIE, San Diego, California, USA, 2011)
398
[20]Hui Gao, Shunzhong Luo, Huaming Zhang, Heyi Wang, Zhonghua Fu, Demonstration,
399
radiation tolerance and design on a betavoltaic micropower, Energy 2013;51:116-122
400
[21] M.AlRashidi, M. AlHajri, K. El-Naggar, A. AlOthman, A new estimation approach for
401
determining the I-V characteristics of solar cells, Solar Energy 2011;85:1543-1550
402
[22] W. Gong, Z. Cai, Parameter extraction of solar cell models using repaired adaptive
403
differential evolution, Solar Energy 2014;94:209-220
404
[23] T. Ikegami, T. Maezono, F. Nakanishi, Y. Yamagata, K. Ebihara, Estimation of
405
equivalent circuit parameters of PV module and its application to optimal operation of PV
406
system, Solar Energy Materials & Solar Cells 2001;67:389-395
407
[24] K.H. Hussein, I. Muta, T. Hoshino, Osakada, Maximum photovoltaic power tracking: an
408
algorithm for rapidly changing atmospheric conditions, IEE Proceedings on Generation,
409
Transmission, Distribution 1995;142: 59–64
410
[25] M.B. Eteiba, E.T.El Shenawy, J.H. Shazly, A.Z.Hafez, A Photovoltaic(Cell, Module,
411
Array) Simulation and Monitoring Model using MATLAB/GUI Interface, International
412
Journal of Computer Applications 2013;69:14-28
413
[26] Villalva, Marcelo Gradella, Gazoli, Jonas Rafael, Filho andErnesto Ruppert.
414
Comprehensive approach to modeling and simulation of photovoltaic arrays, IEEE Trans.
415
Power Electron 2009; 24:1198–1208
416
[27] Moradi, M.H. andReisi,A.R. A hybrid maximum power point tracking method for
417
photovoltaic systems, Solar Energy 2011; 85: 2965-2976
418
[28] Amit Jain, Avinashi Kapoor, Exact analytical solutions of the parameters of real solar
419
cells using Lambert W-funcion, Solar Energy Materials & Solar Cells 2004; 81:269-277
420
[29] F. Ghani, M. Duke, Numerical determination of parasitic resistances of a solar cell using
421
the Lambert W-function, Solar Energy 2011; 85:2386-2394
422
AC C
EP
TE D
M AN U
SC
RI PT
396
ACCEPTED MANUSCRIPT 423
Table 1. Classification of reference by specific parameters Specific parameter
Purpose
Reconfiguration of PV arrays
Improve the output in shading and
[1], [2], [5], [6], [7]
or cell arrangements
reduce a loss
and [8]
Bypass diode circuit added
Reduce hot spot temperature and the
MOSFET
power loss
PV module model including shaded cell
Evaluate the power loss
[9] and [19]
To predict output of shading module
[10]
Comparison on power loss and impact [11]
MPPT algorithm and
Reduce power loss in irradiance
dc-dc
M AN U
Cell of reverse bias voltage
converter
variation
Damaged bypass diode
Analyze damaged diode
424 425
[12], [13] and [14] [16] and [18]
TE D
Table 2. Temperature and resistance according to current of diode Forward direction for normal diode
Reverse direction for damaged diode
Temperature
Resistance
Current
Temperature
Resistance
(A)
(°C)
(Ω)
(A)
(°C)
(Ω)
2
41.6
0.16
2
35.1
0.11
62.4
0.17
4
56.4
0.22
79.5
0.17
6
91.7
0.34
8
100.6
0.18
8
139.7
0.48
10
120.2
0.18
10
219.2
0.7
AC C
6
EP
Current
4
427
RI PT
PV module
[3] and [4]
SC
Impact of partial shading on
426
Reference
Table 3. Characteristics of the test PV module Maximum Power@ A
Maximum Power Point Voltage@< A Maximum Power Point Current@ A
260 W 30.70 V 8.62 A
ACCEPTED MANUSCRIPT Short Circuit Current@L A
9.16 A
Open Circuit Voltage@< L A
428
38 V
Table 4. I-V characteristics of PV modules under STC, one normal, one with 33% shading
430
and one with a damaged diode Characteristics
Normal module
measured
module
@ A
One damaged diode module
171.4
38
25.6
26
30.5
19.8
20.2
@A
9.18
9.17
9.18
8.49
7.69
8.6
175.99
433 434
AC C
EP
TE D
432
M AN U
L @A
SC
< L @
263.8
< @
431
33 % Shaded
RI PT
429
(a) Measured I-V curve for a normal and a damaged bypass diode
SC
RI PT
ACCEPTED MANUSCRIPT
435
437
(b) Calculated power loss of each bypass diode based on I-V curve
M AN U
436
Fig. 1. Characteristics of a bypass diode (SBX 1240/Diotec Semiconductor AG)
440
AC C
439
EP
TE D
438
Fig. 2. Measured surface temperature of a diode
441 442
Fig. 3. One-diode equivalent-circuit model of a PV-cell
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. Equivalent circuit of a PV module under partial shading
445 446 447
AC C
EP
TE D
444
M AN U
443
Fig. 5. Equivalent circuit of a PV module with failure of one bypass diode
RI PT
ACCEPTED MANUSCRIPT
448
Fig. 6. Schematic diagram of partially shaded PV module and an unshaded model
450
representing a damaged bypass diode
452 453
Fig. 7. Estimated resistance of 0_ using resistance measured in the reverse direction
AC C
451
EP
TE D
M AN U
SC
449
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
454 455
456 457 458 459
AC C
EP
TE D
(a) I-V curve
(b) P-V curve Fig. 8. Results of simulation
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
460 461
Fig. 9. I-V curve characteristics of a shaded PV module and one with a damaged bypass
462
diode
(a)
AC C
465
EP
464
TE D
463
466 467
(b)
468
Fig. 10 Current flow of PV module in each condition: (a) Current flow of PV module under
469
partial shading, (b) Current flow of PV module with a damaged bypass diode
470
RI PT
ACCEPTED MANUSCRIPT
471
(b)
Fig. 11. (a) Schematic diagram and (b) Tests bed configuration of PV modules
474 475
AC C
EP
TE D
M AN U
473
(a)
SC
472
Fig. 12 Thermal image of }|~~ in a PV module
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 13 Surface temperature at each point in the PV module shown in Fig. 12
478 479
AC C
EP
TE D
477
M AN U
476
Fig. 14. Intensity of irradiation during the measurement
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 15 Current in PV module with damaged diode while the PV system operated
482 483
AC C
EP
TE D
481
M AN U
480
Fig. 16 Thermal image of PV modules during inverter operation
484
(b) The irradiance condition
Fig. 17 Temperature in the junction box when the system is not operating
489 490
(a) The temperature in junction -box
EP
488
(b) The irradiance condition
Fig. 18 Temperature in the junction box when the system is operating
AC C
487
TE D
M AN U
486
(a) The temperature in junction box
SC
485
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights •
It proposed the mathematical models of PV modules under shading and with damaged bypass diode. Electric and thermal characteristics of PV modules under shading and with damaged bypass diode were discussed.
It is possible to distinguish shading or bypass diode failure in PV modules using the
EP
TE D
M AN U
SC
thermal characteristics.
AC C
•
RI PT
•
S0360-5442(17)30595-9
DOI:
10.1016/j.energy.2017.04.030
Reference:
EGY 10670
To appear in:
Energy
Received Date: 21 November 2016 Revised Date:
7 April 2017
Accepted Date: 8 April 2017
Please cite this article as: Ko SW, Ju YC, Hwang HM, So JH, Jung Y-S, Song H-J, Song H-e, Kim S-H, Kang GH, Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode, Energy (2017), doi: 10.1016/j.energy.2017.04.030. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 2 3
Electric and thermal characteristics of photovoltaic modules under partial shading and with a damaged bypass diode
RI PT
4
Abstract:
6
In this paper, characteristics of PV (photovoltaic) modules under partial shading or with a
7
damaged bypass diode in the junction box were evaluated by comparing a theoretical model
8
and empirical data. For the electrical analysis of the current- voltage (I-V) curve of each
9
module, a PV module with one diode cell model was proposed, and the model closely
10
matched the empirical results. The damaged bypass diode was replaced with an element of
11
resistance in the simulation model. The calculation shows that the open circuit voltage of the
12
PV module with a damaged bypass diode was slightly higher than that of a PV module under
13
shading conditions while the PV system was operating. The I-V curve of each module
14
obtained with the solar simulator was similar to the results of the simulation. From the results
15
of field testing each PV module, when the PV system was operating in connection with the
16
power grid, the internal temperature of the junction box connected to the shaded PV module
17
was 5 °C higher than that of the PV module with the damaged bypass diode. Furthermore
18
when the PV system was not operating, the internal temperature of the junction box in the PV
19
module with the damaged bypass diode was extremely high. This condition caused a short-
20
circuit and the surface temperature of the damaged bypass diode reached 219 °C. In this
21
paper, we theoretically and empirically analyzed the characteristics of a shaded PV module
22
and a module with a damaged bypass diode.
M AN U
TE D
EP
AC C
23
SC
5
24
Keywords: partial shading, damaged bypass diode, photovoltaic module, open circuit voltage,
25
short circuit current
26 27
1. Introduction
28
A bypass diode is widely used in PV (photovoltaic) modules to prevent fire caused by hot
29
spots as well as to decrease energy losses during shading and mismatching. Bypass diodes,
ACCEPTED MANUSCRIPT connected in reverse bias between a solar cell strings positive and negative output
31
terminals, generally is used per small group of series cells. For instance, in the case of a
32
250 W PV module consisting of 60 solar cells, the bypass diodes are integrated into the
33
PV module in parallel for every 20 solar cells. Various studies on bypass diode
34
configurations, summarized in Table 1, have improved the efficiency of PV system under
35
mismatch condition induced by partially shading, abnormally working electrical
36
components and so on. The power loss in a partial shaded module is inevitable as
37
presented as proven in theoretically calculation [1] and [2] and empirically demonstrated
38
its output characteristics [3] and [4]. For overcoming this issue, the method for additional
39
bypass diode configuration of array with a series-connected PV module [5] was proposed.
40
Additionally, an optimal configuration to suppress line and mismatch losses was
41
formulated [6] and [7]. Moreover, various studies were conducted on the electrical
42
variation and impact on PV module and system. A method of reducing output loss using
43
MPPT (Maximum Power Point Tracking) algorithm and dc-dc converter has been studied
44
[8] and various MPPT techniques have been reviewed to mitigate partial shading effects
45
in PV systems [9]. Recently, new MPPT technique using Hybrid DESPO (Differential
46
Evolutionary algorithm and Particle Swarm Optimization) method is developed [10].
47
Among these techniques, the introduction of bypass diode to PV module is a simple yet
48
effective way to circumvent the deficiency in mismatched modules. Previous works about
49
cells and bypass diode in [11] have proved that its installation is an efficient way for
50
improving maximum power and reliability of PV module. Additionally, PV module using
51
a smart bypass was suggested in [12] to minimize the power loss. The influence of the
52
bypass diode can be observed with the change in shape of the I-V (current-voltage) and P-
53
V (power-voltage) curves of PV module. Previous researches, [13] and [14], investigated
54
the characterization of the bypass diode in PV systems or arrays under an influence of
55
mismatch. Normally, mismatch losses in a PV module is occurred by a partial shading or
56
a damaged bypass diode. When the PV is under partial shading, the attainable power from
57
PV cell array is determined by shaded cell. However, the normally working bypass diode
58
makes a new path, thereby resulting in reduced losses. On the other hand, if a bypass
59
diode of the PV module is damaged, the sub-circuit of the PV module will enter a short-
60
circuit state. As a result of such a mismatching in the system, a failure to detect the
61
damaged bypass diode will lead to significant system losses, as reported in [15]. For
62
instance, the voltage of PV module with a damaged bypass diode will drop by 33 %
AC C
EP
TE D
M AN U
SC
RI PT
30
ACCEPTED MANUSCRIPT compared to that of one employing a normally working diode. In the case of a large PV
64
system with a centralized inverter system, the system loss will be larger than that of single
65
PV module, and may even worse depending on the number of series and parallel
66
connections. Also, the failure of a bypass diode on a PV module is severely dangerous
67
because the surface temperature rises dramatically: enough to cause a fire. The Schottky
68
type diode, widely used in PV module, has a lower surge reliability, a lower thermal
69
resistance and high leakage current [16]. For this reason, researches about mechanism of
70
failure in Schottky diode by surge voltage have been conducted [17] and [18].
71
Nevertheless, bypass diode failures, that occasionally occur owing to its low reliability,
72
are difficult to detect because the PV system works normally even with bypass diode in
73
trouble. Until now, in the conventional diagnosis method, the damaged bypass diodes can
74
be only detected by measuring the terminal voltage of bypass diode in junction box
75
during regular maintenance at PV plants using diagnostic sensors. Unfortunately, the
76
failure of a PV system is only recognized, when PV modules show a severe mismatch that
77
induces power loss in the system. According to a PV system at AIST (Advanced
78
Industrial Science and Technology) in Japan [19], where 53 units of 4 kWp PV arrays
79
were operated, only few malfunction diodes were detected among hundreds of PV
80
module with them. The inspection of bypass diodes of 1272 PV modules with 180Wp
81
showed that 47 % of the modules had damaged ones. However, burn marks on sub-
82
modules was observed only 3% of defective PV modules. Moreover, the P-V curve of PV
83
module with damaged diode is similar to that of module under mismatching condition,
84
likes as partially shading. Thus, it is not easy to manage the power loss due to uncertainty
85
of its origin. The loss from shaded module is a temporary one, while the failure of bypass
86
diode causes a continuous loss. However, the systematically exploration of each condition,
87
PV module with shading or failure of bypass diode, has not been conducted so far.
88
Therefore, effective diagnosis method to find out the origin of mismatch is imperative in
89
PV system. In this paper, we demonstrate a new method to discriminate damaged bypass
90
diode in PV module. Electric circuit modeling and simulation were performed to
91
distinguish between a shaded PV module and a module with one damaged bypass diode.
92
Moreover, thermo-electric properties of bypass diode in various conditions were
93
empirically investigated. Through the analysis of each condition of the PV module using
94
experimental data, elements for a fault diagnosis of a bypass diode in a PV module were
95
proposed by monitoring its temperature under operating condition.
AC C
EP
TE D
M AN U
SC
RI PT
63
ACCEPTED MANUSCRIPT 96
2. Experiments on the thermal characteristics of a bypass diode
97
The current through a Schottky barrier diode under a forward bias ‘V’ is given by the
98
following relation [20].
99
=
102
− 1
(1)
where , the reverse saturation current, is given as follows. = ∗∗
∅
RI PT
101
(2)
Here, Ad is the diode area, A∗∗ is the effective Richardson constant (12 A cm−2 K−2), T is the
SC
100
104
temperature, k is the Boltzmann constant, q is the electronic charge, ∅ is the energy barrier
105
Fig. 1 shows the I-V and P-V characteristics of normal and damaged bypass diodes measured
106
using a source meter. The difference in power consumption between a normal and damaged
107
diode is 5.41 W with forward current of 10 A. The bypass diode blocks the current under the
108
normal working condition of a PV module. Meanwhile, in shaded PV modules, the direction
109
of current in the bypass diode is forward. In contrast, in the case with a damaged bypass
110
diode, the current flows in the reverse direction regardless of operation of the PV system.
111
Hence, the thermal characteristics of bypass diodes as a function of current are analyzed
112
under the forward and reverse bias. Here, the surface temperature and the resistance of the
113
bypass diode were measured under various currents by using thermal imaging and resistance
114
measuring equipment, when the temperature was saturated.
115
Measurement results are shown in Fig. 2 and summarized in Table 2. Fig. 2 represents the
116
surface temperature of normal and damaged diodes as a function of current. The surface
117
temperature of the normal diode increased linearly as the forward current increased. On the
118
other hand, the temperature of the damaged diode rose exponentially according to the reverse
119
current and eventually the temperature exceeded 200 °C at 10 A, being approximately 100 °C
120
higher than that of a normal bypass diode with the same current. If the surface temperature
121
exceeds 210 °C, the junction box in the PV module will melt, thereby resulting in a fire. The
122
different current dependence of resistance for the diodes reveals that the high surface
123
temperature of the damaged bypass diode is attributed to its boosted resistance. The
124
resistance of a PV module with a damaged bypass diode increased linearly, and the reverse
EP
TE D
M AN U
height (in eV), and ‘n’ is the ideality factor.
AC C
103
ACCEPTED MANUSCRIPT 125
current increased. But the resistance of the normal diode was constant up to 10 A. To estimate
126
the origin of the high surface temperature of the damaged diode, a PV module with a
127
damaged diode was investigated using a circuit model.
128
3. PV module modeling and simulations
130
3.1 Modeling of PV module
131
The model of the PV -cell consisted of a diode, two resistances, and a current source, as
132
shown in Fig. 3. The typical I-V characteristics of a PV cell are given by the following
133
equation [21], [22] and [23].
134
= − exp #
− 1,- −
SC
()* +
$%& ' %&.
M AN U
135
$%& '
RI PT
129
(3)
where and are the photovoltaic and saturation current, respectively, / is the electron
137
charge, k is the Boltzmann constant, T is the temperature of the p-n junction, 0 is a series
138
The typical I-V characteristics of a PV array, where PV cells are interconnected in a parallel-
139
series, are given by the following equation [24].
140
= 2 − 2 3 − 1
141
If there is considerable parallel and series resistance, I-V characteristics can be expressed as
142
marked in [25], [26] and [27].
143
144 145
TE D
resistance, 0 is a shunt resistance, and 1 is the diode ideality constant.
&
(4)
EP
AC C
136
= 2 − 2 4 5
7 6 $%& ' & 9 78
3&
− 1:; − 4
7
$%& ' &
78 78 %&. 7&
;
(5)
where 2 and 2 are the number of module strings in parallel and the number of modules in each series string. Fig. 4 and Fig. 5 show equivalent circuits of one PV module with 33 %
146
shading and one with a damaged bypass diode, respectively.
147
Assuming that there are no electrical elements except for thermal voltage, the equivalent
148
circuit under partial shading is designed by considering a thermal voltage drop, as illustrated
ACCEPTED MANUSCRIPT 149
in Fig. 4. The voltage of the PV module (V) is given by
150
= 221
152 153
'
= Gexp @
'F
(6)
@ H3 $'%& $ IH3A
AJ − 1
I.
RI PT
151
V = <= + < = 2@2< − 20 A −
(7)
where N is the number of series cells in parallel connected to one bypass diode, and
156
solving Eq. (6)
SC
155
As the terminal voltage of a PV module comprised of 2 cells is equal to the terminal voltage
of the 2 shunt resistance, 2< is equal to 20 . Hence, of Eq. (8) is obtained by
M AN U
154
H $'% $ IH & 3 3 %&.
=
158
The forward voltage drop of the bypass diode is given by the Schottky barrier diode equation.
159
160
ln ' + 1 '
F
TE D
157
Therefore, the output current at Node A is given as follows, ()*
162
V
EP
(9)
()*
^ H3$'%& $ + [ ] Z − 1] − %&. ] \ Y
TU $= VF W X 3
AC C
161
O S H3 $'%& $ + R N = L − Nexp Q I. N M P
(8)
V
TU $= VF W 3
(10)
163
Meanwhile, in the case of a PV module with a damaged bypass diode, an equivalent circuit
164
can be modified as represented in Fig. 5. The voltage of a PV module incorporating a
165
damaged diode can be expressed as follows.
166
< = 2N@n
F
(11)
ACCEPTED MANUSCRIPT 167
' % H3 $'%& ` `H3
= exp #
,- − 1
I.
(12)
169
Similar to the case of a partially shaded module, the terminal voltage (including 2 PV cells)
170
follows.
171
is equal to the terminal voltage of 2 shunt resistance. Therefore, Eq. (13) is represented as < = 22 0 − 220 + _ 0_
RI PT
168
H $'% '` %`H & 3 3 %&.
=
173
The current equations at nodes A and B in Fig. 5 are as follows.
174
_ = L −
SC
172
(13) (14)
(15)
= L − −
176
The output current of a PV module under one bypass diode failure is represented by solving
177
Eq. (12), (14), (15) and (16).
H $'% '` %`H & 3 exp # 3
, I.
M AN U
175
= L −
179
Eq. (17) show the I-V characteristic of a working PV module with a damaged bypass diode.
180
power grid. If the PV plant is not operated, the equation of output current is zero. The
183 184
(17)
The output current in PV plants only flows when the system is properly connected to a = L
EP
current of a damaged bypass diode will then be as follows.
AC C
182
TE D
178
181
− 1- −
H $'% @@V&abVAc` A & 3 d7 %&.
(16)
(18)
185
3.2 Simulation of shaded PV module and one with a damaged bypass diode
186
The characteristics of a PV module for the simulation are described in Table 3. In this
187
simulation, the PV module consists of 60 cells. In addition, the bypass diodes are connected
188
to every 20 cells in the reverse direction, which is the same as the equivalent circuit model, as
189
shown in Fig. 6.
ACCEPTED MANUSCRIPT In the partially shaded PV module, 20 cells are shaded (33% shading) and the other 40 cells
191
are unshaded, and have three properly working bypass diodes. On the other hand, in the PV
192
module with a damaged bypass diode, the PV module consists of 60 unshaded cells with a
193
short circuited bypass diode, as shown in Fig. 6. The resistance of the damaged bypass diode
194
varies by the diode current in the reverse direction as mentioned earlier. Assuming that the
195
maximum operating current of the shaded PV module equals that of the module with the
196
damaged bypass diode, the maximum operating current, 8.44 A, is derived from the shaded
197
condition. Hence, the calculated current that flows to the damaged bypass diode is 0.75 A.
201 202 203 204 205 206 207
(_ A is 0.75 A, the resistance of the damaged bypass diode is 0.05Ω. Therefore, the resistance (0_ A for the simulation, is defined as 0.05Ω.
SC
200
current is derived by interpolation of the measured data, as shown in Fig 7. When the current
M AN U
199
For precise calculation of the current in the damaged bypass diode, the 0_ for reverse
o = 1.5109 =i A, the Boltzmann constant j = 1.381 ∙ 10n Hp , the electron charge The value of each variable for the simulation is given as follows: the saturation current
/ = 1.602 ∙ 10=r @stuvtwxA, the temperature = 298 K, the Diode ideality constant 1 = 1, the shunt resistance of a cell 0 = 7.5 Ω, the series resistance of a cell 0 =
0.007Ω, the series number of a cell in parallel connected to one bypass diode 2 = 20, and
TE D
198
RI PT
190
the resistance of the damaged bypass diode 0_ = 0.05 Ω.
I-V curves of Eq. (10) and (17) are solved by the Lambert W-function, using MatLab tools
209
[28] and [29].
210
Fig. 8 shows the I-V as well as the P-V characteristics of the 33 % shaded module and the
211
module with the damaged bypass diode under the normal stand test condition (STC).
212
Comparing the 33 % shading condition to the case with the damaged bypass diode in Fig. 5,
213
their I-V characteristics are similar, except for a slight difference in the open circuit voltage.
214
This is attributed to the increased series resistance in the 33 % shaded module.
215
3.3 Measured results for I-V
216
Fig. 9 shows the I-V characteristics of each condition: 33 % shading and a damaged bypass
217
diode. Also, the performance of the PV module without shading and with a damaged bypass
218
diode was compared. The results for the PV module in the STC experiment are summarized
AC C
EP
208
ACCEPTED MANUSCRIPT 219
in Table 4. Here the module consisted of 60 silicon PV cells with three bypass diodes, one
220
connected to every 20 cells. The measured results of the I-V curve were consistent with the
221
simulation results, which indicate that the efficiency of the two cases is in the same range.
223
Here, the slightly decreased < L originated from the thermal voltage of the bypass diode
224
difficult to judge between shading and bypass diode failure based on the I-V characteristics.
225
Partial shading caused by clouds and trees is generated instantly or at a specified time.
226
However, the PV module with bypass failure always shows a voltage drop compared to a
227
normal case.
RI PT
during forward conduction. Through the simulation and empirical results, we found that it is
SC
222
228
Table 4 shows the data obtained by measurement for each condition (normal, 33% shading,
230
and one damaged bypass diode). The module voltage under conditions of shading and one
231
bypass diode failure dropped by 33 % compared to the normal condition. However, the open
232
circuit voltage for the PV module with one bypass diode failure was 0.5 V higher and this
233
difference in voltage is the forward voltage drop of the diode.
234
4. Analysis of characteristics PV modules under shading and with bypass diode failure
235
4.1 Diode current and output current of PV module on each conditions
236
Fig. 10 shows the current flowing sub-circuit in the PV module. Under partial shading, the
237
output current ( ) in the PV module flows through the bypass diode in the forward direction.
240 241 242
TE D
EP
239
On the other hand, the current flow in the PV module with a damaged bypass diode, marked
in Fig. 10(b), is different from that of the module under a shading condition. The values are
AC C
238
M AN U
229
similar for the two cases during PV module operation.
The current to bypass diode (_ ), as expressed in Eq. (15), is smaller than that of the bypass
diode under shading in the normally working PV system, as shown in Fig. 10. Because the
244
temperature of the diode (_{ | ) is proportional to _ , the damaged bypass diode is thus less
245
However, the closed electrical loop that occurs due to the damaged bypass diode enables
243
heated (5 °C lower) compared to a regular bypass diode of a partially shaded PV module.
247
continuous L flow in the PV array. In particular, this current may elevate the temperature of
248
Meanwhile, if the PV system is not operating, the damaged bypass diode emits a large
246
the PV cell (}|~~ ), which gives rise to degradation of specific cells, akin to hot spots.
ACCEPTED MANUSCRIPT
250
amount of heat because the L generated by the PV module flows in the reverse direction, as
251
damaged bypass diode may cause fire in the PV system in the case of high irradiance.
249
252
given in Eq. (18).
Because the L is dependent on the light intensity, overheating of the
Consequently, the }|~~ and _{ | should be different depending on the origin of failure.
RI PT
253
4.2 Thermal properties of PV module under outdoor condition
255
4.2.1 Characteristics of PV module with damaged bypass diode
256
For the experiment with the module with a damaged bypass diode, we established a test bed,
257
as seen in Fig. 11 (B). Here, the PV module consisted of 60 cells with three bypass diodes in
258
a junction box, and one of them was damaged. Moreover, the test bed consisted of 10
259
modules (260 Wp) connected in series and controlled by a 3 kW inverter. To analyze the
260
M AN U
SC
254
261
electrical and thermal properties of the PV module with the damaged bypass diode, the
262
was measured. Moreover, }|~~ of the PV modules was checked using a thermos graphic camera (FILER A300). For the measuring conditions, the intensity of solar light was 900 -
263
910 W/m2 and the ambient temperature was 15 - 20 °C. Specifications of the installed PV
265 266
module in the test bed are summarized in Table 3. While the PV system was operating, the _
TE D
264
of the bypass diode was approximately 1 A, which is the same as the value obtained by subtracting from L of the PV module.
269
bypass diode was analyzed after being taken with a thermos graphic camera (FILER A300).
270
shown in Fig. 13 and 14, respectively. In Fig. 13, the }|~~ connected to the damaged bypass
EP
268
Fig. 12 exhibits the }|~~ of the test bed module having a damaged bypass diode (left side of
267
figure) in working mode. For comparison, the thermal image of a PV module with a normal
272
diode was higher than that of other points. In particular, at a specific point having the highest
AC C
271
The }|~~ at each point in the PV module under solar irradiance as a function of time, is
274
0 , it was observed that }|~~ was elevated by 15 - 20 °C. We believe that this can be
275
the PV module with the damaged bypass diode will undergo accelerated degradation
276
compared to other modules.
273
attributed to the current flowing in a closed loop originating from the damaged diode. Hence,
277
Fig. 15 shows the current of the damaged bypass diode, and the output current of the PV
278
system. In the figure, the output and damaged bypass diode currents are approximately 7.5
ACCEPTED MANUSCRIPT and 1 A, respectively. In fact, the current through the bypass diode measured outdoors was
280
different from the simulations results. In the case of simulation, the current of damaged diode
281
0.75 A based on the Eq. (15) was calculated under standard test condition, where the module
282
was under normally incident light 1000 W/m2 at the temperature 25 °C. Moreover, the losses
283
from invertor system and non-ideal circuit including resistance were ignored in the simulation.
284
As a result, the experimentally derived current in the diode is a little bit higher than that of
285
theoretical expectation. Through the experiment, it was verified that the diode current during
286
operation of the PV system was the same that given by Eq. (15).
287
From the results of this experiment, it is possible to predict the temperature dependence of a
288
junction box with a damaged bypass diode. Because of the broken diode, the efficiency of the
289
PV module is lower than that of a module with normal working diodes, and is caused by 33 %
290
reduced output voltage. Furthermore, degradation of the PV will be accelerated due to the
291
elevated surface temperature.
292
M AN U
SC
RI PT
279
4.2.2 Characteristics of PV modules under partial shading and with damaged bypass diode
294
Fig. 16 shows a thermal image of PV modules under partial shading and with a damaged
295
bypass diode during operation of the PV system. The left module in the figure is partially
296
shaded, whereas the center module has a damaged diode. In both the junction box and at the
297
surface of the module, the temperature of the PV module with the damaged diode was higher
298
than that of the partially shaded PV module.
299
The inner temperature in the junction box ( ) was monitored for 2500 seconds to
301 302 303 304 305 306 307 308
EP
AC C
300
TE D
293
determine thermal-light intensity characteristics of the PV module under non- and normaloperating modes. Fig. 17 reveals that was almost constant at 35 – 40 (°C). In contrast,
in the case with the damaged bypass diode, the maximum temperature was 95 (°C) under irradiance of 550 (W/m2). Low variance of regarding the light intensity was observed
in the case of partial shading, while the case with a damaged bypass diode exhibited high variation of from 50 - 100 (°C). Here, it should be pointed out that such a high induced by a damaged diode will lead to a fire in the system.
On the other hand, the values of the two cases were similar regardless of the origin of
low performance, as shown in Fig. 18, when the PV module was operating. The value
ACCEPTED MANUSCRIPT 309 310
of the partially shaded module was 5 °C higher than the with a damaged bypass diode.
This originated from the low current flow into the damaged bypass diode during PV system
311
operation. It will thus be possible to ascertain the origin of faulty PV modules, even though
312
they exhibit similar performance. Furthermore, a possible advantage attainable from
314
monitoring and }|~~ is that we can determine the origin of low performance, whether
RI PT
313
the PV system is operating or not. 5. Conclusions
316
This study was carried out to investigate the characteristics of PV modules under partial
317
shading and with a damaged bypass diode, respectively, by employing simulation of the
318
equivalent circuit model and by empirically measuring the temperature of the PV modules. In
319
the simulation results, the I-V curves of the two cases were similar, and thus it was difficult to
320
clearly distinguish them. However, the thermal characteristics of a bypass diode in a non-
321
operating module allowed us to distinguish between the two cases, because the temperature
322
of the damaged bypass diode is very high, at approximately 200 °C, enough to cause a fire or
323
melt the junction box. Therefore, we believe that effective detection of fault in the bypass
324
diodes is very important. This study can provide a guideline for ensuring a stable PV system.
325
Through this study, we verified the differences in thermal characteristics of a bypass diode
326
under a shading condition (including dust and snow) and bypass diode failure. Below is a
327
summary of the findings of the study.
328
(1) The I-V curve and performance of a 33 % shaded PV module is similar to those of a
TE D
M AN U
SC
315
331
(2) While the PV inverter is operating, the inner temperature of the junction box of the PV
332
module under shading is slightly higher (5 °C) than that of the PV module with a
333
damaged bypass diode, as a result of similar current flowing through the bypass diode in
334
both cases.
EP
330
module incorporating a damaged bypass diode, of which the maximum power and < L is 33% lower than those of a normally operating module.
AC C
329
335
(3) On the other hand, when the PV module is not operating, the current through the damaged
336
bypass diode will short-circuits, thereby resulting in increased temperature inside the
337
junction box; to a level that may be high enough to melt the box and the system, cause a
338
fire, and accelerate degradation of the PV cell.
ACCEPTED MANUSCRIPT 339
(4) Therefore, it is necessary to detect failure of a bypass diode in PV modules, in order to
340
prevent fires. Moreover, it is possible to distinguish shading or bypass diode failure in a
341
PV module by checking the temperature characteristics of the module in operation and
342
non-operation modes. Acknowledgment
344
This work was supported by a grant from the Renewable Energy of Korea Institute of Energy
345
Technology Evaluation and Planning (KETEP) and the Standardization Certification of Korea
346
Energy Agency (KEA) funded by the Korean Ministry of Trade, Industry and Energy.
347
(Project No: 20153010011980, 20143010011820, 71000106)
348
References
349
[1] Silvestre S, Boronat A, Chouder A. Study of bypass diodes configuration on PV modules.
350
Applied Energy 2009;86:1632-1640
351
[2] Alberto Dolara, George Cristian Lazaroiu, Sonia Leva, Giampaolo Manzolini,
352
Experimental investigation of partial shading scenarios on PV(photovoltaic) modules, Energy
353
2013;55:466-475
354
[3] T. H. Jung, J. W. Go, G. H. Kwang and H. K. Ahn, Output Characteristics of PV module
355
Considering Partially Reverse Biased Conditions, Solar Energy 2103; 92:214-220
356
[4] Adria E. Brooks, Daniel Cormode, Alexander D. Cronin, Elsa Kam-Lum, PV System
357
Power Loss and Module Damage due to Partial Shade and Bypass Diode Failure Depend on
358
Cell Behavior in Reverse Bias, Photovoltaic Specialist Conference (PVSC), IEEE 2015; 42
359
[5] Koray Sener Parlak, PV array reconfiguration method under partial shading conditions,
360
Electrical Power and Energy Systems 2014;63:713-721
361
[6]
362
NagamaniChilakapati, Positioning of PV panels for reduction in line losses and mismatch
363
losses in PV array, Renewable Energy 2015;78:264-275
364
[7] Santolo Daliento, Fabio Di Napoli, Pierluigi Guerriero, Vincenzo d’Alessandro, A
365
modified bypass circuit for improved hot spot reliability of solar panels subject to partial
366
shading, Solar Energy 2016;134:211-218
AC C
EP
TE D
M AN U
SC
RI PT
343
Srinivasa
Rao
Potnuru,
Dinesh
Pattabiraman,
Saravana
Ilango
Ganesan,
ACCEPTED MANUSCRIPT [8] Tey Kok Soon, Saad Mekhilef, A Fast-Converging MPPT Technique for Photovoltaic
368
System under Fast-Varying Solar Irradiation and Load Resistance, IEEE 2015; 11:176-186
369
[9] M. Seyedmahmoudian, B. Horan, T. Kok soon, R. Rahmani, A. Muang Than Oo, S.
370
mekhilef, A. Stojcevski, State of the Art Artificial Intelligence-based MPPT Techniques for
371
Mitigating Partial Shading Effects on PV systems – A review, Renewable and Sustainable
372
Energy Reviews, 2016; 64:435-455
373
[10] Mohammadmehdi Seyedmahmoudian, Rasoul Rahmani, Saad Mekhilef, Amanullah
374
Maung Than Oo, Alex Stojcevski, Tey Kok Soon, Alireza Safdari Ghandhari, Simulation and
375
Hardware Implementation of New Maximum Power Point Tracking Technique for Partially
376
Shaded PV System Using Hybrid DEPSO Method, IEEE Transactions on Sustainable Energy,
377
2015;6:850-862
378
[11] Yaw-Juen Wang, Po-Chun Hsu, An investigation on partial shading of PV modules with
379
different connection configuration of PV cells, Energy 2011;35:3069-6078
380
[12] Pieter Bauwens, Jan Doutreloigne, Reducing partial shading power loss with an
381
intergrated Smart Bypass, Solar Energy 2014;103:134-142
382
[13] N.A. Al-Rawi, M.M. Al-Kaisi, D. Asfer, Reliability of photovoltaic modules II.
383
Interconnections and bypass diodes effects. Solar Energy Materials and Solar Cells
384
1994;31:469-480
385
[14] Alonso-Garcia MC, Ruiz JM, Analysis and modelling the reverse characteristic of
386
photovoltaic cells. Solar Energy Materials and Solar Cells 2006;90:1105-1120
387
[15] Daniel Gomez Lorente, Simone Pedrazzi, Gabriele Zini, Alberto Dalla Rosa, Paolo
388
Tartarini, Mismatch losses in PV power plants, Solar Energy 2014;100:42-49
389
[16] Gough Yumu Lui, Richard Corkish, Comparison of Different Bypass Diode
390
Technologies, Solar 2011, the 49th AuSES Annual Conference 2011:1-10
391
[17] Haeberiln. H, Damages at Bypass Diodes by Induced Voltages and Currents in PV
392
modules Caused by Nearby Lightning Currents. 22nd European Photovoltaic Solar Energy
393
Conference, Milano, Italy, 2007
394
[18] Dhere. N, Shiradkar, N. Schneller E. and Gade V., The Reliability of Bypass Diode in
395
PV modules, Proc. SPIE 8825, 2013
AC C
EP
TE D
M AN U
SC
RI PT
367
ACCEPTED MANUSCRIPT [19] K. Kato, “PVRessQ!”: A Research Activity on Reliability of PV System from an user’s
397
viewpoint in Japan, Proc. Optics + Photonics 8112(SPIE, San Diego, California, USA, 2011)
398
[20]Hui Gao, Shunzhong Luo, Huaming Zhang, Heyi Wang, Zhonghua Fu, Demonstration,
399
radiation tolerance and design on a betavoltaic micropower, Energy 2013;51:116-122
400
[21] M.AlRashidi, M. AlHajri, K. El-Naggar, A. AlOthman, A new estimation approach for
401
determining the I-V characteristics of solar cells, Solar Energy 2011;85:1543-1550
402
[22] W. Gong, Z. Cai, Parameter extraction of solar cell models using repaired adaptive
403
differential evolution, Solar Energy 2014;94:209-220
404
[23] T. Ikegami, T. Maezono, F. Nakanishi, Y. Yamagata, K. Ebihara, Estimation of
405
equivalent circuit parameters of PV module and its application to optimal operation of PV
406
system, Solar Energy Materials & Solar Cells 2001;67:389-395
407
[24] K.H. Hussein, I. Muta, T. Hoshino, Osakada, Maximum photovoltaic power tracking: an
408
algorithm for rapidly changing atmospheric conditions, IEE Proceedings on Generation,
409
Transmission, Distribution 1995;142: 59–64
410
[25] M.B. Eteiba, E.T.El Shenawy, J.H. Shazly, A.Z.Hafez, A Photovoltaic(Cell, Module,
411
Array) Simulation and Monitoring Model using MATLAB/GUI Interface, International
412
Journal of Computer Applications 2013;69:14-28
413
[26] Villalva, Marcelo Gradella, Gazoli, Jonas Rafael, Filho andErnesto Ruppert.
414
Comprehensive approach to modeling and simulation of photovoltaic arrays, IEEE Trans.
415
Power Electron 2009; 24:1198–1208
416
[27] Moradi, M.H. andReisi,A.R. A hybrid maximum power point tracking method for
417
photovoltaic systems, Solar Energy 2011; 85: 2965-2976
418
[28] Amit Jain, Avinashi Kapoor, Exact analytical solutions of the parameters of real solar
419
cells using Lambert W-funcion, Solar Energy Materials & Solar Cells 2004; 81:269-277
420
[29] F. Ghani, M. Duke, Numerical determination of parasitic resistances of a solar cell using
421
the Lambert W-function, Solar Energy 2011; 85:2386-2394
422
AC C
EP
TE D
M AN U
SC
RI PT
396
ACCEPTED MANUSCRIPT 423
Table 1. Classification of reference by specific parameters Specific parameter
Purpose
Reconfiguration of PV arrays
Improve the output in shading and
[1], [2], [5], [6], [7]
or cell arrangements
reduce a loss
and [8]
Bypass diode circuit added
Reduce hot spot temperature and the
MOSFET
power loss
PV module model including shaded cell
Evaluate the power loss
[9] and [19]
To predict output of shading module
[10]
Comparison on power loss and impact [11]
MPPT algorithm and
Reduce power loss in irradiance
dc-dc
M AN U
Cell of reverse bias voltage
converter
variation
Damaged bypass diode
Analyze damaged diode
424 425
[12], [13] and [14] [16] and [18]
TE D
Table 2. Temperature and resistance according to current of diode Forward direction for normal diode
Reverse direction for damaged diode
Temperature
Resistance
Current
Temperature
Resistance
(A)
(°C)
(Ω)
(A)
(°C)
(Ω)
2
41.6
0.16
2
35.1
0.11
62.4
0.17
4
56.4
0.22
79.5
0.17
6
91.7
0.34
8
100.6
0.18
8
139.7
0.48
10
120.2
0.18
10
219.2
0.7
AC C
6
EP
Current
4
427
RI PT
PV module
[3] and [4]
SC
Impact of partial shading on
426
Reference
Table 3. Characteristics of the test PV module Maximum Power@ A
Maximum Power Point Voltage@< A Maximum Power Point Current@ A
260 W 30.70 V 8.62 A
ACCEPTED MANUSCRIPT Short Circuit Current@L A
9.16 A
Open Circuit Voltage@< L A
428
38 V
Table 4. I-V characteristics of PV modules under STC, one normal, one with 33% shading
430
and one with a damaged diode Characteristics
Normal module
measured
module
@ A
One damaged diode module
171.4
38
25.6
26
30.5
19.8
20.2
@A
9.18
9.17
9.18
8.49
7.69
8.6
175.99
433 434
AC C
EP
TE D
432
M AN U
L @A
SC
< L @
263.8
< @
431
33 % Shaded
RI PT
429
(a) Measured I-V curve for a normal and a damaged bypass diode
SC
RI PT
ACCEPTED MANUSCRIPT
435
437
(b) Calculated power loss of each bypass diode based on I-V curve
M AN U
436
Fig. 1. Characteristics of a bypass diode (SBX 1240/Diotec Semiconductor AG)
440
AC C
439
EP
TE D
438
Fig. 2. Measured surface temperature of a diode
441 442
Fig. 3. One-diode equivalent-circuit model of a PV-cell
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. Equivalent circuit of a PV module under partial shading
445 446 447
AC C
EP
TE D
444
M AN U
443
Fig. 5. Equivalent circuit of a PV module with failure of one bypass diode
RI PT
ACCEPTED MANUSCRIPT
448
Fig. 6. Schematic diagram of partially shaded PV module and an unshaded model
450
representing a damaged bypass diode
452 453
Fig. 7. Estimated resistance of 0_ using resistance measured in the reverse direction
AC C
451
EP
TE D
M AN U
SC
449
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
454 455
456 457 458 459
AC C
EP
TE D
(a) I-V curve
(b) P-V curve Fig. 8. Results of simulation
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
460 461
Fig. 9. I-V curve characteristics of a shaded PV module and one with a damaged bypass
462
diode
(a)
AC C
465
EP
464
TE D
463
466 467
(b)
468
Fig. 10 Current flow of PV module in each condition: (a) Current flow of PV module under
469
partial shading, (b) Current flow of PV module with a damaged bypass diode
470
RI PT
ACCEPTED MANUSCRIPT
471
(b)
Fig. 11. (a) Schematic diagram and (b) Tests bed configuration of PV modules
474 475
AC C
EP
TE D
M AN U
473
(a)
SC
472
Fig. 12 Thermal image of }|~~ in a PV module
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 13 Surface temperature at each point in the PV module shown in Fig. 12
478 479
AC C
EP
TE D
477
M AN U
476
Fig. 14. Intensity of irradiation during the measurement
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 15 Current in PV module with damaged diode while the PV system operated
482 483
AC C
EP
TE D
481
M AN U
480
Fig. 16 Thermal image of PV modules during inverter operation
484
(b) The irradiance condition
Fig. 17 Temperature in the junction box when the system is not operating
489 490
(a) The temperature in junction -box
EP
488
(b) The irradiance condition
Fig. 18 Temperature in the junction box when the system is operating
AC C
487
TE D
M AN U
486
(a) The temperature in junction box
SC
485
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights •
It proposed the mathematical models of PV modules under shading and with damaged bypass diode. Electric and thermal characteristics of PV modules under shading and with damaged bypass diode were discussed.
It is possible to distinguish shading or bypass diode failure in PV modules using the
EP
TE D
M AN U
SC
thermal characteristics.
AC C
•
RI PT
•