Floating photovoltaics (FPVs) are gaining traction as a land-saving alternative to ground-mounted photovoltaics (GPV). A commonly cited advantage of FPVs is their potential for lower operating temperatures due to the cooling effect of water. However, existing literature shows that the thermal performance of FPV systems may not consistently exceed that of GPV systems, as it is influenced by technology and location. Consequently, studying the thermal properties of a range of FPV systems is crucial to optimize power output and enable accurate energy yield modeling for new sites. This work investigates the thermal properties and calculated heat loss coefficients, or U-values, associated with the Faiman model for an FPV system using Ciel & Terre's Hydrelio Air floats, located in a pond in South Africa. A dependence of U-values on wind direction was observed, with improved cooling when the wind approaches from the rear side of the system. The estimated U-value components were U-0 = 21.6 W/m(2)K and U-1 = 3.60 Ws/m(3)K for wind from the front and U-0 = 19.4 W/m(2)K and U-1 = 7.10 Ws/m(3)K for wind from the rear side. The impact of the observed cooling variation due to wind direction on system performance was also evaluated, revealing a 1.7% increase in median performance ratio when the wind originates from the rear side.

Floating photovoltaic (FPV) modules may face a risk of increased moisture ingress due to their deployment on water surfaces. One way to mitigate this is by using impermeable front- and backsheets, with an edge sealant around the module perimeter. While a suitable sealant should have low bulk permeability, proper sealant application to avoid higher ingress channels at interfaces is crucial. Here, we report on the use of a gravimetric method as a simple way of evaluating moisture ingress through an edge sealant and of identifying application-related issues that lead to increased moisture ingress. The method uses multiple samples that closely mimic the sealant's intended application as part of an FPV design developed by the company Sunlit Sea. Supported by steady-state water vapor transmission rate measurements and finite-element modeling, the method is shown to be capable of determining the order of magnitude of the permeability of two different candidate sealant materials. Moreover, the method detected several application-related sealant failures that were not discernible through visual inspection. Finally, it uncovered potential issues of debonding of one of the sealants in immersion, highlighting a relevant yet understudied stressor for FPV modules.

Floating photovoltaic (FPV) systems are emerging as a promising solution to the scarcity of suitable land for ground-mounted solar PV (GPV) installations. By the end of 2022, global FPV capacity reached 5.7 GWp following a remarkable compound annual growth rate of approximately 87.5% from 2015 to 2022. This growth introduces a significant new frontier for operation and maintenance (O&M) practices in the solar industry. As the industry matures and more FPV assets come under operation, the need for innovative, efficient, and environmentally sensitive O&M strategies becomes imperative. This review presents the existing information on the O&M of FPV systems, highlighting the unique challenges and opportunities that set FPV systems apart from conventional GPV installations. Through an examination of recent advancements, best practices, and areas requiring further research, this study aims to provide valuable insights for optimizing the performance and sustainability of FPV systems.

The interest in floating photovoltaics (FPV) as a land-sparing alternative has been rapidly growing over the last few years. So far, most of the installed FPV systems are inland on relatively calm waters. However, with the expanding market, FPV is moving to nearshore and offshore locations. There, larger waves can be expected resulting in a need for evaluating the impact of wave motion on the performance of FPV systems. Wave-induced movements of the PV panels can lead to varying irradiance levels, also within the string of panels, causing wave-induced loss (WIL). In this work, we have developed a model to simulate WIL for FPV systems. The model consists of three main modelling steps, wave-structure-interaction modelling, irradiance modelling and electrical modelling, where the second step has been experimentally verified. In the model, the in-house software 3DFloat is used to simulate the movement of the PV panels, and existing python libraries are used for the irradiance and electrical modelling. With this method, WIL has been simulated for different scenarios, to study the influence of different locations, sea states, times of year, string lengths, tilt angles and wave directions. The simulated WIL varied greatly with the chosen conditions. As an example, the WIL ranged from 3.3% for a significant wave height of 0.25 m to 6.7% for a significant wave height of 1 m.

Floating photovoltaics (FPV) is rapidly emerging as a promising alternative to ground-mounted PV (GPV) where available land area is scarce or expensive. Improved cooling has often been reported as a benefit of FPV, as cell temperature is an important parameter for the performance of a PV system. However, more recent literature shows that the cooling effect depends strongly on FPV technology and that it is not always superior to that of open rack GPV systems. There is still a need for more information on how to estimate cell temperatures for FPV systems, and how to consider the influence of various environmental factors such as wind speed and direction. Operating cell temperature may be estimated with the PVsyst model, where heat loss coefficients (U-values) denote the heat transfer capabilities of the PV system. In this work, cell temperatures and U-values for a small footprint FPV system with east-west orientation and a 15° tilt located in Sri Lanka are studied using both module temperature measurements and computational fluid dynamics (CFD) modelling. CFD modelling allows for investigating the influence of both wind speed and direction on cell temperatures, as well as to look at the distribution of cell temperatures within the system under different wind conditions. Calculations based on measurements give Uc = 22.6 W/m2K and Uv = 4.9 Ws/m3K and correlate well with CFD calculations. We also show that wind direction, system configuration and sensor placement influence the estimated U-values, complicating the use of tabulated values for any given technology.

Moisture ingress into photovoltaic (PV) modules is one of the main drivers behind module midlife- and wear-out-failures, particularly when modules are installed in locations with high humidity stress. To minimize moisture ingress, impermeable front- and backsheets in combination with an edge sealant around the module perimeter can be used. Besides low water vapor transmission rates through the sealant material, mechanical durability of the sealant is of utmost importance. In this work, we assess the durability of a double edge sealant design used in a floating PV (FPV) concept in which the float and PV module are integrated. Strength of attachment testing, combined with measurements of failure type, are performed on samples taken from an FPV prototype after outdoor exposure and are compared to measurements on lab samples that are unexposed or exposed to indoor accelerated stress. We observe a significant decrease of lap shear strength for the double edge sealant after field exposure, coupled with more adhesive failure. Correlation to accelerated stress test results indicates that the observed adhesion losses can partly be attributed to degradation driven by ultraviolet light. By reporting on observations of FPV field degradation and exploring how indoor accelerated stress testing can be used to understand the origins of the observed degradation, this work constitutes an important early contribution to the field of FPV reliability.

The power output of a photovoltaic system is dependent on the operating temperature of the solar cells. For floating PV (FPV), increased wind speeds can result in increased yield due to lowered operating temperatures, which has long been stated as a key advantage for FPV. So far, this effect has not been included in commercial software packages for yield estimation. Typically, only standard settings are provided, taking into account the mounting type (PVsyst) or mounting and module type (Sandia). This may result in an underestimation of the yield, and consequently, the estimated Levelized Cost of Electricity (LCOE) of the FPV project. In this study, a linkage between recorded module temperatures from FPV systems located in The Netherlands and Sri Lanka and the prevalent models employed within PVsyst and Sandia software for estimating module temperatures are established. Our findings reveal that the models within PVsyst and Sandia tend to overestimate module temperatures by 2.4% and 3%, respectively, for each 1 m/s increment in wind speed. We present two methods for determining the single heat loss coefficient, or U-value, tailored to specific sites accounting for local wind conditions. The first method computes the U-value based on the average monthly wind speed, whereas the second employs the irradiance-weighted average monthly wind speed. The latter method can be advantageous for locations characterized by significant fluctuations in wind speeds between night and day. Through a statistical residual analysis comparing measured and modeled module temperatures, we demonstrate that our proposed methods offer a more accurate representation of module temperature compared to the PVsyst and Sandia models when default settings are used. When we subsequently compute the specific yield using both measured and modeled temperatures, we observe that the approach using irradiance-weighted average wind speed shows a higher yield of up to 2% compared to the traditional methods.

Accurate photovoltaics (PV) energy yield assessments for cold climates necessitate understanding and estimation of snow loss. Estimation of snow loss for a specific system requires a snow loss model. Multiple models to estimate snow loss are suggested in the literature, but extensive validation is lacking. In this work, we describe the effect of snow on PV systems by analyzing signatures in monitoring data and we evaluate the accuracy of a modified adaption of the Marion snow loss model. Eight different systems with a total installed capacity of 1.6 MW p , installed on both tilted and flat roofs, are analyzed. In the modified model we use different snow clearing rate coefficients for thin and thick snow covers to model the natural snow clearing process. The snow depth dependent coefficients yield lower error in the total modeled snow loss and capture climatic variations between locations more accurately compared to the standard constant coefficient. For most of the systems the total absolute snow loss is modeled with an error of less than 11%, on average 23% points lower than with the default implementation of the Marion model. Some of the systems have larger modeling errors, which can be related to effects not taken into account in the model, such as the effect of building heat leakage and shading on snow clearing.

Floating photovoltaics (FPV) is a rapidly emerging technology that provides an alternative to ground-mounted PV (GPV), particularly where land is scarce or expensive. Despite an impressive technological development and growth in installed capacity in recent years, studies on the performance and reliability of FPV are scarce. This work provides insight with respect to the performance, reliability, and operational characteristics of a new FPV technology with the aim to identify innovation opportunities, reduce risks, develop improved solutions, and improve bankability of FPV. We have analysed production and weather data from one year of operation for an open FPV system with a small water footprint located on a water body in Kilinochchi, Sri Lanka. The technology is developed by the company Current Solar. Using established filtering routines and algorithms from pvlib, the yield and performance ratio is calculated and compared to a GPV system installed on the shore of the lake. We find that the technology gives a stable overall performance over the one-year period, and that the period of amphibious operation did not impact the continued performance of the system. Calculations of the U-value of the system, based on the production and weather data, gives a median U-value of 33 W/m2K, slightly higher than the default PVsyst value of 29 W/m2K for freestanding GPV systems. The calculated U-values are used in an energy yield analysis in PVsyst to estimate the energy production of the FPV technology and benchmark it against measured data.

The Norwegian agricultural sector has an ambition to reduce green-house gas emissions and a significant share of the reduction is expected from switching to a fossil-free machine park. Electrification of tractors might be an important step towards sustainable farming. Field operations are energy intensive, and electrification is expected to increase peak electricity demand. Herein, we analyze how distributed energy resources could be designed to optimize cost of electricity on farms. Analyzes were performed for three different farms (Dairy, Grain, and Pork/Grain) where energy demand for both indoor and field operations were included. Data with hourly time resolution was used and simulations were performed with the Homer software to optimize the net present cost of a grid-connected PV-battery system supplying electricity to indoor operations, charging of battery tractors, and operation of electrolyzers for filling of hydrogen fuel cell tractors. Results showed that electrification of tractors altered the total demand for grid-imported electricity differently at the three farms when including distributed energy sources. Grid-imported electricity was 24% higher for Dairy, 122% higher for Grain, whereas Pork/Grain consumed 13% less grid-imported electricity. This implies that the local PV-production can in some cases cost-efficiently cover the increased energy demand associated with electrification of tractors.

Floating photovoltaics (FPV) systems are an emerging and increasingly competitive application of solar PV. Although lack of land is the main reason for assessing the technology against land-based PV, favourable cooling conditions, justified by the fact that air above a water body is typically colder and windier than on land, is often also used as an argument. Few have quantified this effect, and the aim of this work has been to use Computational Fluid Dynamics (CFD) to derive U-values for the most used FPV technology today. Our results show that the presence of a float with large water footprint promotes a significant non-uniform temperature field in the wafer layer. The study confirms that the cell temperature rises by decreased wind and by increased solar irradiation and air temperature. We found that the water temperature does not significantly affect the cell temperature. Radiative heat transfer between the back of the module and water was estimated to reduce the cell temperature typically less than 1 C. Because the technologies are not identical but still represent systems with large water footprint, U-values from the study were considered to compare well to reported values derived from FPV field data in Singapore and the Netherlands.

For floating PV (FPV), the operating temperature of the PV modules has been a major source of uncertainty. As the operating temperature of PV modules affects their efficiency, knowledge of this parameter is critical in order to perform accurate energy yield assessment (EYA). This uncertainty is reflected in the scientific literature but has also hampered the bankability and realization of commercial FPV projects. Our work proposes a model that computes both the efficiency of heat loss to the environment for a given FPV technology and the operational cell temperature, needed to compute the module efficiency. The suggested model is applicable to different FPV technologies. We show that PV modules that are not in direct thermal contact with water have similar heat loss behavior as land‐based PV. We thereafter investigate a specific technology in which the modules are mounted on a membrane resting directly on the water body. The model is validated against actual production and weather data from deployed FPV systems. Our results show that the water temperature impacts both FPV technologies, however to a smaller degree for the air‐cooled system, where the wind is of greater influence. When in direct thermal contact with water, the water body provides superior cooling, and the resulting U‐value of about 86.5 W/m 2 K is significantly larger than typical values reported for land‐based modules.

Enhanced performance of floating PV due to water cooling is widely claimed, but poorly quantified and documented in the scientific literature. In this work, we assess the effect of water cooling for a specific technology developed by Ocean Sun AS, consisting of a floating membrane with horizontally mounted PV modules allowing for thermal contact between the modules and the water. The impact of thermal contact with water on energy yield is quantified using production data from a well-instrumented 6.48 kW installation at Skaft?, Norway. In addition, we apply a thermal model that incorporates the effect of heat transport from the module to the water to estimate the module temperature. By comparing a module string in thermal contact with water with a module string with an air gap between the water and the modules, we find that the water-cooled string had on average 5?6% higher yield compared to the air-cooled string. Also, we find that the system in thermal contact with water has a U-value of approximately 70?80 W/m2K, and that it is necessary to consider the water temperature for a more accurate calculation of the module temperature.

To achieve accurate PV energy yield assessments for cold climate locations where snowfalls can lead to significant power losses in the wintertime, improvement and validation of snow loss modeling is necessary. In this paper we introduce a snow depth dependent clearing rate coefficient in the Marion snow loss model to account for the effect of different snow conditions on the natural snow clearing process. To evaluate the model, seven roof mounted PV plants with two different system configurations are analyzed. For both system configurations the snow depth dependent clearing rate coefficient yields lower error in the total modeled snow loss and capture climatic variations between locations better compared to the standard constant coefficient.

As cost reductions have made photovoltaics (PV) a favorable choice also in colder climates, the number of PV plants in regions with snowfalls is increasing rapidly. Snow coverage on the PV modules will lead to significant power losses, which must be estimated and accounted for in order to achieve accurate energy yield assessment and production forecasts. Additionally, detection and separation of snow loss from other system losses is necessary to establish robust operation and maintenance (O&M) routines and performance evaluations. Snow loss models have been suggested in the literature, but developing general models is challenging, and validation of the models are lacking. Characterization and detection of snow events in PV data has not been widely discussed. In this paper, we identify the signatures in PV data caused by different types of snow cover, evaluate and improve snow loss modeling, and develop snow detection. The analysis is based on five years of data from a commercial PV system in Norway. In an evaluation of four snow loss models, the Marion model yields the best results. We find that system design and snow depth influence the natural snow clearing, and by expanding the Marion model to take this into account, the error in the modeled absolute loss for the tested system is reduced from 23% to 3%. Based on the improved modeling and the identified data signatures we detect 97% of the snow losses in the dataset. Endogenous snow detection constitutes a cost-effective improvement to current monitoring systems.

Reliable monitoring of PV systems is essential to establish efficient maintenance routines that minimize the levelized cost of electricity. The existing solutions for affordable monitoring of commercial PV systems are however inadequate for climates where snow and highly varying weather result in unstable performance metrics. The aim of this work is to decrease this instability to enable more reliable monitoring solutions for PV systems installed in these climates. Different performance metrics have been tested on Norwegian installations with a total installed capacity of 3.3 MW: (i) comparison of specific yield, (ii) temperature corrected performance ratio, and (iii) power performance index based on both physical modelling and machine learning. The most influential effects leading to instability are identified as snow, low light, curtailment, and systematic irradiance differences over the system. The standard deviation of all the performance metrics is reduced when filters targeting these four effects are applied. Compared to general low irradiance or clear sky filtering, a greater reduction in the variation of the metrics is achieved, and more data remains in the useful dataset. The most suitable performance metrics are comparison of specific yield and performance index based on machine learning modelling. The analysis highlights two paths to accomplish increased reliability of PV monitoring systems without increased hardware costs. First, better reliability can be achieved by selecting a suitable performance metric. Second, the variability of the performance metric can be reduced by utilizing filters that specifically target the origin of the variability instead of using standard literature thresholds.

Methods for quick and accurate detection and diagnosis of defects in PV systems are increasingly important as the global photovoltaic (PV) capacity continues to grow at a rapid pace. Two of the most used methods for defect detection involve aerial infrared thermography and data analysis of production data. In this work, we combine the two methods to analyze two utility scale PV plants, providing new understanding about the two methods. We report on the percentage and distribution of thermal anomalies of different categories and quantify their relationship with performance on string level. We find that the most important parameter for determining production losses on string level is the number of module substrings containing thermal anomalies. Due to the large variability of the effect of different thermal signatures, as well as uncertainties in the estimate of the string performance, we find no clear correlation between performance and thermal signature category or temperature. However, a whole hot cell is the thermal signature that on average has the smallest impact on the power output on string level. Finally, in our data, the performance of a string of 20 modules with 3 bypass diodes is on average reduced by 1.16 +/- 0.12% per module substring containing thermal anomalies.