The explicit model of the energy yield with respect to irradiance and cell temperature of a photovoltaic (PV) system can be apprehended using pvsyst software. Building on this data, this paper addresses performance challenges for JA Solar, JAP6 (DG) 60-235 solar PV module driving a load of Enphase, IQ6-60-x-240 grid inverter. The data modeling reflects correlation of 62% between panel temperature and output efficiency. Researchers in the past have claimed that extreme temperature exposure as one of the main impediment in decline of solar panel's life span and figured 25 °C as the ideal temperature for optimum yield. This research proposes the Internet of things (IoT)-based smart solar energy system (SES) for smart cities that automatically tune the low-powered cooling unit to lower panel's temperature to outmatch energy yield and augment solar panels life. The analog design of the cooling mechanism is set up for temperature range from −10 °C to 85 °C using hybrid op-amp proportional–integral–derivative (PID) controller and heat sink/fan with surface mount temperature sensor to maintain module temperature. The experiment analysis showed improvement of 1.7% to 2.99% in output efficiency after considering 1.8 W total power intake of the cooling circuit relative to the pvsyst v6.74 results. To access temperature data of solar panel and output current along with in-built system's current consumption, IoT accreditation is done using node MCU and Wi-Fi module.

Introduction

The exigency in energy demands over 500 million terajoules/year [1] and scarcity in conventional resources have paved the way for the development of renewable energy resources especially solar energy. This copious energy resource with the annual potential of 1575–49,837 exajoules [1,2] has proved to be the best resource as an environment friendly, safe, clean, and green energy alternative [26]. To harness this everlasting energy, solar photovoltaic cell (PVC) is widely used. The power conversion efficiency (PCE) of the solar cell accounts for the intrinsic parameter to quantify the solar cell performance [7]. The standard test conditions (STCs) to govern the PCE of the solar module impels (i) ideal temperature requirement of 25 °C (or 300 K for simulations), (ii) perpendicular direction of incident light, (iii) solar irradiation of 1000 W m–2, and (iv) standard global solar spectrum AM1.5 g.

However, various practical issues [8] on field account for significant deviation from STCs and low PCE of module resulting in degradation of overall solar energy system (SES) performance. With other restraints, viz., different solar irradiance, the tilt angle of module affecting the direction of the incident light, and the temperature rise of solar PVC becomes inevitable owing to practical conditions [912]. In the past, much work has been done with the objective of removing the heat factor of the solar panel to upgrade efficiency. Most of the researchers have suggested employing a thermoelectric effect for bringing down the temperature of the solar panel [1317]. He et al. [18] have reported thermoelectric cooling and heating system driven by a heat pipe PV/thermal module. The output efficiency of the panel reported is 16.7%, with 23.5% thermal efficiency. Marandi et al. [19] worked on an economical solution to cool the solar panel. In their research, a solar cavity is designed and packed with hybrid PV-thermoelectric generator modules. Soltani et al. [20] reported a new nanofluid-based cooling method for a hybrid PV/thermoelectric system.

The aforementioned publications have undoubtedly enhanced the output efficiency by cooling the solar panel, but the deployment of thermoelectric devices to cool the solar panel affects the power intake and impact system cost too [2123]. The accurate implementation of any cooling technique is, however, a function of the temperature of the solar module. To ascertain this value, in most cases, the trial-and-error method is employed. Evidently, temperature maintenance on the solar PV module needs a more robust approach. Accounting for daily hour's variation in solar irradiation intensity, ambient temperature, the direction of sunlight, load condition, and impact of these uncertain parameters on output efficiency, the contemplation of smart SES becomes a requisite. This research work proposes a smart SES that incorporates:

  • pre-data analysis, so that correlation in terms of various parameters and optimum yield of complete year is built up to minimize uncertainties in terms of solar irradiance and PV module properties;

  • design accession considering impacting parameter control (temperature in our case) and overall enhancement of output power conversion efficiency;

  • optimized controller design in assessing added active components to SES to control temperature;

  • comprehension of accurate and fast automatic temperature control of the energy system using the Internet of things (IoT);

If the dependence model correlating maximum power output Pmax to input irradiance and temperature is available, then system to enhance efficiency at low output power due to high temperature can be worked out. This research work focuses on the data modeling of JA Solar, JAP6 (DG) 60–235 PV module driving a load of Enphase, IQ6-60-x-240 grid inverter using the commercial software program pvsyst v6.74 [24].1 This software has marked its stamp worldwide in the prediction and analysis of solar PV parameters. The present work is an IoT-based experimental attempt, which aims the use of cooling mechanism in the case where the output efficiency as per pvsyst v6.74 data analysis results is predicted low due to the nominal high temperature. The proposed work focuses the enhancement of the PCE in months where the temperature of the panel can be considerably brought down. The month where solar panel temperature is at extremes is out of the scope of this research work. The smart SES setup gives optimum results with an exemption of the months (May, June, and July in this work), where panel temperature crosses extreme limits. The usage of IoT-based cooling setup with surface mount resistance temperature detectors (RTDs) is further proposed to be used in other months. The cooling mechanism developed combines temperature sensor with an analog hybrid model of proportional–integral–derivative (PID) and heat sink/fan on the solar PV. The proposal aims in lowering the solar module temperature through IoT smart automation setup for smart cities.

This paper is structured as follows: in Sec. 2, the temperature dependence of solar parameters is briefed. Section 3 shows the data modeling on pvsyst v6.74 simulation results, followed by the design-integrated circuits that define the cooling mechanism and its simulation on proteus software with the experimental setup. IoT enabling of the complete design is then examined. Finally, Sec. 5 presents results and comparison analysis.

Theory: Temperature Dependence of Solar Photovoltaic Cell

The increase in temperature (T) of the solar cell is due to the absorption of solar irradiation as heat, which further impacts material parameters and solar cell performance metrics. Faiman [11] accessed the outdoor operating temperature of photovoltaic modules and reported 50 °C module temperature with ambient temperatures around 25 °C for irradiation intensity of 1000 Wm–2. Ross and Smokler [12] reported 30 °C and 60 °C deviation in the module temperature (with 800 Wm–2 irradiance, wind velocity 1 ms–1 and air temperature 20 °C) with respect to the ambient temperature. The variation in the bandgap energy of the photoactive material paramounts for the temperature dependence of silicon solar cells. Using Varshni relation [25], the temperature dependence of the bandgap in semiconductors can be described as Eq. (1), which shows the linear temperature dependence at high T and an asymptotic behavior with quadratic temperature dependence. 
Eg(T)=Eg(0)αT2T+β
(1)
where Eg(T) is the bandgap of the semiconductor at some temperature T, which may be direct or indirect, Eg(0) denotes its value at T ∼ 0 K, and α and β are constants. The temperature dependence of the bandgap energy affects the output performance efficiency (ɳ) of the solar cell given by Eq. (2). 
η=VocIscFFPin
(2)
where Voc is the open circuit voltage, Isc is the short circuit current, fill factor (FF), and Pin is the input radiated power. The temperature coefficient of efficiency (βη) can be calculated as in Eqs. (3) and (4) [7,26]: 
βη=βVOC+βISC+βFF
(3)
 
1ηdηdT=1VOCdVOCdT+1ISCdJSCdT+1FFdFFdT
(4)
The temperature variation of the open circuit voltage (βVOC) is important to consider as this parameter is most affected by an increase in temperature of the solar cell. The 80–90% dependency of solar cell on temperature is due to high impact on Voc in cases where the PV cell is not unduly restricted figuring FF or resistance losses. From Eq. (5) with theoretical considerations, it can be deduced that βVOC is linearly dependent on temperature (T) [7] 
βVoc=dVocdT=VocEg/qT+kTq[1nednedT+1nehdnedT]
(5)
The fill factor relates Pmax that can be extracted from a cell to the product of its Voc and Isc. Neglecting n, RS, and RSh dependence on temperature, the temperature coefficient of the fill factor (βFF0) can be approximated as Eq. (6) [7]: 
βFF0=1FF0dFF0dT(11.02FF0)(1VOCdVOCdT1TC)
(6)
Now, βFF0 considering the series resistance can be rewritten as Eq. (7) 
βFF0=1FFdFFdTβFF0RSVOC/ISCRS(1RSdRSdT)
(7)
In certain amorphous silicon and nanocrystalline dye cells, the FF increases with temperature (which is the opposite of the usual trend) due to decreasing resistance effects or increasing “mobility-lifetime” products [27]. The life span of the electronic device is also directly impacted by temperature as given by an Arrhenius equation [28] (Eq. (8)) 
R=damagetime=Aexp(EAkT)
(8)
where A is a constant related to reaction, EA is the activation energy associated with the reaction, k is the Boltzmann constant (8.617 × 10−5 eV/K), and T is the absolute temperature. Equation (8) can be rearranged to develop an acceleration factor that relates the life of a component when it is operated at its use temperature Tuse to a test time at temperature Ttest as in Eq. (9) 
AF=timeusetimetest=exp(EAk(1Tuse1Ttest))
(9)

Proposed Smart Solar Energy System

The proposed architecture of smart SES is elucidated in Fig. 1; basically, the whole process is divided into three units: (i) data monitoring unit: to specify the correlation metrics with the specification when the cooling mechanism is needed, (ii) cooling unit: with myriad cooling mechanism proposed in the past, the research work suggests simple, cost-effective, and low-powered cooling unit, (iii) IoT setup: for system monitoring: the operation of cooling can be completely automated counting on RTD sensor mounted on the solar PV module. The complete setup can be further applied on the complete solar plant as the ambient temperature remains the same.

Here, data analysis of the previous data is worked on using pvsyst software. The analysis sets up correlation of output efficiency with various dependent parameters, viz., global horizontal input irradiance (GHI), temperature (T), energy into the grid (Energy), array current (Current), and array voltage (Voltage). The analysis based on the last 1 year data figured temperature as one of the main impediments. Uncertainty in irradiance measurements in the past research was related to the measuring instruments, but this research work focuses on pvsyst software-based data. While existing research work reduces operational issues, the uncertainties related to the instruments are reduced. The second module focuses on controlling this parameter and suggests the cooling mechanism of the solar panel. The experimental setup consists of an aluminum sheet attached at the back side of the solar panel. The panel transfers heat to the sheet. The sheet is connected to the heat sink where it channels out. For sensing T, a wide range of instruments were characterized, and many systematic errors were identified and quantified. The third module works on the automation of this cooling setup using IoT. The heat sink module only starts when the temperature of the PV module goes above 25 °C. The following sections are an elaborative discussion of these modules.

Data Monitoring.

The pvsystv6.74 simulation software is used to estimate the performance of the solar module placed at the J.C Bose University of Science and Technology—location: Faridabad, Haryana, latitude: 28.45 and longitude: 77.35. Along with this software, the metrological data as well as the load data of the complete year 2017 was observed. The evaluation of the performance of the grid inverter Enphase, IQ6-60-x-240 connected with module JA Solar, JAP6 (DG) 60-235 was done. The data acquired from the software is an accurate prediction of the output system yield on hourly simulation of each day. The prediction model of the annual energy output of the solar module is used to figure out the duration with low output efficiency by building a correlation matrix between various parameters as shown in Sec 3.1.1.

Solar Irradiance and Energy Injected Into the Grid.

Solar PV module absorbs solar irradiance as input and converts it into power. The PCE variation is directly proportional to solar insolation and PV temperature. Figure 2(a) illustrates data to understand the impact of solar irradiance and temperature on output. With the increase in temperature, the power generation decreases slightly even when there is a constant solar irradiance.

As the temperature increases, a clear decline in efficiency stats is observed in Fig. 2(b). The temperature sets with Δ of 7 °C, viz., (0, 7), (7, 14), and so on, are set up of easy computation.

Variations in Output Efficiency Versus Consumption Every Month.

The energy into the grid may vary depending upon the irradiance and myriad other parameters. However, the consumption load for either day or night remains constant. The correlation between output efficiency and temperature will help in remarking the dependency of the output energy on T. The correlation coefficient R2 is found to be 84.79% between T and energy into the grid (Energy) and is illustrated in Fig. 3(a). This depicts a high impact of T on output Energy. Figure 3(b) shows the plot of the variation of output efficiency in every month. This data plot in terms of month-wise efficiency gives a deeper insight in the aforementioned correlation analysis. The maximum conversion efficiency is generated in the month of December, i.e., 18% and minimum energy is in the month of June, i.e., 10.175% as shown in Fig. 3(b). The total energy injected to the grid is 375.509 kW h for a complete year.

The cumulative analysis of every month with an average of performance metrics is shown in Table 1. The exposition on table data for the month of May shows the lowest output ɳ, i.e., 5.904% with the highest solar irradiance of 259.41 W/m2 received owing to the high solar module temperature. In the month of May, June, and July, undoubtedly solar panel at location Faridabad, Haryana, is exposed to high irradiation intensity and high atmospheric heat. This results in an increase of T in a solar panel resulting in low output ɳ. Further, the data analysis exhibited January and December month with the lowest irradiance and better power conversion.

The same trend can be clearly observed on the hourly/day data; a false color plot of change in efficiency Δɳ(T) with temperature as reference conditions can be visualized from Fig. 4. The plot is categorized in two zones: (i) wherein GHI is high and T reaches its peak with low output ɳ and (ii) maximum ɳ zone of the day is highlighted, although GHI is low but downswing in T is observed resulting in better output ɳ.

Correlation of Performance Parameters With Output Efficiency (ɳ).

Correlation test between various performance parameters can be performed with the cor.test function in the native stats package. This research uses linear regression with the lm function in the native stats package. The pie-chart correlation matrix and tabular representation of various parameters with the false color chart are illustrated in Fig. 5. The Performance Analytics on temperature versus efficiency plot shows multiple R2 values = 0.6282 signifying 62% correlation coefficient.

The performance analysis contents the strong proof of high temperature, as key stumbling block in effective solar irradiance harvesting for maximum output ɳ. This accounts for the robust cooling mechanism to lower the solar module temperature. Section 3.2 explicates a low powered and cost-effective design of the cooling unit.

Cooling Mechanism Proposed.

Considering the need of low power consumption, circuit accuracy, and practical feasibility, the cooling mechanism constitutes an analog hybrid model of PID, heat sink, and fan controlled by RTD surface mount temperature sensor mounted on the solar PV. This setup is tested for January month considering the results of data analysis, which showed it to be the best month for optimum production. The approximate drop by 2 °C in T is observed in case of T > 25 °C. The increase in output ɳ of solar PV with respect to pvsyst prediction model after reducing power consumption is reported to be 1.7–2.99%. This increase in output η‌ of low powered proposed set up accounts for optimal cooling solution of solar panel, in terms of power consumption. Figure 6 illustrates the working mechanism of the proposal.

To sense temperature on the solar PV module, RTD surface mount sensor with 100 Ω DIN Class A (±0.06 Ω or ±0.15 °C at 0 °C) Accuracy Standard and sensing temperature range −73 to 260 °C continuous, 290 °C (554 °F) is employed. The op-amp PID controller to switch on the fan above the particular reference temperature and switch it off below that is exercised. The proposed work was simulated for testing and verification on proteus 8.3 and shown in Figs. 7(a) and 7(b). The simulation setup was tested for the temperature at −10 and at 85 °C. The equation relating the output and input is given as Eq. (10) 
Vout=200*(V2V1)
(10)
where V2 is the voltage across RT1 and V2 is the voltage across R4, the differential op-amp gain Av = 200. Based on the equation, the output voltage related to the resistance as well as temperature is given in Table 2.

The complete circuit simulation combining temperature sensing and generation circuit with the PID controller and the heat sink/fan cooling circuit is shown in Fig. 8 with the experimental setup shown in Fig. 9, and the power consumption of the cooling unit in Table 3. The total power intake of a cooling unit calculated is 1.996 mW, which is further implied in total power conversion efficiency calculation of the module.

Internet-of-Things-Based Platform for System Monitoring.

IoT streams sensing devices to analyze and share data using a suitable pre-programmed environment. Using this intelligent automation, not only cost parameter is worked upon but the overall efficiency of the device is also improved. To boost up energy yield of the solar PVC by efficient photonic energy harvesting, the use of IoT has proven useful [29,30]. The node MCU, a minicomputer, is used to implement IoT in the proposed system. The converted output from the temperature sensor is given to Ain (analog input of Node MCU-Wi-Fi module). The ESP 8266 Wi-Fi module is used to transmit and receive data to local HTML page and to a remote server. The test is conducted on JA Solar, JAP6 (DG) 60-235 PV module to account (i) reference solar panel temperature (Tref) and (ii) tested solar panel temperature (T). The resultant logged temperature data are obtained through automatic monitoring on the cloud via IoT and are examined further to determine whether cooling is required or not. If T exceeds the reference value, the cooling mechanism is activated. Withal to data correlation developed, the solar panels under inspection can be endorsed for the rise in temperature to enhance the output efficiency and the life of solar module too. The comprehensive workflow representation of the proposed scheme with IoT is illustrated in Fig. 10.

Results and Discussion

The experimental evaluation of the proposed IoT-driven SES reported a temperature drop of 2 °C from the experiment analysis and an improvement of 1.6–2.99% in the output efficiency, which are calculated considering the power consumption of 1.996 W for values of T > 25 °C. In this case, the power consumption of 1.996 W of the cooling unit is calculated, which got triggered by the PID controller circuit and the temperature error generation circuit. The reduction in temperature will allow efficient capture of solar radiation, thereby improving net solar-to-electrical energy conversion efficiency. The results obtained from the pvsyst software are correlated, and temperature as a key parameter for lowering optimum yields is figured out. In context to this, a cooling mechanism to lower the module temperature is developed and connected to IoT. The mathematical analysis considering the power consumption of the cooling mechanism is done, and a comparative result analysis on the output efficiency is presented in Table 4. The cooling unit was in the non-activated mode for temperature less than 25 °C and a temperature greater than 40 °C. The updated efficiency was calculated for the complete month, and 390.99 W additional power was recorded and analyzed. Figure 11 illustrates everyday data of the entire month with updated efficiency and compares output efficiencies before and after applying the cooling setup. The days where the temperature of the solar panel did not exceed 25 °C, the cooling setup was in off state. At the outset of the research work, the return on investment (RoI) time was calculated with 2.44 INR per watt. With the addition of 390.99 W and approximate cost of setup 3000 INR, the RoI (time) for this work is around 3 months. We have assumed negligible energetic costs for operation and maintenance for the RoI calculations.

Conclusion and Performance Comparison

This research work presents a comprehensive smart solar energy system for smart cities. The data analysis contemplates temperature as the rudimentary hurdle behind low output conversion efficiencies and short life span resulting in high monetary valuation of PVCs. Accounting for the fact that, rise in temperature of solar module leads to power degradation in the output of the solar panels, as well as inflict life span of solar panel; the cooling/temperature maintenance mechanism becomes essential. The temperature-dependent IoT-driven cooling unit is modeled for improving output PCE of the solar module. The proposed smart SES design incorporates the following advantages: (i) comprehensive data prediction model: data analysis of SES for a complete year including nearly all aspects of performance parameters are made available; (ii) design simplicity: a simple low-powered cooling mechanism, which overall makes the system cost-effective, is proposed; (iii) outdo performance: considerable effect on the system efficiency is achieved by maintaining its optimal temperature as per need from the predictive data modeling; (iv) design automation: an IoT enabled and node MCU operated system is set up for monitor temperature, power consumption, and generation, send and display the temperature gradient/power loss in an easy and concise manner. Although the data prediction model is set up for JA Solar, JAP6 (DG)60-235 PV module driving load of Enphase, IQ6-60-x-240 grid inverter as a case study in this paper, the correlation matrix and data prediction modeling method can be pertained to other circuits by simply changing the power loss equations.

Acknowledgment

The authors would like to thank Mr. Amresh Mahajan, General Manager at ACME Cleantech Solution Private Limited, for the resourceful data and constant support and guidance.

References

References
1.
IEA
,
2015
, “
IEA—Key World Energy Statistics, 2015
”, https://www.iea.org/weo2017. Accessed February 21, 2018.
2.
Wong
,
K. V.
,
2016
, “
Sustainable Engineering in the Global Energy Sector
,”
ASME J. Energy Resour. Technol.
,
138
(
2
), p.
024701
.
3.
Smith
,
D. D.
,
Cousins
,
P.
,
Westerberg
,
S.
,
De Jesus-Tabajonda
,
R.
,
Aniero
,
G.
, and
Shen
,
Y. C.
,
2014
, “
Toward the Practical Limits of Silicon Solar Cells
,”
IEEE J. Photovolt.
,
4
(
6
), pp.
1465
1469
.
4.
Green
,
M. A.
,
Emery
,
K.
,
Hishikawa
,
Y.
, and
Warta
,
W.
,
2011
, “
Solar Cell Efficiency Tables (Version 37)
,”
Prog. Photovolt.: Res. Appl.
,
19
(
1
), pp.
84
92
.
5.
Garcıa
,
J. M.
,
Padilla
,
R. V.
, and
Sanjuan
,
M. E.
,
2017
, “
Response Surface Optimization of an Ammonia–Water Combined Power/Cooling Cycle Based on Exergetic Analysis
,”
ASME J. Energy Resour. Technol.
,
139
(
2
), p.
022001
.
6.
Sobri
,
S.
,
Koohi-Kamali
,
S.
, and
Rahim
,
N. A.
,
2018
, “
Solar Photovoltaic Generation Forecasting Methods: A Review
,”
Energy Convers. Manage.
,
156
, pp.
459
497
.
7.
Steinkemper
,
H.
,
Geisemeyer
,
I.
,
Schubert
,
M. C.
,
Warta
,
W.
, and
Glunz
,
S. W.
,
2017
, “
Temperature-Dependent Modeling of Silicon Solar Cells—EG, NI, Recombination, and VOC
,”
IEEE J. Photovolt.
,
7
(
2
), pp.
450
457
.
8.
Chawla
,
R.
,
Singal
,
P.
, and
Garg
,
A. K.
,
2018
, “
A Mamdani Fuzzy Logic System to Enhance Solar Cell Micro-Cracks Image Processing
,”
3D Res.
,
9
(
34
), pp.
1
12
.
9.
Dupré
,
O.
,
Vaillon
,
R.
, and
Green
,
M. A.
,
2016
, “
Experimental Assessment of Temperature Coefficient Theories for Silicon Solar Cells
,”
IEEE J. Photovolt.
,
6
(
1
), pp.
56
60
.
10.
Lu
,
X.
,
Zhao
,
Y.
,
Wang
,
Z.
,
Zhang
,
J.
, and
Song
,
Y.
,
2016
, “
Influence of Environmental Temperature and Device Temperature Difference on Output Parameters of C-Si Solar Cells
,”
Sol. Energy
,
136
, pp.
333
341
.
11.
Faiman
,
D.
,
2008
, “
Assessing the Outdoor Operating Temperature of Photovoltaic Modules
,”
Prog. Photovolt.: Res. Appl.
,
16
(
4
), pp.
307
315
.
12.
Ross
,
R. G.
Jr
, and
Smokler
,
M. I.
(
1986
).
Flat-Plate Solar Array Project. Vol. 6: Engineering Sciences and Reliability
. Jet Propulsion Lab., Pasadena, CA.
13.
Taylor
,
R. A.
, and
Solbrekken
,
G. L.
,
2008
, “
Comprehensive System-Level Optimization of Thermoelectric Devices for Electronic Cooling Applications
,”
IEEE Trans. Compon. Packag. Technol.
,
31
(
1
), pp.
23
31
.
14.
Singh
,
P.
, and
Ravindra
,
N. M.
,
2012
, “
Temperature Dependence of Solar Cell Performance—An Analysis
,”
Sol. Energy Mater. Sol. Cells
,
101
(
1
), pp.
36
45
.
15.
Cotfas
,
D. T.
,
Cotfas
,
P. A.
,
Ciobanu
,
D.
, and
Machidon
,
O. M.
,
2017
, “
Characterization of Photovoltaic–Thermoelectric–Solar Collector Hybrid Systems in Natural Sunlight Conditions
,”
J. Energy Eng.
,
143
(
6
), p.
04017055
. Y.
16.
Al-Hadban
,
K.
Sreekanth
,
H.
Al-Taqi
, and
R.
Alasseri
,
2018
.
Implementation of Energy Efficiency Strategies in Cooling Towersa Techno-Economic Analysis
,
ASME J. Energy Resour. Technol.
,
140
(
1
), p.
012001
.
17.
Sainthiya
,
H.
, and
Beniwal
,
N. S.
,
2019
, “
Efficiency Enhancement of Photovoltaic/Thermal (PV/T) Module Using Front Surface Cooling Technique in Winter and Summer Seasons: An Experimental Investigation
,”
ASME J. Energy Resour. Technol.
,
1
.
18.
He
,
W.
,
Zhou
,
J.
,
Chen
,
C.
, and
Ji
,
J.
,
2014
, “
Experimental Study and Performance Analysis of a Thermoelectric Cooling and Heating System Driven by a Photovoltaic/Thermal System in Summer and Winter Operation Modes
,”
Energy Convers. Manage.
,
84
(
10
), pp.
41
49
.
19.
Marandi
,
O. F.
,
Ameri
,
M.
, and
Adelshahian
,
B.
,
2018
, “
The Experimental Investigation of a Hybrid Photovoltaic-Thermoelectric Power Generator Solar Cavity-Receiver
,”
Sol. Energy
,
161
, pp.
38
46
.
20.
Soltani
,
S.
,
Kasaeian
,
A.
,
Sarrafha
,
H.
, and
Wen
,
D.
,
2017
, “
An Experimental Investigation of a Hybrid Photovoltaic/Thermoelectric System With Nanofluid Application
,”
Sol. Energy
,
155
, pp.
1033
1043
.
21.
Moradgholi
,
M.
,
Nowee
,
S. M.
, and
Abrishamchi
,
I.
,
2014
, “
Application of Heat Pipe in an Experimental Investigation on a Novel Photovoltaic/Thermal (PV/T) System
,”
Sol. Energy
,
107
, pp.
82
88
.
22.
Rosa-Clot
,
M.
,
Rosa-Clot
,
P.
, and
Tina
,
G. M.
,
2011
, “
TESPI: Thermal Electric Solar Panel Integration
,”
Sol. Energy
,
85
(
10
), pp.
2433
2442
.
23.
Hajji
,
M.
,
Labrim
,
H.
,
Benaissa
,
M.
,
Laazizi
,
A.
,
Ez-Zahraouy
,
H.
,
Ntsoenzok
,
E.
,
Meot
,
J.
, and
Benyoussef
,
A.
,
2017
, “
Photovoltaic and Thermoelectric Indirect Coupling for Maximum Solar Energy Exploitation
,”
Energy Convers. Manage.
,
136
(
1
), pp.
184
191
.
24.
Sauer
,
K. J.
, and
Roessler
,
T.
,
2012
, “
June. Systematic Approaches to Ensure Correct Representation of Measured Multi-Irradiance Module Performance in PV System Energy Production Forecasting Software Programs
,”
38th IEEE Photovoltaic Specialists Conference (PVSC)
,
IEEE
,
New York
, pp.
000703
000709
.
25.
Nguyen
,
H. T.
,
Baker-Finch
,
S. C.
, and
Macdonald
,
D.
,
2014
, “
Temperature Dependence of the Radiative Recombination Coefficient in Crystalline Silicon From Spectral Photoluminescence
,”
Appl. Phys. Lett.
,
104
(
11
), p.
112105
.
26.
Vainshtein
,
I. A.
,
Zatsepin
,
A. F.
, and
Kortov
,
V. S.
,
1999
, “
Applicability of the Empirical Varshni Relation for the Temperature Dependence of the Width of the Band Gap
,”
Phys. Solid State
,
41
(
6
), pp.
905
908
.
27.
Green
,
M. A.
,
2003
, “
General Temperature Dependence of Solar Cell Performance and Implications for Device Modelling
,”
Prog. Photovolt.: Res. Appl.
,
11
(
5
), pp.
333
340
.
28.
Uno
,
M.
, and
Tanaka
,
K.
,
2012
, “
Accelerated Charge–Discharge Cycling Test and Cycle Life Prediction Model for Supercapacitors in Alternative Battery Applications
,”
IEEE Trans. Ind. Electron.
,
59
(
12
), pp.
4704
4712
.
29.
Makkar
,
A.
,
Raheja
,
A.
,
Chawla
,
R.
, and
Gupta
,
S.
,
2019
, “
IoT Based Framework: Mathematical Modelling and Analysis of Dust Impact on Solar Panels
,”
3D Res.
,
10
(
3
), pp.
1
15
.
30.
Singh
,
N.
,
Dayama
,
P.
,
Randhawa
,
S.
,
Dasgupta
,
K.
,
Padmanaban
,
M.
,
Kalyanaraman
,
S.
, and
Hazra
,
J.
,
2017
, “
Photonic Energy Harvesting: Boosting Energy Yield of Commodity Solar Photovoltaic Systems via Software Defined IoT Controls
,”
Proceedings of the ACM Future Energy Systems
,
Shatin, Hong Kong
,
May 16–19
, ACM, pp.
56
66
.