1

Health Monitoring of Aluminum Electrolytic

Capacitor to Minimize Payback Period of Single
Phase Solar PV System
AbstractSingle stage, single phase solar photovoltaic (PV)
inverters are used to directly feed solar power to the grid. DClink voltage of these inverters have oscillations of twice the grid
frequency. Due to high energy density and low cost, Aluminum
Electrolytic Capacitors (AEC) are popularly employed in dc-link
to suppress the second harmonic and switching voltage ripples.
With ageing AECs loses its ripple suppression quality due to
increase in its equivalent series resistance (ESR) and decrease
in its capacitance value. This causes oscillations in PV operating
point. The oscillating PV power around it maximum possible
value, results in reduction of PV power extraction efficiency.
Technique for real time monitoring of PV power extraction
efficiency is proposed in this paper. The technique is based on
average and maximum power calculations. Based on status of
extraction efficiency, replacement of capacitor is suggested at
a suitable time so as to minimize the payback period of PV
system. Maintenance cost and discount factor of PV system are
also included in payback period evaluation. Detailed simulation
studies are carried out using Simulink. A laboratory prototype
of single phase inverter is developed. The proposed technique is
implemented along with maximum power point tracking (MPPT)
controller. Results of the experimental studies are included in this
paper.
Index TermsDiagnostics; online monitoring; payback period;
PV power extraction efficiency; second harmonic ripples; solar
PV inverter

I. I NTRODUCTION

OWER Electronic Converters (PECs) are used to interface

solar photovoltaics (PV) with grid. High efficiency of
PEC is desired due to high cost of solar PV panels and low
panel efficiency [1]. Efficiency of PEC is affected by (i) power
conversion losses and (ii) power extraction losses. Power conversion losses include switching and conduction losses in the
PEC. Literature exist on improving the conversion efficiency of
PECs used for solar PV system [2]. The extraction loss is due
to deviation of solar PV operating point from its maximum
power point (MPP). This deviation is either due to MPP
tracking (MPPT) algorithm or oscillations in solar PV voltage.
Power extraction efficiency of various MPPT techniques typically varies between 90% to 99.5% [3][5]. Further, extraction
efficiency is also affected by second harmonic and switching
harmonics ripple component appearing across PV voltage in
single phase solar PV system. Aluminum electrolytic capacitor
(AEC) is used as an energy storage element to limit second
harmonic voltage ripple to a low value and to maximize the
PV power extraction efficiency (PEE) [6]. Technique to reduce
dc-link capacitance without affecting PV extraction efficiency
is suggested in [7]. However, due to two stage conversion,
efficiency of system is less than that of single stage conversion.
In [8], reliable circuit topology is proposed for photovoltaic

AC modules by avoiding the use of AEC. This increases PEC

complexity and reduces conversion efficiency due to three
stages of power conversion. In [9], a technique is proposed
to eliminate the use of AEC at dc-link of three phase grid
connected PV system. The modified Current Source Inverter
(CSI) is used to feed power to grid. However, the technique
is not viable for single phase PV system, due to large value
of dc-inductor used to suppress low frequency current ripples.
AECs are popularly used in converters due to their low
cost and volume as compared to film type capacitors [10].
However, AEC degrades due to its electrolyte loss with time,
which leads to rise in equivalent series resistence (ESR)
and fall in capacitance value [11][14]. Further, the rate
of loss in electrolyte increases with temperature [15]. This
leads to increase in ESR of capacitors due to high operating
temperature, thereby increasing magnitude of ripples in PV
voltage [16]. This reduces the PEE, which results in less
power availability from solar PV system Therefore, timely
replacement of electrolytic capacitor is necessary for high
power extraction efficiency.
Several approaches for online health monitoring of electrolytic capacitor are suggested in literatures [16][22]. In
[16], online monitoring is suggested based on monitoring of
ripple voltage across capacitor. If the ripple voltage crosses
the set limit, it gives the indication of capacitor failure. The
limitation of this technique is that under transient condition
the ripple voltage may grow leading to error in detection.
Moreover, the technique is not suitable if the capacitor core
temperature does not remain constant. In [17], a predictive
maintenance technique for electrolytic capacitor is proposed
for boost converter. The technique is non-invasive and uses
double estimation of ESR and capacitance for higher accuracy.
The software-added approach is proposed for switch mode
power supply to predict the health of AEC [18]. The voltage
and current ripples across capacitor are sampled to estimate
ESR. The technique also incorporates the temperature and
load variation. A technique is proposed for three phase PWM
converter to monitor the ESR of dc-link AEC [19]. AC current
is injected from three-phase source side due to which ripples
are produced across DC-link capacitor. Recursive least square
method is used for the estimation of ESR. In [20], RMS value
of switching frequency ripple voltage and current through
capacitor are calculated. Ratio of these two is used to obtain
ESR value. In [21], the ratio of power loss in capacitor to
square of RMS ripple current through capacitor is evaluated
to determine ESR. ESR exceeding predefined limit gives
indication of replacement of capacitor. The aforementioned
techniques [16][21], require current sensor of high bandwidth

START

to determine the ESR value, thereby increasing system cost.

In [22], a technique based on PV power extraction efficiency
is proposed. However, experimental results are not included.
Implementation challenges related to experimentation were not
discussed. Further, mathematical analysis of the technique is
not included in [22].
The objective of this paper is to propose accurate method
for timely replacement of capacitor on the basis of PV PEE
to minimize the payback period. For mathematical analysis,
Double Fourier Series Analysis (DFSA) is carried out to
determine the ripples in dc-link current. A scaled down laboratory prototype is developed to validate the proposed scheme.
Results are discussed in-detail in this paper.
The paper is organised as follows. In section II, proposed
technique is presented. Section III covers the mathematical
modeling of the PV system including proposed technique.
Section IV discusses the criterion for replacement of capacitor
on the basis of payback period. Detailed simulation and experimental results are presented in sections IV and V, respectively.
Section VI concludes the paper.

wait for
MPPT to
settle
Delay

S3

S1

iL
S4

S2

MPPT

Fig. 1.

L-filter

S3 S4

PIi

-1

+1
-1

+
Vpv*

S 1 S2

AC

___

vpv

ipv

Lo

+-

vpv

ic

PIv

Sin
o

A grid feeding solar PV system.

is
count = n

Yes

Sin

PLL

PV PEE

Check Payback
Period if
capacitor is
replaced at this
instant

Is
Yes Replace the
Operate with No
Payback period
the same
capacitor
minimum
capacitor
?

Fig. 2.

Flow chart of proposed technique.

and sensor bandwidth. Another challenge to estimate the PEE

is due to MPPT algorithm. Reference value of solar PV voltage
changes periodically to track the maximum power. The actual
PV voltage follows its reference value with some delay. The
samples if taken in this transient period may lead to error in
PV PEE calculation.
Flow chart of the proposed technique is shown in Fig. 2.
To address the above issue, sampling of vpv and ipv is done
after settling of dc-link voltage, as shown in flow chart. In
next step, n no. of samples of vpv and ipv are captured with
sampling rate of 1/Ts . In next step, instantaneous PV power
(ppv ) is calculated and Psum and Psum1 is find out in order
to evaluate PV PEE by the following equations:
pav =

Display

PV PEE
evaluation

A. PV Power Extraction Efficiency (PEE) calculation

No

DC-link idc

Proposed technique has two submodules: (a) PV Power

Extraction Efficiency (PEE) calculation, and (b) Criterion for
replacement of capacitor.

ipv

count = count+1

II. P ROPOSED T ECHNIQUE

Fig. 1 shows a single phase full bridge inverter, interconnecting solar PV to the grid. The inverter is controlled
by the MPPT, which generates the reference PV voltage.
This voltage is realized using voltage and current controllers.
Power supplied by solar PV panel oscillates at twice the grid
frequency. In the proposed technique the ratio of average PV
power to its maximum value is calculated to find the PV
PEE. For this, sampling of instantaneous PV voltage (vpv )
and PV current (ipv ) is required to calculate instantaneous
PV power. Large number of samples per second would offer
higher accuracy in PV PEE. However, the sampling rate is
limited due to the speed of analog to digital converter (ADC)

0 ,count=0,
0 ,n

psum
n

(1)

where, pav is the average power. Due to noise, the direct

extraction of maximum power (pm ) from all the samples of
Ppv may contain an error. To eliminate the effect of noise on

2.8175

pm , the maximum power is calculated indirectly by obtaining

ripple content in power. The RMS power, prms and ripple
power rms, pripp,rms are calculated as given below:
r
prms =

pripp,rms =

psum1
n

(2)

p2rms p2av

(3)

Hc1

Fref

+1

Hc2

sinx

Ts

/2

O
-1
-1

3/2

y sinx, x 0,2
Hc1

S1

The peak value of ripple power, pripp,max is given by:

pripp,max = 2.pripp,rms
(4)

S2

Fref

+
S4
Hc2

S3

The maximum value of PV power is:

pm = pripp,max + pav

(5)

The ratio of average PV power to the maximum PV power

is PV PEE:
pav
(6)
P V P EE =
pm
After evaluation of PV PEE, criterion for replacement of
capacitor based on PV PEE is discussed in subsequent section.
B. Criterion for replacement of capacitor
The degradation of electrolytic capacitor results in larger dclink voltage ripples due to which average PV power decreases.
The reduced PV PEE results in lower revenue generation
which increases the payback period of PV system. Therefore,
timely replacement of capacitor is required for maximum PV
PEE. However, the frequent replacement of capacitor will
add to the system cost. To quantify this, payback period of
the system is determined for different period of capacitor
replacements. The efficiency at which capacitor is replaced
versus the payback period is determined. The minimum value
of this curve, would indicate the PV PEE at which capacitors
should be replaced. The details about the payback period
calculations is given in Appendix A. Detailed calculations of
this criteria for a sample system is explained in section IV.

Fig. 3.

Unipolar PWM switching scheme.

c t

c t
=

c t
Hc2 (c t) =
1

c t
=
1

Fref (o t) = M sin(o t)
Hc1 (c t) =

< c t < 0

(8a)

0 < c t <

(8b)

< c t < 0

(8c)

0 < c t <

(8d)
(8e)

where, c and o are angular frequency of carrier and reference waveforms, respectively. After substituting, x = c t + c
and y = o t + o for o = c = 0, the equations of carriers
and reference waveforms are given by,
x

x
=

x
Hc2 (x) = 1

x
= 1

Fref (y) = M sin(y)

Hc1 (x) =

< x < 0

(9a)

0<x<

(9b)

< x < 0

(9c)

0<x<

(9d)
(9e)

III. M ATHEMATICAL MODELING

Analytical equation for PV PEE is derived for the single
phase grid connected PV system. The full bridge inverter
shown in Fig. 1, is operated by unipolar sinusoidal pulse width
modulation switching scheme, as shown in Fig. 3. DC-link
voltage includes oscillation of twice the grid frequency and
switching frequency. Therefore, double fourier series analysis
is used to determine various harmonic components [23].
DC-link current (idc ) is obtained using switch states and
inductor current (iL ),
idc = iL (S1 .S4 S2 .S3 )

(7)

Equations of the carrier waveforms, Hc1 and Hc2 and

reference waveform, Fref , used for PWM generation are given
below,

TABLE I
C ONDITION FOR DETERMINING DC LINK CURRENT
idc
il
0
il

When < x < 0

M sin(y) > x

x
1 < M sin(y) <

M sin(y) < x
1

When 0 < x <

x
M sin(y) >
x
1 < M sin(y) <

x
M sin(y) <
1

By using (7) and (9), the condition for idc is tabulated

in Table I. Unit cell of x-y plot describing the boundary
conditions for idc is shown in Fig. 4. The various frequency
components in dc current are given by,
Cmn = Q1 Q2 Q3
where,

(10)

2 x
3/2

ii=iL

ii = 0

ii = 0

/2

(M Sin( y) + )

ii= -iL

Parameter
Maximum Power
Voltage at Pmax
Current at Pmax
Short circuit current
Open circuit voltage
Temperature coefficient of Voc
Temperature coefficient of Isc
Grid frequency
Grid Voltage

M Sin ( y ) y = o x o +
c
o

ii= 0

y sinx, x 0,2

sinx

M S in ( y )
2.8175

TABLE II
S PECIFICATION OF SYSTEM

ii= -iL

Fig. 4.

Parameter
Inductor
DC link Capacitor
ESR of input Cap.
Switching frequency

X-Y plot of unit cell.

Q1 =

1
2 2

Q2 =

1
2 2

Q3 =

1
2 2

nR

M sin(y)
R

iL .ej(mx+ny) dxdy

0 M sin(y)
M sin(y)+
R

n R0

iL .ej(mx+ny) dxdy

n R0

iL .ej(mx+ny) dxdy

(M sin(y)+)

and, iL = IL sin(y )
where, IL is the peak inductor current and is the phase
difference between fundamental component of inverter output
voltage and inductor current. Ripples in idc is absorbed by
electrolytic capacitor. By using (7), (9) and (10), PV voltage
is given by,
vpv = Vpv +

X
X

Value
2 kW
365 V
5.48 A
6.3 A
390 V
.03%
-.33%
50 Hz
230 V

TABLE III
C ONVERTER COMPONENT

O
-1

( M Sin( y ) + )

Variable
PM P P
VM P P
IM P P
ISC
VOC

f0
vg

Vmn sin(mc t + no t)

(12)

2
1
(mc no )C

(13)

Variable
L
C
ESR
fs

Value
3 mH
4700 F
0.02
10kHz

platform. Specifications of system and converter components

are given in Tables II and III, respectively.
Reference MPP voltage and actual DC link voltage are
shown in Fig 5. Simulated result shows that the dc link voltage
is following the reference voltage with some delay. As already
discussed in section II, this delay depends on the plant and
controller parameters. To apply the proposed technique, this
transient period is excluded for PV PEE calculations.

m=0 n=

where,
s
Vmn = Cmn

ESR2 +

PV current (ipv ) for ideal solar PV array is given by,

vpv

ipv = I + Io (1 e N )

(14)

where,
I = photo-generated current (A)
Io = diode reverse saturation current (A)
= .026 at 25o C
N = number of solar cells in series
The instantaneous PV power is given as,
p(t) = vpv ipv

Fig. 5.

(15)

The ratio of average and maximum of p(t) over a 2nd

harmonic time period are solved using MATLAB. This ratio
is PV PEE. Details are provided in the next section.
IV. S IMULATION R ESULTS AND D ISCUSSION
A single phase grid feeding solar PV inverter along with
controller, shown in Fig. 1 is simulated in Matlab/Simulink

Actual and reference PV voltage during transient.

The inverter output voltage, grid voltage and current are

shown in Fig. 6. With ageing ESR and C changes, depending
on the temperature. Detailed relation of ESR and C with
temperature is provided in [24], [25]. Due to reduction in
C and increase in ESR with time, variation of PV PEE
is observed. PV PEE is determined by the following two
methods, 1) by implementing proposed technique in Simulink
model, and 2) by implementing mathematical model derived
in section III on MATLAB. The results are tabulated in Table
IV. Following are the key observations:
1) As the capacitor degrades, PV PEE detoriates. Difference
in values of extraction efficiency estimated by simulink model
(column 4 of Table IV) and mathematical model (column
5) increases. This is because mathematical model dont in-

10000
Inverter output voltage (V)
Grid voltage (V)

0
-200

0.2

0.22

0.24

0.26

0.28

0.3

0.32

0.34

0.36

0.38

0.4

9000
Payback Period (days)

200

inductor current (A)

20

8000
7000
6000
5000
4000
3000
0.5

0.6
0.7
0.8
Power Extraction Efficiency

0.9

-20

-40
0.2

0.22

0.24

0.26

0.28

0.3
Time

0.32

0.34

0.36

0.38

0.4

Fig. 6. Simulink waveform of inverter output voltage, grid voltage and

inductor current.
TABLE IV
S IMULATION R ESULTS @ 50o C O PERATING T EMPERATURE
Time
(Hours)

ESR
()

Capacitance
(F)

0
10,000
20,000
31,120
38,900
40,308
41,057
41,217
41,260
41,296
41,330
41,364
41,375
41,385
41,400
41,420

.02
.026
.04
.08
.32
.7
1.9
3
3.55
4.2
5.067
6.4
7
7.63
8.85
11.248

4700
4247
3760
3290
2937
2874
2840
2832
2831
2829
2827
2826
2826
2825
2824
2824

%PV PEE( by
(Simulink Model of
Proposed technique)
99.94
99.93
99.90
99.87
99.87
99.69
96.38
90.38
86.86
82.64
77.3
70.2
67.45
64.85
60.52
54.08

%PV PEE( by
(Mathematical
model)
99.99
99.99
99.99
99.99
99.97
99.88
99.09
97.60
96.51
94.85
91.85
84.96
81.14
76.38
67.89
59.43

corporates with non idealities of the system, such as switch

resistance, diode resistance, inductor resistance etc.
2) Observation of Table IV reveals that for a given criteria
of capacitor life from datasheet ( i.e. ESR becoming twice of
initial value), the PEE is still 99.90%. So, it is uneconomical to
remove the capacitor based on the life provided by the capacitor manufacturers. So as discussed in section II criterion for
replacement of capacitor based on payback period calculation
is suggested. Payback period calculation is alraedy discussed
in [22].
Capacitor must be replaced if PV PEE drops to a set value.
Payback versus PV PEE at which capacitor must be replaced
is shown in Fig. 7. The payback period of PV system is high if
the capacitor is replaced at higher PV PEE. Though, extraction
efficiency increases, this leads to marginal increase in revenue.
However, frequent replacement of capacitor would result in
larger GSC. Therefore, payback period increases. In case the
capacitor is replaced at very lower value of PV PEE, GSC
would reduce but revenue collection per unit time reduces
significantly. This results in higher PV system payback period.
The replacement of capacitor at certain PV PEE results in

Fig. 7.
PEE.

Change in payback period on capacitor replacement at different PV

optimum value of both GSC and revenue collection per unit

time. At this instant the payback period of the PV system is
minimized.
V. E XPERIMENTAL V ERIFICATION
A scaled down laboratory prototype of single phase grid
connected PV system, shown in Fig. 1, is developed. The built
prototype of single phase inverter feeds 120W of power at
unity power factor. Specifications of PV module and inverter
components are given in Tables V and VI, respectively. The
experimentations are performed with three different values of
dc-link capacitors as tabulated in Table VI.
FSBB20CH60C (IGBT module) is used as inverter, which
includes integrated short circuit protection and inbuilt gate
drivers. Agilent Technologies made E4360A is used to emulate
solar PV source. DSP TMS320F2808 is used for the closed
loop control and to implement the propsed technique. PWM
signals for IGBTs are generated using interrupt-based timer
modules of the controller. OPAMP-based differential amplifer
circuit is used to attenuate feedback signals in order to make
it compatible with controllers ADC. Inductor and PV currents
are measured using LEM-LA-25P current transducers. For the
operation of OP-AMPS, transducers, IGBT driver circuit and
TABLE V
S PECIFICATION OF PV MODULE
Parameter
PV Voltage at Pmax
PV Current at Pmax
Short circuit current
Open circuit voltage

Value
80 V
1.7 A
2.2 A
100 V

TABLE VI
I NVERTER PARAMETERS
Parameter
DC link Cpacitance(AEC)
Inductor(CRGO core)
Grid frequency
Grid Voltage
Switching Frequency

Value
Capacitor C1 : C=1860 F, ESR=61 m
Capacitor C2 : C=820 F, ESR=62 m
Capacitor C3 : C=410 F, ESR=134 m
L=3.1 mH and R=0.2 at 100Hz
50 Hz
45 V
10 kHz

PV
voltage
Inverter output
voltage

PV
current

Grid voltage

Inductor
current

Fig. 8. Waveform of inverter output voltage (50 V/div), grid voltage (50
V/div) and inductor current (5 A/div), Time: 5 ms/div

controller, a regulated 15V supply is generated from a linear

power supply circuit.
The inverter output voltage, inductor current and grid voltage are shown in Fig. 8. Inductor current is of the same
phase as that of grid voltage, and power factor is found to
be 0.95. Total harmonic distortion (THD) of the grid voltage
and inductor current are in the range of 1.7%- 1.8% and 2.5%2.8% for three different dc capacitors, respectively.
The perturb and observe MPP algorithm is applied to
extract maximum PV power. The waveform of reference PV
voltage and actual PV voltage is shown in Fig. 9. Since the
power fluctuate at twice the grid frequency (50Hz), the dclink voltage has the dominant ripple of twice the fundamental
frequency (100Hz). Furthermore, the dc-link ripple magnitude
depends on the value of dc-link capacitor.
For Capacitor C1 and Capacitor C2 , dc-link voltage ripple
and PV current ripple are shown in Figs. 10 and 11 ,
respectively. From these waveforms it is clear that as dclink capacitance decreases the magnitude of ripple voltage

Actual PV voltage

Reference PV
voltage

Fig. 10. Waveform of dc-link voltage ripple (500 mV/div) and current ripple
(50mA/div) for Capacitor C1 : C=1860 F, ESR=61 m

PV
voltage
PV
current

Fig. 11. Waveform of dc-link voltage ripple (1 V/div) and current ripple
(100 mA/div) for Capacitor C2 : C=820 F, ESR=62 m

increases. For capacitor C1 and C2, dc link voltage ripple is

2.94 V and 3.36 V, respectively.
The technique of online monitoring of PV PEE is validated experimentally on the prototype. For three values of
capacitance average and maximun solar power is evaluated
as discussed in Section II. PV PEE is obtained and tabulated
in Table VII along with simulation results for the same value
of capaciances and system specifications used in prototype.
The maximum power remains same in the three cases, as
expected. Further, with reduction in capacitance value and increase in ESR value, average power decreases. Experimentally
the extraction efficiency deteriorates from 98.3% to 93.126%.
TABLE VII
E XPERIMENTAL AND SIMULATION R ESULTS

Fig. 9. Reference (10 V/div) and actual voltage (10 V/div) during maximum
power point tracking.

Pavg
Pmax
PV PEE

C=1860 F
ESR=61 m
Exp.
Sim.
133.2W 133.7W
135.5W 135.9W
98.3%
98.38%

C=820 F
ESR=62 m
Exp.
Sim.
129.9W
130.8W
135.97W
135.9W
95.53%
96.24%

C=410 F
ESR=134 m
Exp.
Sim.
126W
127.9W
135.3W
135.9W
93.126%
94.11%

VI. CONCLUSION
Degradation of electrolytic capacitor leads to rise in dclink voltage ripples. Larger magnitude of dc-link voltage
ripples reduces the average PV power, thereby reducing PV
power extraction efficiency (PEE). A technique is suggested
for online monitoring of PV PEE. It is concluded from the
experimental results that at twice the value of initial ESR, PV
PEE is still greater than 93%. It is uneconomical to replace the
capacitor at that instant. Replacement of capacitor is suggested
on the basis of PV PEE for minimum system payback period.
Payback period calculation of system including maintenance
cost for replacement of capacitor at different PV PEE is
included. The merits of the proposed technique are, i) Evaluation of PV PEE is unaffected from change in PV reference
voltage due to MPPT, ii) no additional circuits or sensors are
required to estimate the life of AEC. The proposed technique
is validated by experimentation. PV PEE is found to decrease
with reduction in C and increase in ESR values.
A PPENDIX A
PAYBACK P ERIOD A NALYSIS
Payback period is the recovery time period of initial investment of system [26]. Calculation of payback period depends
on whether the cash flow per period from the project is even
or uneven. For even cash flow, payback period is the ratio of
initial investment to the total cash inflow per period.
For a solar PV system, non uniformity in solar radiations
and degradation of capacitor leads to uneven cash flow per
period. Total cash inflow (CF) till the last negative cumulative
cash flow is given by,
CF =

n
X

Ei Ci (1 + r) 365

(16)

i=1

where,
Ei = the output energy per period (KWh)
Ci = Feed in Tariff (FIT) ($/kWh)
r = rate of rise in FIT (per unit)
i = period in days.
Initial PV system cost (PC) includes the cost of solar panels,
inverter, MPPT tracker, cables, system controllers etc. Solar
panels cost around 50% of the initial PV system cost [27].
Gross system cost (GSC) is defined as initial PV system cost
plus the cost of replaced electrolytic capacitors till the total
system cost is redeemed. Discount rate of PV system is also
included in GSC calculation.
n

GSC = P C(1 + d) 365 +

m
X

CC[(1 + d)

niT0
365

(17)

i=1

where,
P C = initial cost of PV system ($)
CC = cost of replaced electrolytic capacitor ($)
d = discount rate of solar PV system (per unit)
m = number of replaced electrolytic capacitor
T0 = time period of electrolytic capacitor replacement
Payback period (N) of solar PV system can be evaluated
by saving to investment ratio (SIR), which is defined as,

CF
(18)
GSC
Following conclusions can be made from (18)
1) For n < N : SIR < 1 investment cost > revenue
generated from system.
2) For n = N : SIR = 1 investment cost = revenue
generated from system. At this instant, the value of n will
give the payback period of PV system.
3) For n > N : SIR > 1 investment cost < revenue
generated from system. Total revenue generated minus total
investment cost of PV system will give the net profit generated
from the PV system.
SIR =

R EFERENCES
[1] M. A. Green, K. Emery, Y. Hishikawa, W. Warta and E. D. Dunlop,
Solar cell efficiency tables (version39), in Progress in Photovoltaics:
Research and Applicat., vol. 20, no. 1, John wiley & sons, ltd., Jan.
2012, pp. 12-20.
[2] Sukesh, N., Pahlevaninezhad, M., Jain, P.K., Analysis and Implementation of a Single-Stage Flyback PV Microinverter With Soft Switching,
IEEE Trans. Ind. Electron., vol.61, no.4, pp.1819, 1833, April 2014.
[3] P. Manganiello, M. Ricco and G. Petrone, Optimization of Perturbative
PV MPPT Methods Through Online System Identification, IEEE Trans.
Ind. Electron., vol. 61, no. 12, pp. 6812-6821, 2014.
[4] Raj, J.S.C.M., Jeyakumar, A.E., A Novel Maximum Power Point
Tracking Technique for Photovoltaic Module Based on Power Plane
Analysis of I-V Characteristics, IEEE Trans. Ind. Electron., vol.61,
no.9, pp.4734, 4745, Sept. 2014.
[5] Paz, F.; Ordonez, M., Zero Oscillation and Irradiance Slope Tracking
for Photovoltaic MPPT, IEEE Trans. Ind. Electron. , vol.61, no.11,
pp.6138, 6147, Nov. 2014.
[6] S. Zengin, F. Deveci, and M. Boztepe, Decoupling capacitor selection
in DCM flyback PV microinverters considering harmonic distortion,
IEEE Trans. on Power Electron., vol. 28, no. 2, pp. 816-825, Feb. 2013.
[7] S. K. Yarlagadda, W. Shireen, A maximum power point tracking
technique for single-phase PV systems with reduced dc-link capacitor,
in Proc. of IEEE Appl. Power Electron. Conf. and Exposition, 16-20
March 2014, Fort Worth, TX, pp. 2950-2957.
[8] C. Rodriguez and G. A. J. Amaratunga, Long-Lifetime Power Inverter
for Photovoltaic AC Modules, IEEE Trans. Ind. Electron., vol. 55, no.
7, pp. 25932601, June. 2008.
[9] Anand, S., Gundlapalli, S.K., Fernandes, B.G., Transformer-Less Grid
Feeding Current Source Inverter for Solar Photovoltaic System, IEEE
Trans. Ind. Electron., vol.61, no.10, pp.5334,5344, Oct. 2014.
[10] H. Ma and L. Wang,Fault diagnosis and failure prediction of aluminium
electrolytic capacitors in power electronic converters, in Proc. of IEEE
31st Annu. Conf. of Ind. Electron. Soc. (IECON), 6 Nov. 2005, Raleigh,
NC.
[11] K. Harada, A. Katsuki, and M. Fujiwara, Use of ESR for deterioration
diagnosis of electrolytic capacitor, IEEE Trans. Power Electron., vol.
8, no. 4, pp. 355361, Oct. 1993.
[12] S. G. Parker, Jr., Improved spice models of aluminum electrolytic
capacitors for inverter applications, IEEE Trans. Ind. Appl., vol. 39,
no. 4, pp. 929935, July/Aug. 2003.
[13] A. M. R. Amaral and A. J. M. Cardoso, A Simple Offline Technique for
Evaluating the Condition of Aluminum Electrolytic Capacitors, IEEE
Trans. Ind. Electron., vol. 56, no. 8, pp. 32303237, July. 2009.
[14] Y. Zhou, X. Ye and G. Zhai, Degradation Model and Maintenance
Strategy of the Electrolytic Capacitors for Electronics Applications,
in Proc. of Prognostics & Syst. Health Manage. Conf. (PHM), 2011,
Shenzhen.
[15] M. L. Gasperi, Life prediction model for aluminum electrolytic capacitors, in Proc. of IEEE Ind. Applicat. Conf., 6-10 October 1996, San
Diego, CA, pp. 1347-1351.
[16] K. Harada and A. Katsuki, Life detector for smoothing capacitor,
Japanese Patent Appl., JP63081277, 1988.
[17] G. M. Buiatti, J. A. M. Ramos, C. H. R. Garca,A. M. R. Amaral
and A. J. M. Cardoso, An Online and Noninvasive Technique for the
Condition Monitoring of Capacitors in Boost Converters, IEEE Trans.
on Instrumentation and Measurement, vol. 59, no. 8, pp. 2134-2143,
July. 2010.

[18] A. Lahyani, P. Venet, G. Grellet, and P. J. Viverge, Failure prediction of

electrolytic capacitors during operation of a switchmode power supply,
IEEE Trans. Power Electron., vol. 13, no. 6, pp. 1197-1207, Nov. 1998.
[19] X. S. Pu, T. H. Nguyen, D. C. Lee, K. B. Lee and J. M. Kim,
Fault Diagnosis of DC-Link Capacitors in Three-Phase AC/DC PWM
Converters by Online Estimation of Equivalent Series Resistance, IEEE
Trans. Ind. Electron., vol. 60, no. 9, pp. 41184127, Sept. 2013.
[20] K. P. Venet, F. Perisse, M. H. El-Husseini and G. Rojat, Reliazation of
a smart electrolytic capacitor circuit, IEEE Ind. Appl. Mag., vol. 8, no.
1, pp. 16-20, 2002.
[21] M. A. Vogelsberger, T. Wiesinger and H. Ertl, Life cycle monitoring
and voltage managing unit for dc-link electrolytic capacitors in PWM
converters, IEEE Trans. on Power Electron., vol. 26, no. 2, pp. 493503, Feb. 2011.
[22] A. Arya, W. Ahmad and S. Anand, Online Monitoring of Power Extraction Efficiency for Minimizing Payback Period of Solar PV System,
IEEE International Conference on Industrial Technology, ICIT2015,
Seville, Spain, March 2015.
[23] K.S. Gopalakrishnan and G. Narayanan,Harmonic Analysis of DC-Link
Capacitor Current in Sinusoidally Modulated Neutral-Point-Clamped
Inverter, National Power Electronics Conference,2013, Dec. 2013,
Kanpur.
[24] K. Abdennadher, P. Venet and G. Rojat, A real time predictive maintenance system of aluminium electrolytic capacitors used in uninterrupted
power supplies, IEEE Trans. on Ind. Applicat., vol. 46, no. 4, pp.16441652, August 2010.
[25] C. Kulkarni, G. Biswas, X. Koutsoukos, J. Celaya and K. Goebel,
Experimental studies of ageing in electrolytic capacitors, in Proc. of
Ann. Conf. of the prognostics and health manage.Soc., 2010.
[26] J. R. Williams, S. F. Haka, M. S. Bettner and J. V. Carcello, Financial
& managerial accounting, Ed., 16th ed. New York: McGraw-Hill, 2012,
p. 1117.
[27] A. M. Iqbal, R. A. Abd-Alhameed, M. A. Elgadal, H. M. Gabasa and M.
M. Ehbal, Investigation of using photovoltaic system at local residential
community level, in Proc.of IEEE first Int. conf. on Green Energy, Mar.
2014, Sfax, pp. 1-7.