0
Research Papers

# Experimental and Modeling Study on Solar System Using Linear Fresnel Lens and Thermoelectric ModulePUBLIC ACCESS

[+] Author and Article Information
Yavuz Köysal

Yeşilyurt Demir Çelik Vocational School,
Department of Electricity and
Energy Technologies,
Ondokuz Mayıs University,
Samsun 55330, Turkey
e-mail: yavuzk@omu.edu.tr

Ali Ekber Özdemir

Fatsa Faculty of Marine Sciences,
Marine Science and Technology Engineering,
Ordu University,
Ordu 52400, Turkey

Tahsin Atalay

Yeşilyurt Demir Çelik Vocational School,
Department of Electronics and Automation,
Ondokuz Mayıs University,
Samsun 55330, Turkey
e-mail: atalayt@omu.edu.tr

1Corresponding author.

Contributed by the Solar Energy Division of ASME for publication in the JOURNAL OF SOLAR ENERGY ENGINEERING: INCLUDING WIND ENERGY AND BUILDING ENERGY CONSERVATION. Manuscript received November 29, 2017; final manuscript received March 22, 2018; published online April 16, 2018. Assoc. Editor: Gerardo Diaz.

J. Sol. Energy Eng 140(6), 061003 (Apr 16, 2018) (11 pages) Paper No: SOL-17-1471; doi: 10.1115/1.4039777 History: Received November 29, 2017; Revised March 22, 2018

## Abstract

This work is concerned with an experimental design for generating power from thermoelectric generator (TEG) and linear Fresnel lens collector with one-axis solar tracking system. Main purpose of this experimental design is to measure the performance of the TEG with linear Fresnel lens collector. This work also aims to create a mathematical model by using adaptive neuro fuzzy inference system (ANFIS) model so that the electrical production estimates of the constructed system can be made for a given data set. For this reason, two individual systems, selective surface adapted for achieving medium temperature scale and nonselective surface for low temperatures, were constructed. There are two different coolant systems, which are passive and active, to create effective open circuit voltage values. Experimental results show that the maximum open circuit voltages were obtained as 0.442 V and 1.413 V for experimental system with selective surface adapted, as 0.341 V and 0.942 V with nonselective surface adapted when the received radiated power on Fresnel lens was measured nearly 625 W/m2 on average in the noon time. Experimental values were collected for the selective surface adapted system on 11th and 12th of September, 2017 and for nonselective surface on 13th of September, 2017, respectively, in Samsun/Turkey with location 41°14′N and 36°26′E. The collected data such as solar irradiation, wind speed, ambiance temperature, and open circuit voltage were used for (ANFIS) modeling. Obtained result shows that experimental calculations and modeling are consistent with each other.

<>

## Introduction and Background

Energy is an important factor for rural and residential areas in many fields from agriculture to industrialization, from health to the requirements of modern life. Especially for developed and developing countries, the sustainability of energy resources is very important. Energy policies of the countries have come to the forefront due to increasing environmental concerns, both from environmental pollution and the sufficiency of energy resources. Peura and Hyttinen [1] show the effect that humanity has produced in generating energy as “big picture” in their article. This picture is related with the dynamics between the environment and society.

Today, a large part of the energy requirement is still provided from fossil sources. But it is a fact that fossil fuels are finite. Environmental pollution increases day by day with the use of fossil fuels and this causes worries about the future of the world. These concerns led researchers to explore new sources of energy that are environmentally clear, namely renewable energy. There are five commonly used renewable energy sources which are clear: solar, wind, geothermal, hydropower, and biomass. Of course sun is the most intended energy source for clean energy production and so solar energy is arises out of alternative source of fossil fuels. It is possible to produce large-scale electrical power by using solar rays. Nowadays, solar applications are used in industrial and commercial sectors such as solar thermal electric power plants, solar heating systems, solar lighting, and photovoltaic. Especially, photovoltaic applications for generating electrical power by using solar energy are very popular academically.

###### The Importance of Effective Solar Radiation.

In order to generate electrical power, it is necessary to focus on sunlight in our system to create a usable temperature. Thus, it is vital to use effective solar concentrators. The concentration of sunlight in an area can limit surface of the material and reduce of cost of electricity. These solar concentrators can be listed as Fresnel lens (circular and linear), parabolic solar collectors, quantum dot concentrator, parabolic trough, and solar collector tubes. Khamooshi et al. [2] gives the types of solar photovoltaic concentrators, their technologies and their characteristics, and properties. Kumar et al. [3] mentioned solar concentrated types such as linear Fresnel reflector, central receiver, parabolic dish, and parabolic trough type and mainly focused on challenges and advantages of the Fresnel lens. A review by Xie et al. [4] gives information about imaging systems and nonimaging systems of concentrated solar energy systems using Fresnel lenses. In an experiment, Sobhansarbandi et al. [5] used an improved evacuated solar collector tube that was carbon nanotube sheet coating structure for collecting solar radiation and found that it was more effective than standard evacuated solar collector tubes. The obtained results show that carbon nanotube sheets improve the performance of solar collector by providing additional spectral absorption. Swanson [6] gave information about the role of concentrating solar systems in the future. In this review, advantages of concentrating over flat-plate systems are discussed under headings such as lower cost, superior efficiency, and higher annual capacity factor. It is possible to reach the desired temperature values with these mentioned solar concentrators. Many applications use medium temperature scale (350–450 K) for experimental designs [7]. In our previous work, Özdemir et al. [8] studied the solar thermoelectric generators (TEGs) with wind chimney that consists of an evacuated solar collector tube that was in the range of medium temperature scale. The obtained temperature values were observed near the 450 K.

It is known that Fresnel lens is a specially designed optical device, which converges light to a focal point or line with grooved structure. These optical devices are lightweight, thin, and cost-effective, alternative to conventional optical lenses. There are many types of linear concentrator technologies such as linear reflected Fresnel concentrator, parabolic trough solar concentrator, and linear refracted Fresnel lens. Zhai et al. [9] have analyzed a linear Fresnel lens experimentally. They have also modeled an evacuated tube heated by a Fresnel lens for making analysis. Sing et al. [10] have developed linear Fresnel reflector system for concentration of solar radiation and they studied the performance of developed system. Li and Wang [11] have used parabolic trough concentrator for concentrating solar radiation on evacuated tube to observe heating efficiency of the working fluids used. Al-Jumaily and Al-Kaysi [12] have used flat linear Fresnel lens with sun tracking system for analyzing thermal and optical losses in the constructed system. The concentrated solar radiation is directed to the thermoelectric systems in which photovoltaic panels or special designed thermoelectric generators used in electrical production. Thermoelectric generator used in our constructed system is a semiconductor material used for generating electricity.

###### Electrical Energy Source Thermoelectric Generator.

Thermoelectric generator is a specially designed solid device that converts thermal energy into electrical energy. This conversion is mainly based on Seebeck effect. In a TEG module, there are a number of semiconductor elements that are thermally parallel and electrically in series. TEG modules have many advantages such as being lightweight, noise-free, environmentally friendly, having nonmoving parts but low electrical efficiencies (about two or three percent). A standard TEG module has two sides, namely hot side and cold side. Generally, they are considered where the waste heat is present such as power plants, gas turbine plants, and heat engine plants. Orr and Akbarzadeh [13] have discussed the power generation from TEG where the waste heat exists such as exhaust gases of a car engine and an open-loop gas turbine power plant. Remeli et al. [14] designed and constructed a special system lab scale bench-top prototype for recovering industrial waste heat to generate electricity by using TEG. Yodovard et al. [15] have studied waste heat thermoelectric power generator model. The mentioned waste heat was provided from the diesel cycle and gas turbine cogeneration system. They have also analyzed system prices, system life spans, maintenance costs, etc. for the generating system.

There are a lot of papers in the literature related with TEG modules, which are experimental or theoretical. Date et al. [16] have combined TE module with heat pipes and analyzed the constructed system with experimental and theoretical methods. Liu et al. [17] have studied the optimization of thermoelectric generators by using computer-aided analysis. Eswaramoorthy et al. [18] work on a small-scale electrical generator that contains TE modules and solar parabolic dish.

###### Research Approach.

In our present study, we have analyzed the performance of the commercial TE module coupled with absorber surface adapted to copper plate and only copper plate (nonselective surface) under the usage of linear Fresnel lens. One axis solar tracking system was used for obtaining optimal concentration ratio with the used Fresnel lens. We have also carried out electrical performance of the experimental systems while the active cooling systems were used. A finned water cooled aluminum heat sink that contains coolant tunnels was used for active cooling system, which is used in heat sink of thermoelectric generator. The aluminum heat sink was placed on the cold side of TE module outside the cabinet. Running water was used for coolant with the flow rate of the coolant water, which was set to constant value of 10 g/s during the experiment. The cooling water is transferred to the heat sink tunnels by a pump to reduce the cold side temperature of the TEG. But neither pumping power nor related calculations are considered in this study.

## Methodology

The entire work is divided in to two stages. Stage 1 is related with the experimental calculations of the constructed system such as generated electrical open circuit voltages, Seebeck coefficient, maximum output power, and electrical efficiency. Stage 2 is related with creating a mathematical model by using adaptive neuro fuzzy inference system (ANFIS) model. Electrical generation capacity will be determined for any given data set by using created model. In addition, as an example of this stage, the electrical production values of the system were given for the entire July data by using the generated model. And also, interpretations of these calculations and modeling were presented in this stage.

###### Configuration of the System.

In this paper, electrical production of selected TE module was investigated by experimental methods. The constructed experimental system consists of commercially available TE modules (TEG1-12611-8.0) that are made of bismuth telluride with module size 56 × 56 mm, commercially available linear type Fresnel lens with size of 200 mm × 500 mm and focal distance of 200 mm, selective surface (with the dimension of 200 mm × 50 mm × 0.12 mm) adapted copper plate and glass wool thermal insulation materials. Integrated system is mounted on a one-axis solar tracking system. The schematic diagram of the constructed system is shown in Fig. 1. Cooling system of this setup with insulation material is shown in Fig. 2. In this experiment, a high temperature selective plate with the dimension of 200 mm × 50 mm × 0.12 mm was used to reach the intended temperature values. The heat emission coefficient and absorption coefficient for used selective plate were α = 5–7% and β = 95–97%, respectively. This selective surface was adapted to 2 mm thick copper plate that contains TEG module with the thermal paste so that there was no gap between the selective plate and copper plate. Temperature measurements are made with K-type thermocouples that are placed at linear focus point of the concentrated radiation, hot side and cold side of TEG module used, inlet and outlet of the cooling system. All the meteorological data such as solar radiation, wind speed, and ambient temperature were collected by meteorological data system that was installed on the rooftop of our education building.

In this setup, a linear polymethyl—methacrylate Fresnel lens was used for concentrating solar radiation onto surface of copper plate where the TEG was placed. For this reason, a linear Fresnel lens with size of 200 mm × 500 mm and focal distance of 200 mm was mounted on the top of the container box.

A copper plate with selective surface adapted and nonselective surface was placed to the bottom of the container box to the focal distance of used Fresnel lens. Heat sinks for active cooling were assisted with water flow channels. Finally, the whole system was adapted to one-axis sun tracking system. Related chart of the used TEG for open circuit voltage versus hot side temperature (Th) under various cold side temperature (Tc) is given in Fig. 3.

All these components of the experimental system were combined in an isolated container box. The constructed systems were assisted with solar tracking system for the best performance of generating electrical power. It is also important to adjust the tilt angle of the container box to the sun. The system must be mounted to at the angle of φ−15 for summer time where ϕ is the angular distance from the equator of the experiment location [20]. The container box is mounted to solar tracking system under tilt angle of 26 deg for obtaining optimal summer performance according to location of experimental system. The whole system of the construction is shown in Fig. 4.

The concentrating performance of a selected linear Fresnel as a solar collector that performed on generating electricity with thermo-electrical generator (TEG1-12611-8.0) is presented in this study. From these, the following electrical calculations have been carried out for the used TEG module.

###### Performance of Used Thermoelectric Generator Module.

It is possible to calculate some electrical quantities such as Seebeck coefficient (αTEG), maximum output power (Pmax), and electrical efficiency (ηe) by using the collected data from experimental constructions [2125].

The quantities mentioned above can be calculated by using the below equation: Display Formula

(1)$αTEG=VocTh−Tc$

where Voc is open circuit voltage, Th and Tc are hot and cold side temperature of the used TEG module used, respectively.

The current flowing in the circuit can be calculated with the given internal resistance and load resistance of used TE module. It is assumed that load resistance must be equal to internal resistance of TE module for calculating maximum power output (Pmaxout). This situation is called matched load. In this manner, it is also possible to calculate current flowing in the circuit theoretically Display Formula

(2)$I=VocRL+Rint$

In the equation above, RL is load resistance and Rint (1.8Ω) is internal resistance of the TE module. The maximum output power is given as Display Formula

(3)$Pmaxout=I2×RL$

The maximum output power can be expressed in matched load situation by substituting Eq. (2) into Eq. (3)Display Formula

(4)$Pmaxout=Voc24Rint$

The electrical efficiency (ηe) can be calculated by using incident power (Pin). Incident power is obtained by mean value of solar flux irradiation. Solar flux irradiation was assumed to be nearly 625 W/m2 for the days when the experimental measurements were taken in the noon time so that the incident power can be expressed as follows: Display Formula

(5)$Pin=Isfi×A×X$

where “A” is the area of selective surface on the TE module. “X” is calculated as 20. That is the concentration ratio of used Fresnel lens and calculated with aperture area and absorber area. Electrical efficiency (ηe) can be expressed with the following equation by using obtained and calculated parameters: Display Formula

(6)$ηe=PmaxoutPin×100$

Another important concept is thermal resistance of the used TEG module. Thermal resistance (RT) represents the temperature rise per unit rate of heat dissipation, measured with (°C/W) Display Formula

(7)$RT=ΔTQ$

where ΔT is the temperature difference and Q is dissipation in W. Nowadays, thermal resistance is specified by the manufacturers but in our experimental system it is observed that Q and ΔT quantities are changing continuously with respect to the sun's location activity which also changes RT quantity.

## Results and Discussions

###### Thermal Resistance, Generated Open Circuit Voltage, Maximum Power Output, and Electrical Efficiency of Constructed System.

Thermal resistance around the located TEG module is an important indicator that shows the performance of generating power of generator system. Figure 5 gives information about variation of thermal resistance of the used generator as a function of solar irradiation flux related with experimental time. This figure shows that thermal resistance of the both sides of the system, with active and passive cooling sides, increasing solar irradiation flux decreases the thermal resistance of the generator system.

The generator system having active cooling has a higher thermal resistance than the passive system. Thermal resistance values have the highest values at the end of experiment. Also, it is shown that thermal resistance values have the lowest values when the solar flux is maximum.

Open circuit voltage and maximum power output indicators are an important aspect of how the generator works. Open circuit voltage and therefore power output values are related with the TEG module systems, shown in Fig. 6. The efficiency of the TEG module depends on the continuous and uniform temperature difference between the cold and hot surfaces. For this reason, cooling systems were integrated with generator systems. In order to have more efficiency, active cooling system with coolant tunnels was used. Thus, it is possible to make a comparison between active and passive systems. In both systems, the open circuit voltage increases and decreases with increasing and decreasing solar flux. The maximum obtained open circuit voltage values are 1.413 V and 0.942 V for both of the active cooling systems with selective surfaces and nonselective surfaces, respectively, in the afternoon time. The generated open circuit voltage values were higher in the systems with selective surfaces due to the wide scale of temperature gradient. The temperature difference between the hot side and cold side of TEG module was reached at 48 °C and 38.4 °C for active cooling systems, respectively, when the open circuit voltages were maximum. The calculated power outputs were obtained as 0.277 W and 0.123 W for the maximum open circuit voltage values. Under these conditions, Seebeck coefficients were obtained as 0.029 V/°C and 0.003 V/°C, for the systems that contain active cooling, respectively.

In Fig. 7, it is shown that electrical efficiencies were increasing with the start of collection of experimental data for the active cooling systems. When the solar flux continued to rise, it started to decrease in the morning time. Increasing solar irradiation flux decreases the electrical efficiencies of the active cooling sides of the generator. It started to rise again in the afternoon time. This is interpreted as the generator system temperatures are stabilized after the noon time. These conditions were not observed in passive cooling side of the generator and electrical efficiencies were nearly the same values for data collection process. The calculated maximum electrical efficiencies were 0.595% and 0.22% for active cooling sides of selective surface and nonselective surface, respectively.

Figure 8 gives information about variation of electrical power according to hot side and cold side temperatures of the TEG module. In this figure, red line represents hot side temperature and blue line represents cold side temperature of the module. Black line represents electrical power output of the TEG. It is shown in this figure, that when the hot side and cold side temperatures are near to each other, electrical power is obtained in a very small amount, but as the difference between the red and blue lines increases, the electrical power increases. Maximum value of the electrical power was obtained as nearly 0.273 W in the afternoon time of 12th of September. Some calculated parameters that belong to experimentally collected data were given in detail in Tables 1 and 2.

###### System Modeling.

There are many mathematical models about TEG in the literature [2629]. These models are used for prediction of system behavior under changing of selected parameters. Development of any mathematical model requires many complex calculations. However, artificial neural network structures have a learning and generalization capacity without any complex mathematical calculations. An obtained data set, which is divided into two parts called training and testing, is used for construction of neural network structure. Training data set is used for tuning of network parameters and this stage is called training stage.

Training capacity is accomplished by testing data set for evaluation of network. In this paper, a well-known artificial neural network structure that is called as ANFIS was used for modeling of TEG based energy production system. ANFIS has been used as a powerful tool in system modeling for many years. It was developed in 1993 by Jang [30]. ANFIS is based on fuzzy inference system and its general structure can be seen in Fig. 9.

Artificial neural network structures have layers. Structures of these layers vary in accordance with the features of neural network structures. For example, multilayer perceptron network can contain many layers that resemble to each other. However, radial basis function neural network has only one layer. ANFIS has a special structure including many layers.

Layer 1: This layer is called as the fuzzifier layer and produces adaptive outputs. These adaptive outputs come from membership functions. In this paper, membership functions used are Gaussian functions.

Layer 2: In this layer, outputs of membership functions are combined via some basic operators (multiplication, max, min, etc.).

Layer 3: This layer is called as normalization layer. Outputs of layer 2 are normalized in this layer.

Layer 4: The normalized values of the layer 3 are weighed in the layer 4.

Layer 5: In this layer, all outputs of layer 4 are summed for obtaining final values of the ANFIS output. The details of ANFIS can be found in Ref. [28].

In this paper, solar radiation (W/m2), ambient temperature (° C), and wind velocity (m/s) are used as input parameters of ANFIS. Open circuit voltages of system were selected as output parameters of ANFIS. Data collected on 12th and 11th of September were used for the training and testing stages of ANFIS, respectively. In this work, the ANFIS rules have been selected with subtractive clustering. Training parameters are selected as range of influence = 0.9, squash factor = 1.25, accept ratio = 0.5, reject ratio = 0.15, and epoch = 100. At the end of training stage, constructed Sugeno—type ANFIS structure is presented in Fig. 10.

First step of training stage is about tuning of membership functions parameters that are centers and widths. Tuned membership functions of constructed ANFIS structure are shown in Fig. 11.

The experimental data and output of the ANFIS for the training and testing processes are shown in Fig. 12.

In Fig. 12, obtained mean squared errors for training and testing stage are 0.036 and 0.0804, respectively. This means that testing error is about 5.3%, which is an acceptable error ratio. Therefore, constructed ANFIS model can be used for the prediction of system output under the different meteorological data in any day of the year with 5.3% prediction error. As shown in Fig. 10, the used ANFIS structure includes only two rules. It means that, unlike many mathematical models, the computational complexity of ANFIS model is too low. As already mentioned, produced output matched power of single TEG module is calculated by equality (4). After the establishment of the ANFIS structure, this structure can be used for the estimation of daily average output matched power. According to figure mentioned above, daily average output matched powers estimated by ANFIS for July are shown with %5.3 error ratio in Fig. 13.

A simple method is used for showing correlation between input parameters and ANFIS outputs. Daily normalized input parameters (daily average solar radiation, daily average wind speed, and daily average ambiance temperature) and ANFIS outputs (daily average output matched powers) were used in this simple method. Monthly maximum values of input parameter were used for normalization. After this stage, all normalized input parameters were multiplied with each other for each day of July. Multiplied input parameters and ANFIS outputs are shown in Fig. 14.

In Fig. 14, it is clearly seen that there is a correlation between ANFIS outputs and multiplied normalized input parameters. These results show that ANFIS model is reliable and can be used for prediction of daily average matched load power for any given day of a year with meteorological data.

## Conclusions and Future Work

In this experimental construction, we aimed to investigate electrical quantities such as open circuit voltage, electrical power output, electrical efficiency, and thermal resistance of the system. Two different experimental systems, which were selective surface adapted and nonselective surface, were constructed for obtaining and comparing electrical data. Also, constructed systems were placed to one axis solar tracking system on those experimental days for obtaining maximum efficiency. The achieved maximum temperatures were 73 °C for active cooling side and 89 °C for passive cooling side on experimental day 12th of September for the selective adapted system, while they were 66 °C for active cooling side and 81 °C for passive cooling side on experimental day 13th of September for the system nonselective surface. The maximum achieved electrical open circuit voltages were 1.413 V and 0.942 V for the obtained temperatures values, respectively. The obtained voltages that were lower are related to cooling side that only contains heat sink and the obtained voltages that were higher are related to active cooling system that contains heat sink with cooling tunnels for the coolant. The obtained open circuits voltages and matched load powers are in small scale because of using single module TEG. The main aim of this study is to investigate the generation of electrical power production capacity of constructed system. However, this system is open to any improvement by increasing the number of TEG modules on the system for large-scale power generation. Besides, the ANFIS model was used as a tool for estimating the output parameters of the system. And also, constructed model can predict the system outputs by using meteorological data with the error of 5.3% for a given day of a year. This model can be used for the improvement of the presented experimental system.

In the future, the construction of systems using different concentrators that mentioned in Sec. 1 is planned so that it is possible to compare different systems with each other. It will be interesting to use many more TEG modules to see what the available electricity generation is. Additionally, the constructed systems can be modeled by neural network structures to predict generated power.

## Funding Data

• Ondokuz Mayıs University Research Fund Grant (PYO.YMY.1901.16.001).

## Nomenclature

• I =

short circuit current, A

• Isfi =

• Pin =

incident power, W

• Pmax =

output power, W

• Q =

heating power of the system, W

• RL =

• RT =

thermal resistance of the used TE module, °C/W

• Rint =

internal resistance of the TE module, Ω

• Tc =

thermoelectric cold side temperature, °C

• Th =

thermoelectric hot side temperature, °C

• Voc =

open circuit voltage, V

• αTEG =

Seebeck coefficient

• ηe =

electrical efficiency, %

## References

Peura, P. , and Hyttinen, T. , 2011, “ The Potential and Economics of Bioenergy in Finland,” J. Cleaner Prod., 19(9–10), pp. 927–945.
Khamooshi, M. , Salati, H. , Egelioğlu, F. , Faghiri, A. H. , Tarabishi, J. , and Babadi, S. , 2014, “ A Review of Solar Photovoltaic Concentrators,” Int. J. Photoenergy, 2014, p. 958521.
Kumar, V. , Shrivastava, R. L. , and Untawale, S. P. , 2015, “ Fresnel Lens: A Promising Alternative of Reflectors in Concentrated Solar Power,” Renewable Sustainable Energy Rev., 44, pp. 376–390.
Xie, W. T. , Dai, Y. J. , Wang, R. Z. , and Sumathy, K. , 2011, “ Concentrated Solar Energy Applications Using Fresnel Lenses: A Review,” Renewable Sustainable Energy Rev., 15(6), pp. 2588–2606.
Sobhansarbandi, S. , Martinez, P. M. , Papadimitratos, A. , Zakhidov, A. , and Hassanipour, F. , 2017, “ Evacuated Tube Solar Collector With Multifunctional Absorber Layers,” Sol. Energy, 146, pp. 342–350.
Swanson, R. M. , 2000, “ The Promise of Concentrators,” Prog. Photovolt. Res. Appl., 8(1), pp. 93–111.
Rowe, D. M. , ed., 1995, Handbook of Thermoelectrics, CRC Press, Boca Raton, FL.
Özdemir, A. E. , Köysal, Y. , Özbas, E. , and Atalay, T. , 2015, “ The Experimental Design of Solar Heating Thermoelectric Generator With Wind Cooling Chimney,” Energy Convers. Manage., 98, pp. 127–133.
Zhai, H. , Dai, Y. J. , Wu, J. Y. , Wang, R. Z. , and Zhang, L. Y. , 2010, “ Experimental Investigation and Analysis on a Concentrating Solar Collector Using Linear Fresnel Lens,” Energy Convers. Manage., 51(1), pp. 48–55.
Singh, P. L. , Ganesan, S. , and Yàdav, G. C. , 1999, “ Performance Study of a Linear Fresnel Concentrating Solar Device,” Renewable Energy, 18(3), pp. 409–416.
Li, M. , and Wang, L. L. , 2006, “ Investigation of Evacuated Tube Heated by Solar Trough Concentrating System,” Energy Convers. Manage., 47(20), pp. 3591–3601.
Al-Jumaily, K. E. J. , and Al-Kaysi, M. A. K. A. , 1998, “ The Study of the Performance and Efficiency of Flat Linear Fresnel Lens Collector With Sun Tracking System in Iraq,” Renewable Energy, 14(1–4), pp. 41–48.
Orr, B. , and Akbarzadeh, A. , 2017, “ Prospects of Waste Heat Recovery and Power Generation Using Thermoelectric Generators,” Energy Procedia, 110, pp. 250–255.
Remeli, M. F. , Date, A. , Orr, B. , Ding, L. C. , Singh, B. , Affandi, N. D. N. , and Akbarzadeh, A. , 2016, “ Experimental Investigation of Combined Heat Recovery and Power Generation Using a Heat Pipe Assisted Thermoelectric Generator System,” Energy Convers. Manage., 111, pp. 147–157.
Yodovard, P. , Khedari, J. , and Hirunlabh, J. , 2001, “ The Potential of Waste Heat Thermoelectric Power Generation From Diesel Cycle and Gas Turbine Cogeneration Plants,” Energy Sources, 23(3), pp. 213–224.
Date, A. , Date, A. , Dixon, C. , and Akbarzadeh, A. , 2014, “ Theoretical and Experimental Study on Heat Pipe Cooled Thermoelectric Generators With Water Heating Using Concentrated Solar Thermal Energy,” Sol. Energy, 105, pp. 656–668.
Liu, Z. , Zhu, S. , Ge, Y. , Shan, F. , Zeng, L. , and Liu, W. , 2017, “ Geometry Optimization of Two-Stage Thermoelectric Generators Using Simplified Conjugate-Gradient Method,” Appl. Energy, 190, pp. 540–552.
Eswaramoorthy, M. , Shanmugam, S. , and Veerappan, A. R. , 2013, “ Experimental Study on Solar Parabolic Dish Thermoelectric Generator,” Int. J. Energy Eng. (IJEE), 3(3), pp. 62–66.
TEC, 2017, “ Specifications TEG Module TEG1-12611-8.0,” TECTEG MFR, Aurora, ON, Canada, accessed Aug. 13, 2017,
Roger, A. , 2004, Messenger and Jerry Ventre, Photovoltaic Systems Engineering, CRC Press, London.
Nia, M. H. , Nejad, A. A. , Goudarzi, A. , Valizadeh, M. , and Samadian, P. , 2014, “ Cogeneration Solar System Using Thermoelectric Module and Fresnel Lens,” Energy Convers. Manage., 84, pp. 305–310.
CATRENE Working Group on Energy Autonomous Systems, 2009, Energy Autonomous Systems: Future Trends in Devices, Technology, and Systems, CATRENE Leuven, Paris, France.
Mastbergen, D. , 2008, “ Development and Optimization of a Stove-Powered Thermoelectric Generator,” Ph.D. thesis, Colorado State University, Fort Collins, CO.
Beeby, S. , and White, N. M. , 2010, Energy Harvesting for Autonomous Systems, Artech House, Boston, MA.
Candadai, A. A. , Kumar, V. P. , and Barshilia, H. C. , 2016, “ Performance Evaluation of a Natural Convective-Cooled Concentration Solar Thermoelectric Generator Coupled With a Spectrally Selective High Temperature Absorber Coating,” Sol. Energy Mater. Sol. Cells, 145(Pt. 3), pp. 333–341.
Miljkovic, N. , and Wang, E. N. , 2011, “ Modeling and Optimization of Hybrid Solar Thermoelectric Systems With Thermosyphons,” Sol. Energy, 85(11), pp. 2843–2855.
Xiao, J. , Yang, T. , Li, P. , Zhai, P. , and Zhang, Q. , 2012, “ Thermal Design and Management for Performance Optimization of Solar Thermoelectric Generator,” Appl. Energy, 93, pp. 33–38.
Shanmugam, S. , Eswaramoorthy, M. , and Veerappan, A. R. , 2014, “ Modeling and Analysis of a Solar Parabolic Dish Thermoelectric Generator,” Energy Sources, Part A: Recovery, Util., Environ. Eff., 36(14), pp. 1531–1539.
Chen, W. H. , Wang, C. C. , Hung, C. I. , Yang, C. C. , and Juang, R. C. , 2014, “ Modeling and Simulation for the Design of Thermal-Concentrated Solar Thermoelectric Generator,” Energy, 64, pp. 287–297.
Jang, J. S. R. , 1993, “ ANFIS Adaptive-Network-Based Fuzzy Inference Systems,” Man Cybern., 23(3), pp. 665–685.
View article in PDF format.

## References

Peura, P. , and Hyttinen, T. , 2011, “ The Potential and Economics of Bioenergy in Finland,” J. Cleaner Prod., 19(9–10), pp. 927–945.
Khamooshi, M. , Salati, H. , Egelioğlu, F. , Faghiri, A. H. , Tarabishi, J. , and Babadi, S. , 2014, “ A Review of Solar Photovoltaic Concentrators,” Int. J. Photoenergy, 2014, p. 958521.
Kumar, V. , Shrivastava, R. L. , and Untawale, S. P. , 2015, “ Fresnel Lens: A Promising Alternative of Reflectors in Concentrated Solar Power,” Renewable Sustainable Energy Rev., 44, pp. 376–390.
Xie, W. T. , Dai, Y. J. , Wang, R. Z. , and Sumathy, K. , 2011, “ Concentrated Solar Energy Applications Using Fresnel Lenses: A Review,” Renewable Sustainable Energy Rev., 15(6), pp. 2588–2606.
Sobhansarbandi, S. , Martinez, P. M. , Papadimitratos, A. , Zakhidov, A. , and Hassanipour, F. , 2017, “ Evacuated Tube Solar Collector With Multifunctional Absorber Layers,” Sol. Energy, 146, pp. 342–350.
Swanson, R. M. , 2000, “ The Promise of Concentrators,” Prog. Photovolt. Res. Appl., 8(1), pp. 93–111.
Rowe, D. M. , ed., 1995, Handbook of Thermoelectrics, CRC Press, Boca Raton, FL.
Özdemir, A. E. , Köysal, Y. , Özbas, E. , and Atalay, T. , 2015, “ The Experimental Design of Solar Heating Thermoelectric Generator With Wind Cooling Chimney,” Energy Convers. Manage., 98, pp. 127–133.
Zhai, H. , Dai, Y. J. , Wu, J. Y. , Wang, R. Z. , and Zhang, L. Y. , 2010, “ Experimental Investigation and Analysis on a Concentrating Solar Collector Using Linear Fresnel Lens,” Energy Convers. Manage., 51(1), pp. 48–55.
Singh, P. L. , Ganesan, S. , and Yàdav, G. C. , 1999, “ Performance Study of a Linear Fresnel Concentrating Solar Device,” Renewable Energy, 18(3), pp. 409–416.
Li, M. , and Wang, L. L. , 2006, “ Investigation of Evacuated Tube Heated by Solar Trough Concentrating System,” Energy Convers. Manage., 47(20), pp. 3591–3601.
Al-Jumaily, K. E. J. , and Al-Kaysi, M. A. K. A. , 1998, “ The Study of the Performance and Efficiency of Flat Linear Fresnel Lens Collector With Sun Tracking System in Iraq,” Renewable Energy, 14(1–4), pp. 41–48.
Orr, B. , and Akbarzadeh, A. , 2017, “ Prospects of Waste Heat Recovery and Power Generation Using Thermoelectric Generators,” Energy Procedia, 110, pp. 250–255.
Remeli, M. F. , Date, A. , Orr, B. , Ding, L. C. , Singh, B. , Affandi, N. D. N. , and Akbarzadeh, A. , 2016, “ Experimental Investigation of Combined Heat Recovery and Power Generation Using a Heat Pipe Assisted Thermoelectric Generator System,” Energy Convers. Manage., 111, pp. 147–157.
Yodovard, P. , Khedari, J. , and Hirunlabh, J. , 2001, “ The Potential of Waste Heat Thermoelectric Power Generation From Diesel Cycle and Gas Turbine Cogeneration Plants,” Energy Sources, 23(3), pp. 213–224.
Date, A. , Date, A. , Dixon, C. , and Akbarzadeh, A. , 2014, “ Theoretical and Experimental Study on Heat Pipe Cooled Thermoelectric Generators With Water Heating Using Concentrated Solar Thermal Energy,” Sol. Energy, 105, pp. 656–668.
Liu, Z. , Zhu, S. , Ge, Y. , Shan, F. , Zeng, L. , and Liu, W. , 2017, “ Geometry Optimization of Two-Stage Thermoelectric Generators Using Simplified Conjugate-Gradient Method,” Appl. Energy, 190, pp. 540–552.
Eswaramoorthy, M. , Shanmugam, S. , and Veerappan, A. R. , 2013, “ Experimental Study on Solar Parabolic Dish Thermoelectric Generator,” Int. J. Energy Eng. (IJEE), 3(3), pp. 62–66.
TEC, 2017, “ Specifications TEG Module TEG1-12611-8.0,” TECTEG MFR, Aurora, ON, Canada, accessed Aug. 13, 2017,
Roger, A. , 2004, Messenger and Jerry Ventre, Photovoltaic Systems Engineering, CRC Press, London.
Nia, M. H. , Nejad, A. A. , Goudarzi, A. , Valizadeh, M. , and Samadian, P. , 2014, “ Cogeneration Solar System Using Thermoelectric Module and Fresnel Lens,” Energy Convers. Manage., 84, pp. 305–310.
CATRENE Working Group on Energy Autonomous Systems, 2009, Energy Autonomous Systems: Future Trends in Devices, Technology, and Systems, CATRENE Leuven, Paris, France.
Mastbergen, D. , 2008, “ Development and Optimization of a Stove-Powered Thermoelectric Generator,” Ph.D. thesis, Colorado State University, Fort Collins, CO.
Beeby, S. , and White, N. M. , 2010, Energy Harvesting for Autonomous Systems, Artech House, Boston, MA.
Candadai, A. A. , Kumar, V. P. , and Barshilia, H. C. , 2016, “ Performance Evaluation of a Natural Convective-Cooled Concentration Solar Thermoelectric Generator Coupled With a Spectrally Selective High Temperature Absorber Coating,” Sol. Energy Mater. Sol. Cells, 145(Pt. 3), pp. 333–341.
Miljkovic, N. , and Wang, E. N. , 2011, “ Modeling and Optimization of Hybrid Solar Thermoelectric Systems With Thermosyphons,” Sol. Energy, 85(11), pp. 2843–2855.
Xiao, J. , Yang, T. , Li, P. , Zhai, P. , and Zhang, Q. , 2012, “ Thermal Design and Management for Performance Optimization of Solar Thermoelectric Generator,” Appl. Energy, 93, pp. 33–38.
Shanmugam, S. , Eswaramoorthy, M. , and Veerappan, A. R. , 2014, “ Modeling and Analysis of a Solar Parabolic Dish Thermoelectric Generator,” Energy Sources, Part A: Recovery, Util., Environ. Eff., 36(14), pp. 1531–1539.
Chen, W. H. , Wang, C. C. , Hung, C. I. , Yang, C. C. , and Juang, R. C. , 2014, “ Modeling and Simulation for the Design of Thermal-Concentrated Solar Thermoelectric Generator,” Energy, 64, pp. 287–297.
Jang, J. S. R. , 1993, “ ANFIS Adaptive-Network-Based Fuzzy Inference Systems,” Man Cybern., 23(3), pp. 665–685.

## Figures

Fig. 1

Schematic diagram of the experimental system

Fig. 2

Active and passive cooling system view of experimental system with insulation material (left). Internal view of selective adapted copper plate in the container box (right).

Fig. 3

The chart of the used TEG module that given by manufacturer. The chart for open circuit voltage versus hot and cold side temperatures of the used TEG [19].

Fig. 4

Constructed experimental setups. The left one is nonselective surface, the right one is adapted to selective surface.

Fig. 5

Thermal resistance of experimental systems: (a) the graph of selective surface adapted system and (b) the graph of the system nonselective surface

Fig. 6

Variation of open circuit voltage of used TEG modules with solar irradiance and experimental time. (a) the graph of higher voltage values is related with selective surface adapted system and (b) the graph related to lower voltage values of the system nonselective surface.

Fig. 7

Variation of Electrical efficiency of experimental systems with solar irradiance and experimental time: (a) the graph of selective surface adapted system and (b) the graph of the system nonselective surface

Fig. 8

Variation of electrical power of experimental systems with hot and cold side of the used TEG module: (a) the graph of selective surface adapted system and (b) the graph of the system nonselective surface

Fig. 9

General structure of ANFIS and its layers

Fig. 10

Used ANFIS structure

Fig. 11

Membership functions of used ANFIS network

Fig. 12

The experimental data and output of the ANFIS, (a) for training data and (b) for testing data

Fig. 13

Estimated output matched load powers by ANFIS for July

Fig. 14

Correlation input parameters and ANFIS outputs for days of July

## Tables

Table 1 Some calculated parameters for active and passive cooling sides of selective surface adapted system on experimental day of 12th of September
Table 2 Some calculated parameters for active and passive cooling sides of nonselective surface system on experimental day of 13th of September

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections