Elsevier

Renewable Energy

Volume 156, August 2020, Pages 57-67
Renewable Energy

Implicit regression-based correlations to predict the back temperature of PV modules in the arid region of south Algeria

https://doi.org/10.1016/j.renene.2020.04.073Get rights and content

Highlights

  • Empirical models of PV back temperature are developed for Algerian Sahara Region.

  • PV temperature is correlated to air temperature and global irradiance.

  • Used data from two stations at Adrar city of hot and arid climate.

  • Best performing models are identified based on statistical analysis.

Abstract

The determination of the PV module temperature is a key parameter for the assessment of the actual performance of the PV systems. The application of available models for PV module temperature estimation in literature can be verified, but the application of these correlations for different climate conditions does not lead to unequivocal results. The main objective of this study is to suggest new empirical models for estimating the back surface module temperature under outdoor hot dry climatic conditions of Adrar province (Algerian Sahara) and to compare the developed models to different existing models in the literature. The models are developed based on meteorological and irradiance data collected from two different plants with different module technologies. The best site-specific approach uses a simple formula to derive the PV-back module temperature from the meteorological variables such as ambient temperature, and irradiance. The relative root mean square error and the Pearson’s correlation coefficient of the best developed model are 10.662% and 0.955, respectively. In addition, MAPE and RMSE values are considerably small for the studied stations. A general model for predicting the PV-back temperature was also recommended for simple PV modules or open rack systems in rural locations with no measurement equipment nearby. The results are quite useful for studying PV system performance and estimating its energy output.

Introduction

Commonly used resources of electric power generation nowadays, i.e. coal, oil, natural gas, and nuclear fuels, are causing a noticeable globally harmful impact on the environment such as the release of greenhouse gases or the discharges of radioactive wastes [1]. Renewable energy can overcome these forms of damage to the environment while contributing to sustainable development. Among these renewable energy sources, photovoltaic (PV) systems are among the most promising technologies that have been proven to be reliable and cost-efficient and are expected to be deployed widely in the few next decades. According to the IEA (International Energy Agency) mid-term renewable energy market report [2], only onshore wind and PV are driving the majority of renewable energy growth till the end of this decade. They are also the only two technologies that are on track to reach the 2DS (2 °C) goal before 2025. Compared to 2015, global solar PV power production is projected to be doubled by 2020, driven by policy support and cost reduction, especially in countries like China, US, and Japan. According to the IEA PVPS (Photovoltaic Power Systems Programme) mid-decade report [3], the installed photovoltaic power generation capacity has been significantly increased for several years at the global level, and the prospect is for growth of about 40 GW/year from 2015 to 2020 to meet the target of installing 600 GW of capacity by 2025. Late 2014, the worldwide installed solar PV power was in the order of 177 GW with shares of 51, 36, 12, and 1% in Europe, Asia, Americas, and Africa, accordingly [4]. One of the reasons for the growth in the solar PV market has been attributed to the steady decline in the price of photovoltaic systems, namely decreased by 50% between 2010 and 2014 [5]. Nevertheless, this decline in the costs provides new opportunities to electrify millions of locations around the world who have never been benefited before. In fact, solar PV systems are being continuously recognized as an energy- and cost-efficient way of supplying electricity to remote locations, rather than extending electrical grids. According to the aforementioned IEA reports, in developing countries (Latin America and Africa, in particular), the challenge of providing generous amounts of electricity for lights and appliances, including the Internet, will speed up the expansion of the PV market in the near future.

Algeria is one of the North African Mediterranean countries, which has tremendous potential for solar energy applications [6]. According to satellite sunshine data, Algeria is considered one of the sunniest countries in the world. For that reason, the potential of solar resources is optimal for the execution of solar projects [7]. A program supporting the development of renewable energy, namely: Promotion of the Renewable Energies for Sustainable Development (Act of 17 August 2004), has been initiated to increase the overall share of renewable energy sources in the final gross energy consumption to 22 and 37% by 2020 and 2030, accordingly [8,9]. To achieve this, the Algerian government has shown its interest to actively participate in the renewable energy sector’s growth in the next few years and encourage the installation of new PV power generation systems all over the country. Over 70% of the Algerian territory has relatively sufficient and useful solar irradiation that can be easily utilized to promote the development of solar-based energy systems (e.g. the Algerian Sahara, located at north-central Africa region) [10].

The Electrical performance of a PV system depends on the operating temperature of the modules, which plays a central role in the PV conversion process because it affects the basic electrical quantities [11,12]. The back temperature of the module depends on the combined effects of astronomical and meteorological events [13,14]. In the literature, it has been demonstrated that, for every 1 °C increase in PV module temperature, a corresponding 0.3–0.5% decrease in the module’s efficiency takes place [15,16]. Therefore, for the optimum design of PV power systems, it is desirable to measure their performances at the installation site. Measuring module temperature is the first step in estimating cell temperature. However, in most cases, there are no local measurements at the specific site of installation and approximate methods to predict the back temperature of PV modules are used instead. In order to estimate the modules’ back temperature, many models have been suggested to correlate the back temperature of PV modules to the ambient temperature. The first model was suggested by Ross and Smokler [17] in the form of linear expression, based on simple heat transfer concepts. The functional form of this model has been extensively utilized in so many studies. David and Fanney [18] extended this model by including ambient air temperature, incident irradiance on the plane of the module, and wind speed as explanatory variables. Then, the Ross-Smokler model was developed by many researchers, who have correlated the module temperature with some meteorological parameters such as ambient temperature, relative humidity, and wind velocity [[19], [20], [21]].

Despite the relatively old approach of empirical modeling of PV modules’ back temperature, the interest in developing such models is continuous in last few years due to the great spread of PV technology and the fact that those models are handy for non-specialists and do not require deep knowledge on the physics of the PV cells and/or the atmospheric constituents. Most notably, Obiwulu et al. [22] developed and analyzed the prediction accuracy of 48 empirical models for PV modules tilted at four different angles (12 models for each tilt angle) at Lagos (Nigeria). The prediction accuracy and the type of inputs were dependent on the tilt and angle of the modules. Coskun et al. [23] used collected data from a PV plant in Turkey to evaluate the sensitivity of the prediction accuracy of 17 empirical correlations suggested in the literature to the ambient temperature, wind speed, and global irradiance. Based on the results, the evaluated correlations were adjusted and 12 new correlations were suggested. Muzathik [24] proposed a simple multi-linear correlation for estimating the back temperature based on ambient temperature, wind speed, and irradiance, which showed a non-dimensional root mean square error of 0.9763 when tested using data of a small plant at Kuala Terengganu, Malaysia.

Algerian Sahara weather is hot and dry throughout the year, which is favorable for large-scale implementations of PV power generation systems. The location is a candidate to host many solar energy projects. Adrar district is considered a representative site of the Algerian Sahara as long-term ground measurements are available from several small-scale solar PV power plants station in the region. Despite the potential of large-scale installations of PV plants in North Africa and although considerable research has been devoted to in the energy field in the Algerian Sahara and North Africa (MENA) region [[25], [26], [27], [28], [29], [30], [31], [32], [33]], no models have been developed to predict the PV operating temperatures under such hot arid climates, including the location studied here. The scope of this study is to develop and statistically compare implicit correlations of PV operating temperature based on meteorological and solar radiation data at the province of Adrar in Algeria. Despite being site-dependent, the developed models are expected to well-extrapolate (with acceptable errors) to nearby locations in the Sahara Region due to the relatively more uniform spatial distribution of solar radiation, ambient temperature, and relative humidity (compared to e.g. Mediterranean or European locations). Hence, these models can be used at an early stage for planning PV systems under such climate, or for simulating existing systems for overall performance assessment.

Section snippets

Descriptions of the study site

Algeria is located in the north-central portion of Africa, between longitudes of 9° and 12° E, and latitudes of 19° and 37° N. It is the largest country in Africa with an area of 2,381,741 km2, more than four-fifths of which is desert, and the climate varies considerably from north to south borders. The Algerian part of the Sahara Desert features a hot desert climate, where the sky is usually clear and the sunshine duration is extremely high (around 3600 h/year) [9].

The PV modules used in this

Developed models

In response to the crucial need for the PV module’s operating temperature information, techniques have been formed to predict such temperature in data-short areas. The back temperature on the plane of the module depends on the combined effects of astronomical and meteorological events. Generally, modules’ back temperature models are classified as either linear or multi-linear models, depending on the number of inputs required for making predictions. Among the models existing in the literature,

Performance evaluation of developed models

This section presents the evaluation of the developed models. Regression analyses were performed using the MATLAB® v16.0 software. The calculated regression coefficients (a to e) are shown in Table 5. The performance of the models is evaluated through comparing estimated and measured module back temperature values and the statistical test results of the 24 models obtained are presented in Table 6. The bold font in the table indicates the best values of the statistical indicators.

The evaluation

Conclusions

Anticipated operating temperatures of PV models are essential information for reliable design, planning, and integration of solar PV systems since it is one of the main influential parameters on the overall performance and productivity of the modules. In this study, new simple linear and multi-linear models for predicting the back temperature of PV modules from measured meteorological parameters and global irradiance have been established based on two-years datasets, collected from two stations

CRediT authorship contribution statement

Nadjem Bailek: Conceptualization, Data curation, Formal analysis, Supervision, Writing - original draft, Writing - review & editing. Kada Bouchouicha: Conceptualization, Data curation, Formal analysis, Supervision, Writing - original draft, Writing - review & editing. Muhammed A. Hassan: Formal analysis, Data curation, Writing - review & editing. Abdeldjalil Slimani: Formal analysis, Writing - original draft, Writing - review & editing. Basharat Jamil: Formal analysis, Writing - review &

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

The authors are highly thankful to the authority of the URERMSCDER for providing the support and facilities for the research work and giving access to the PV system data. We would like to thank all the members of electronic system team in UREMS.

References (46)

  • M.A. Hassan et al.

    Potential of four different machine-learning algorithms in modeling daily global solar radiation

    Renew. Energy

    (2017)
  • M. Said et al.

    Photovoltaics energy: improved modeling and analysis of the levelized cost of energy (LCOE) and grid parity–Egypt case study

    Sustain. Energy Technol. Assessments

    (2015)
  • M. Theristis et al.

    Energy yield in photovoltaic systems

  • M. Yorukoglu et al.

    A critical review on the estimation of daily global solar radiation from sunshine duration

    Energy Convers. Manag.

    (2006)
  • B. Jamil et al.

    Estimation of monthly mean diffuse solar radiation over India: performance of two variable models under different climatic zones

    Sustain. Energy Technol. Assessments

    (2018)
  • B. Jamil et al.

    Estimation of diffuse solar radiation in humid-subtropical climatic region of India: comparison of diffuse fraction and diffusion coefficient models

    Energy

    (2017)
  • M.A. Hassan et al.

    Exploring the potential of tree-based ensemble methods in solar radiation modeling

    Appl. Energy

    (2017)
  • J.D. Mondol et al.

    Comparison of measured and predicted long term performance of grid a connected photovoltaic system

    Energy Convers. Manag.

    (2007)
  • J.D. Mondol et al.

    Long-term validated simulation of a building integrated photovoltaic system

    Sol. Energy

    (2005)
  • J. Gale et al.

    Greenhouse Gas Control Technologies-6th International Conference

    (2003)
  • International Energy Agency IEA

    Energy Technology Perspectives: Catalysing Energy Technology Transformations

    (2017)
  • IEA Pvps

    Trends 2015 in Photovoltaic Applications

    (2015)
  • K. Komoto et al.
    (2015)
  • Cited by (24)

    • Techno-economic analysis for optimal configurations of PV systems with back reflectors

      2022, Energy Reports
      Citation Excerpt :

      The solar panel’s temperature is highly impacted by the incident solar irradiance, as well as the ambient air temperature. Specifically, the panel’s efficiency decreases by a rate of 0.4–0.6% per 1 °C increase in panel temperature (Bailek et al., 2020; Dubey et al., 2013). Consequently, one of the adopted methods to enhance the overall performance of the PV power system is to overcome the temperature increase in solar panels.

    View all citing articles on Scopus
    View full text