Search

Multijunction solar cells offer a path to very high conversion efficiency, exceeding 60% in theory. Under ideal conditions, efficiency increases monotonically with the number of junctions. In this study, we explore technical and economic mechanisms acting on tandem solar cells. We find that these mechanisms produce limitations that are the more pronounced the greater the number of junction is and, hence, limit the ideal number of junctions, as well as the corresponding efficiencies. Spectral variations induce current losses in series-connected tandem solar cells. For Denver, we find that these losses reduce achievable harvesting efficiencies to 51% for non-concentrated light, and that they restrict the ideal number of junctions to less than nine. Independently operated solar cells suffer from optical losses with similar consequences. Optical efficiencies of 99% restrict the ideal number of junctions to below ten, and reduce achievable efficiencies by more than 10%. Only architectures with a sequential cell illumination are more resilient to these losses. Restricting available materials reveals that a sufficiently low band gap for the bottom cell of 0.9 eV or below is expedient to realize high efficiencies. Economic considerations show that five junctions or less are economically ideal for most conceivable applications.

To identify abnormal photovoltaic (PV) modules in large‐scale PV plants economically, drone‐mounted infrared (IR) cameras and automated video processing algorithms are frequently used. While most related works focus on the detection of abnormal modules, little has been done to automatically localize those modules within the plant. In this work, we use incremental structure‐from‐motion to automatically obtain geocoordinates of all PV modules in a plant based on visual cues and the measured GPS trajectory of the drone. In addition, we extract multiple IR images of each PV module. Using our method, we successfully map 99.3% of the 35,084 modules in four large‐scale and one rooftop plant and extract over 2.2 million module images. As compared to our previous work, extraction misses 18 times less modules (one in 140 modules as compared to one in eight). Furthermore, two or three plant rows can be processed simultaneously, increasing module throughput and reducing flight duration by a factor of 2.1 and 3.7, respectively. Comparison with an accurate orthophoto of one of the large‐scale plants yields a root mean square error of the estimated module geocoordinates of 5.87 m and a relative error within each plant row of 0.22 m to 0.82 m. Finally, we use the module geocoordinates and extracted IR images to visualize distributions of module temperatures and anomaly predictions of a deep learning classifier on a map. While the temperature distribution helps to identify disconnected strings, we also find that its detection accuracy for module anomalies reaches, or even exceeds, that of a deep learning classifier for seven out of ten common anomaly types. The software is published at https://github.com/LukasBommes/PV-Hawk.

Energy losses of photovoltaic (PV) plants because of soiling are a problem in all regions, including Germany. Soft soiling, caused by a uniform dust film, is shading a PV module and is reducing yield depending on the thickness of the debris layer. Hard soiling, caused by agglomerations of dust or dirt, only covers parts of a PV module, and it causes local shading. A spot check of modules from different PV power plants in Germany revealed a power reduction because of soft- and hard soiling of up to 6%. Determination of soiling type and amount is a prerequisite for decisions about cleaning and for optimizing yield. In this article, we show that luminescence imaging can be used to detect and characterize soiling and quantify losses. For this purpose, we compare luminescence images before and after cleaning and we show that soiling becomes detectable and power losses quantifiable from difference images. We also show that the impact of hard soiling can be quantified from a single photoluminescence (PL) image. This technique may enable a fully automated quantification of power losses because of hard soiling in PV- modules and strings in the future.

Feasibility of the identification of backsheet (BS) composition of field-aged Si photovoltaic (PV) modules based on the analysis of characteristic patterns of UV-excited fluorescence (UVF) of ethylene vinyl acetate copolymer encapsulant is shown. The approach includes BS identification by vibrational spectroscopies and the formation of a database of BS-related characteristic UVF patterns. A multivariate principal component analysis of the UVF images of unknown samples combined with the UVF pattern database allows the BS type to be identified only from UVF images. The geometric place of particular samples in the principal component plots is indicative of the general degradation status of PV modules. The proposed method can be applied for high-throughput noncontact characterization of solar PV plants.

It is commonly assumed that photovoltaic (PV) modules have a 25-year lifetime. Warranties last this long, and techno-economic projections use this figure when calculating system performance. Yet, PV modules do not stop functioning after this time. Many modules still perform at above 80% of their original power rating after 25 years, and there is neither an economic nor an ecological reason to retire them. In this study, we explore the implications of transitioning from a mindset with finite module-life to a long-term perspective for PV-modules. We introduce an economic model to reflect such a long-term view. Using this model, we show that degradation gains in importance compared to module price. We calculate a cost-entitlement of up to 12ct/W for utility installation when improving annual degradation rates from 0.5% to 0.2%. We show that maintenance becomes more valuable – about 33% compared to a system with 30-year lifetime - and that time-constraints for maintenance disappear. We define a new lifetime metric - the minimum economically useful lifetime of a PV module as the time the module needs to be operated before replacement becomes economically preferable. We find that modules currently installed should be operated for at least 25 years and ideally for 50 years. Our metric links module life with degradation rate and, we calculate that reducing annual degradation rates from 0.5% to 0.2% increases the useful lifetime by a factor of 1.69. Such an increase would defer module end of life by decades and reduce resources required for maintenance by 40%.

Degradation of backsheets (BSs) of commercial silicon PV modules is currently recognized as a source of reduced module performance and module failure. Monitoring of the BS state in the field is possible by using non-destructive and highly informative near-infrared absorption (NIRA) spectroscopy. Application of NIRA for the analysis of multi-layer polyethylene terephtalate (PET) based BSs, which dominate the PV module market, is challenging due to a large variety of possible BS configurations that show only small differences in NIRA spectra. In the present work, a spectroscopic tool for the structural identification of PET-based BSs is introduced. The method is based on a principal component analysis of a database of 250 representative NIRA spectra of BSs of different types. It allows a BS with an unknown structure to be assigned to one of 12 different types based solely on its NIRA spectrum. The identification was successfully validated on a test collection of 45 selected BSs and shown to be feasible for the field deployment. Further automation of NIRA measurements and spectral analysis are expected to elevate the proposed tool to the level of a non-intrusive high-throughput field analysis of the BS composition and state in operating PV module grids.

In this study, we introduce a weather data independent self-referencing algorithm for performance analyses of PV power stations based on monitoring data. We introduce this method as an alternative to standard performance ratio analyses, in the case that no meteorological data are available. The self-referencing algorithm utilizes the string with maximum output as a reference. The output of other strings is set in relation to the corresponding value of the reference string, resulting in a quantity we call self-referencing ratio SR. We benchmark this ratio against the established performance ratio, which requires solar irradiance data for calculation. Based on a 9 MWp PV power station with 1,719 strings, we show that we are able to quantify, locate, and visualize the performance of each string in the PV power station. About 10% of the strings are underperforming; most of them are located in the bottom rows of the mounting racks. Shading from adjacent mounting racks is a possible explanation for this finding. Furthermore, IR-images and voltage data indicate a frequent occurrence of strings with a reduced number of modules. An underperformance of less than 2% is challenging to detect, due to the uncertainty range of the chosen dataset.

In field operation, solar cells are exposed to constantly changing operating conditions. These changing conditions have an impact on energy yield. Present-day yield predictions mostly use linear correction coefficients derived from lab experiments. These corrections neglect interactions between meteorological parameters like temperature and humidity. In this study, we reverse this approach by analyzing simulated solar cell performance under varying conditions worldwide. We use meteorological data measured between 2006 and 2015 to establish trends in the development of meteorological conditions and solar cell performance. From these two trends, we obtain linear correlation coefficients. The obtained implied temperature coefficient, on average, has a value of −0.52 ± 0.03%/K. This value is 15% higher than the tabulated temperature coefficient (−0.45%/K) used in the simulation, demonstrating the impact of coinciding meteorological factors. Light absorption due to elevated humidity levels is likely the strongest contributor to the deviation. One application of these findings is a projection of how today's crystalline silicon solar panels would perform due to rising temperature at the end of the 21st century. Using the established implied temperature coefficient, we project performance reductions of between 0.7% and 2.5%, depending on the warming scenario. The effect is reduced in higher efficient, upcoming photovoltaic technologies, providing further motivation to develop and improve these solar cells.

Increasing deployment of photovoltaic (PV) plants requires methods for automatic detection of faulty PV modules in modalities, such as infrared (IR) images. Recently, deep learning has become popular for this. However, related works typically sample train and test data from the same distribution ignoring the presence of domain shift between data of different PV plants. Instead, we frame fault detection as more realistic unsupervised domain adaptation problem where we train on labeled data of one source PV plant and make predictions on another target plant. We train a ResNet‐34 convolutional neural network with a supervised contrastive loss, on top of which we employ a k‐nearest neighbor classifier to detect anomalies. Our method achieves a satisfactory area under the receiver operating characteristic (AUROC) of 73.3% to 96.6% on nine combinations of four source and target datasets with 2.92 million IR images of which 8.5% are anomalous. It even outperforms a binary cross‐entropy classifier in some cases. With a fixed decision threshold, this results in 79.4% and 77.1% correctly classified normal and anomalous images, respectively. Most misclassified anomalies are of low severity, such as hot diodes and small hot spots. Our method is insensitive to hyperparameter settings, converges quickly, and reliably detects unknown types of anomalies making it well suited for practice. Possible uses are in automatic PV plant inspection systems or to streamline manual labeling of IR datasets by filtering out normal images. Furthermore, our work serves the community with a more realistic view on PV module fault detection using unsupervised domain adaptation to develop more performant methods with favorable generalization capabilities.

The significant capital expense of photovoltaics manufacturing has made it difficult for new cell and module technologies to enter the market. We present two technoeconomic models that analyze the sustainable growth of perovskite manufacturing for an R2R single-junction technology and a perovskite-silicon tandem module, focusing on the impacts of economies of scale and average selling price on profitability. We establish a cost range of $3.30/W to $0.53/W for flexible modules manufactured in factory sizes ranging from 0.3 MW/year to 1 GW/year. In addition, we model the cost to manufacture a tandem module consisting of a single-junction perovskite cell stacked in 4-terminal configuration onto a silicon cell and show how an existing manufacturer can grow at a faster rate by co-investing in tandems. Our analyses highlight potential routes to market for perovskite photovoltaics and the possibility to sustainably grow a photovoltaics manufacturing company even in markets with higher labor rates. We show how technoeconomic modeling of cleantech products versus scale can be an important tool in assisting a more rapid uptake of new energy technologies that often struggle to leave the lab. Our analyses highlight potential routes to market for perovskite photovoltaics and the possibility to sustainably grow a photovoltaics manufacturing company even in markets with higher labor rates. More generally, although technoeconomic modeling has proven to be a useful tool for assessing cleantech industries as they are and the long-term potential of new technologies once they reach scale - we encourage other cost modelers to quantify the impact of economies of scale during manufacturing growth to help in the search for viable and sustainable market on-ramps for their technologies. The significant capex of photovoltaics manufacturing has made it difficult for new cell and module technologies to enter the solar power market. We show how technoeconomic modeling of cleantech products versus scale can be an important tool in assisting the commercialization of new energy technologies that often struggle to leave the lab with our analyses focusing on potential routes to market for perovskite photovoltaics.

Richard E. Smalley, in 2003, defined the Terawatt (TW) Challenge as “Adapting our energy infrastructure to simultaneously address diminishing oil resources and rising levels of atmospheric CO 2.” Smalley, best known for the discovery of C 60, for which he received the 1996 Nobel Prize in Chemistry, continued to address the challenges of anthropomorphic and natural global energy flows until he passed away in 2005. Smalley challenged the world to transform the energy sector. He envisioned electricity transmitted by high-voltage direct current (DC) lines from massively deployed solar plants in sunny areas and remotely sited nuclear plants. He also envisioned using advanced batteries for local storage of energy. To meet the needs of ~10 people in a world with a dwindling oil supply, Smalley asserted that the world would need to transform its fossil-fuel-driven 14-TW (average power) energy used in 2003 to a largely renewable-energy-driven 30–60 TW (average power) in 2050. This would be possible only if solar-electricity costs could be drastically reduced. The challenges associated with this transition have been called the “Terawatt Challenge.” Fifteen years later, solar-module costs have been reduced by tenfold and annual deployment of solar photovoltaic (PV) modules has grown by a factor of 100,from ~1 gigawatt (GW) in 2004 to ~100 GW in 2018, with a total of 500 GW installed worldwide, producing 2% of the planet’s electricity. As global installed solar generating capacity approaches1 TW, we revisit Smalley’s TW challenge to identify what has changed and quantify the TW Challenge for a baseline scenario and for two scenarios designed as upper and lower bounds determined by the degree we implement electrification and storage. In this paper, we show that the energy choices we make today will dramatically affect the magnitude of future global energy requirements.

A techno-economic analysis of perovskite-silicon tandem solar modules is presented, outlining the most viable pathway for designing cost-effective, commercially viable tandems. We explore the cost-performance trade-off for silicon bottom cells in perovskite-silicon tandems, and evaluate the potential of using low-cost, lower-efficiency silicon bottom cells, on the basis of levelized cost of electricity (LCOE), compared to the higher-efficiency, higher-cost bottom cells that have been the primary focus of most perovskite-silicon tandem research efforts. We fabricate a cost-effective four-terminal silicon-perovskite tandem using a low-cost multicrystalline bottom cell and calculate the device LCOE. We then extend this analysis by modeling performance and LCOE of similar tandems instead using high-efficiency silicon bottom cells, enabling direct comparison of a low-cost and a high-efficiency tandem. Lastly parametric analyses are performed to more broadly examine the bottom-cell cost-performance trade-off. We show that low-cost silicon, even at the detriment of efficiency, is the more likely path to make perovskite-silicon tandems commercially viable and enable future reductions in LCOE, given both current and near-future silicon technology. We lay out a clear economic motivation for pursuing low-cost silicon bottom cells in perovskite-silicon tandems, showing that they can achieve a 15–20% relative LCOE reduction compared to the single-junction sub-cells. This is a 2–3 times greater relative LCOE reduction compared with using high-efficiency silicon. Furthermore, we show that the advantage of using low-cost silicon bottom cells is robust to and benefits from expected market trends, such as falling system costs and advanced, low-cost manufacturing. This work provides a clear pathway to cost-effective tandems, outlines the benefits for existing multicrystalline silicon manufacturers to investing in tandem development, and points out a clear mismatch between commercial viability and current research efforts.

Reliable and durable solar power plants require PV modules with high-grade polymer encapsulants and backsheets (BSs). For performance analyses of PV installations fast, reliable and non-destructive methods for determining composition and degradation state of polymer components need to be developed. Here, we show that the structure of some common polymer BSs can be determined in the field in real time by analyzing near-infrared transmission (NIRT) spectra collected under illumination with natural sunlight. The potential of this “Solar-NIRT” method was probed by field measurements on a multi-MW PV power plant where four major BS types were identified by multispectral cross-sectional Raman imaging. Additionally, degradation of a particular BS type was found to result in distinct changes in NIRT spectra allowing the degraded BSs to be classified as a separate type. Principal component analysis (PCA) applied to a collection of 62 Solar-NIRT spectra allowed to create a map of five clusters, each corresponding to a particular BS type. The feasibility of using the PCA cluster map for the identification of unknown samples was shown on a test set of 13 different BSs. The Solar-NIRT is relatively fast, non-invasive, selective, can be upgraded to a non-contact regime making it a promising tool for high-throughput characterization.

Thermography is a frequently used and appreciated method to detect underperforming Photovoltaic modules in solar power stations. With the review, we give insights on two aspects: (a) are the developed measurement strategies highly efficient (about 1 module s−1) to derive timely answers from the images for operators of multi-Mega Warr peak power stations, and (b) do Photovoltaic stakeholders get answers on the relevance of thermal anomalies for further decisions. Following these questions, the influence of measurement conditions, image and data collection, image evaluation as well as image assessment are discussed. From the literature it is clear that automated image acquisition with manned and unmanned aircrafts allow to capture more than 1 module s−1. This makes it possible to achieve almost identical measurement conditions for the modules; however, it is documented to what extent the increase in speed is achieved at the expense of image resolution. Many image processing tools based on machine learning (ML) have been developed and show the potential for analysis of infrared (IR) images and defect classification. There are different approaches to evaluating IR anomalies in terms of impact on performance, yield or degradation, of individual modules or modules in a string configuration. It is clear that the problem is very complex and multi-layered. On the one hand, information on the electrical interconnection is necessary, and on the other hand, there is a lack of sufficient and suitable data sets to adapt existing computer vision tools to Photovolatics. This is where we see the greatest need for action and further development to increase the expressiveness of IR images for PV stakeholder. We conclude with recommendations to improve the outcome of IR-images and encourage the generation of suitable public data sets of IR-footage for the development of ML tools.

Over the last years, a significant number of inverter shutdowns most likely due to degraded backsheets have been reported by operators, investors, and manufacturers. We investigated water ingress into different backsheets, and the resulting risk for inverter shutdowns. For studying pending insulation issues of inverters, we analyzed exemplarily a 5-MWp photovoltaic (PV) power station with 20,530 PV modules and 314 inverters. For backsheet identification on-site near-infrared absorption spectroscopic measurements of 518 PV modules from 20 inverters were carried out. In the lab, wet leakage tests provide kinetic data of insulation resistances for different backsheet materials. Historic ground impedance data logged by the inverter were evaluated to show the temporal evolution of inverters with different backsheets. We observe different kinetics and drops in leakage resistance for different backsheet types based on polyamide (PA) and air-side fluorinated coating (FC), and distinguish between two types of behavior. “PA-like” is characterized by moderate leakage resistance loss and slower performance reduction with regard to insulation in the field. “FC-like” is characterized by high susceptibility to water and a high leakage resistance loss rate for PV modules after 6–8 years of operation. These characteristics are indicated when measuring inverter GI in dependence of temperature and humidity. For PA-based BSs, the reduction in ground impedance (GI) is steady over 8 years. We observe incidences with GI below the critical value of 400 kΩ on 2.8% of all days. For FC-based BSs, we also observe steady decrease of average GI initially, yet after 6 years in the field, we see a sudden and rapid decline for 20% of those inverters. GI falling below the critical value is observed on up to 5% of all days, with the trend increasing. Therefore we, expect this problem to become even more severe for longer operating times.

Polymers are essential for long-term green energy production by photovoltaics (PV). We created a unique dataset that links electrical performance data of 28,030 PV modules, combined into 1295 strings and 423 inverters with the identity of the backsheet (BS) materials of each of the tested modules. This dataset gives an insight into BS-versatility and resulting impact on the insulation resistance of strings and inverters. The BSs were analyzed using near infrared absorption spectroscopy (NIRA) plus visual inspection in the field. The insulation resistance (Riso) of selected modules and strings was measured on-site. Furthermore, reported ground impedance (GI) values were evaluated with respect to an inverter-specific threshold below which the inverters do not start operation. We differentiated seven BS-types, the three most frequent being polyamide (PA), fluorinated coating (FC), and polyvinylidene fluoride (PVDF). Although modules with PA show often BS cracking, the insulation resistance was above the threshold. Similar, the GI of inverters with modules with PVDF-BSs was high. Modules with FC-BSs feature generally low Riso–values on module-, string-, and on inverter level. Evaluating historic data reveals an increase in frequency of GI-values below the threshold for inverters connected to modules with exclusively FC-BSs. While inverters with modules having PA- or PVDF-BSs function as expected, inverters with modules having FC-BSs require maintenance to keep power yield on an acceptable level. The dataset collected in the present work highlights that the choice of BS-material of PV-modules is essential for reliable operation of PV power stations.

PV module stability, in terms of reduced degradation rate and increased lifetime, provides an important lever for reducing the levelized cost of energy and life cycle environmental impacts of PV systems. Adapting an earlier value of efficiency methodology, the PV module cost per watt entitlement for a 30-year system lifetime is estimated to be ＄0.0125/W per 0.1% reduction in annual degradation rate, based on LCOE calculations. From an environmental perspective, the life cycle carbon footprint of a ground-mount PV system in a high solar resource location can be reduced by 0.3-1.0 g CO 2 -eq/kWh per 0.1% reduction in annual degradation rate. Increasing average PV module lifetime from 30 to 50 years will further increase these benefits, would reduce annual replacements by 40% and would result in net deferment of 62% of the projected module decommissioning through 2050 for PV modules installed in 2020. Increasing lifetime of state-of-the-art PV modules by 20 years to harvest the value of stability fully will require reducing PV module degradation rates to 0.2%/yr.

Non-concentrating tandem solar are on the verge of entering the PV market. One particular challenge for tandem solar cells is the interconnection scheme of the sub-cells in a module, and the implications on techno-economic factors. Here, we theoretically compare four possible tandem module architectures using the material combinations GaAs and perovskite on silicon to their single-junction counterparts. We model energy yield and performance ratios for operating conditions around the globe, and we explore the differences in cost structures. In general, we find the techno-economic performance of three-terminal and mechanically voltage-matched modules to be the most promising. Tandems are especially well suited for high-value markets and arid climates. In such conditions, the manufacturing cost premium for a record-level tandem is up to three times that of a silicon PV module.

Todays commercial and utility photovoltaic (PV) installations predominantly use monofacial PV-modules in a mounting structure with a fixed-tilt. In recent years, front and rear efficiencies of bifacial PV modules have improved and their cost has dropped, as has the cost for single-axis and dual-axis trackers. These advances make those two technologies interesting options to improve the yield of PV installations. In this work, we present a comparison of the calculated techno-economic performance of conventional installations and different combinations of bifacial and tracking configurations. We find that installations with bifacial modules and one-axis trackers provide the greatest techno-economic benefit over monofacial fixed-tilt installation by increasing yield by, on average 35%, and reducing LCOE on the vast majority of locations in the world by more than 10%. Dual axis trackers achieve even higher yields, yet the high cost for these trackers results in an overall higher LCOE. Sensitivity analyses reveal that our conclusions are robust in general, but the best configuration in a given location still requires analyzing local conditions in detail.

The individual causes for power loss of photovoltaic modules are investigated for quite some time. Recently, it has been shown that the power loss of a module is, for example, related to the fraction of inactive areas. While these areas can be easily identified from electroluminescense (EL) images, this is much harder for photoluminescence (PL) images. With this work, we close the gap between power regression from EL and PL images. We apply a deep convolutional neural network to predict the module power from PL images with a mean absolute error (MAE) of 4.4 ± 4.0 % or 11.7 ± 9.5 W P . Furthermore, we depict that regression maps computed from the embeddings of the trained network can be used to compute the localized power loss. Finally, we show that these regression maps can be used to identify inactive regions in PL images as well.

Increasing deployment of photovoltaics (PV) plants demands for cheap and fast inspection. A viable tool for this task is thermographic imaging by unmanned aerial vehicles (UAV). In this work, we develop a computer vision tool for the semi‐automatic extraction of PV modules from thermographic UAV videos. We use it to curate a dataset containing 4.3 million IR images of 107,842 PV modules from thermographic videos of seven different PV plants. To demonstrate its use for automated PV plant inspection, we train a ResNet‐50 to classify ten common module anomalies with more than 90% test accuracy. Experiments show that our tool generalizes well to different PV plants. It successfully extracts PV modules from 512 out of 561 plant rows. Failures are mostly due to an inappropriate UAV trajectory and erroneous module segmentation. Including all manual steps our tool enables inspection of 3.5 MWp to 9 MWp of PV installations per day, potentially scaling to multi‐gigawatt plants due to its parallel nature. While we present an effective method for automated PV plant inspection, we are also confident that our approach helps to meet the growing demand for large thermographic datasets for machine learning tasks, such as power prediction or unsupervised defect identification.

Summary Herein, we introduce the value of efficiency (VOE)—a metric defining the allowable cost for innovation by analyzing the co-dependence of levelized cost of electricity (LCOE) on manufacturing cost and efficiency. Historically, VOE values have dropped rapidly and will further, yet slower, decline in the future. Additionally, VOE values for residential and utility installations converge at different levels, indicating a stronger diversification of these market segments. Similarly, local differences in VOE suggest a regional market diversification. Advanced concepts like n-type monocrystalline PERC silicon solar cells could be economically preferable to multicrystalline silicon solar cells in some countries but may fare worse in others, including China and India. Extending the analysis to tandem photovoltaics, we confirm its potential to improve LCOEs for residential applications but foresee challenges for utility markets. Finally, we explore detrimental impacts of degradation and show that competing for lower LCOEs demands matching degradation rates of established technologies.

Summary Accelerating the experimental cycle for new materials development is vital for addressing the grand energy challenges of the 21st century. We fabricate and characterize 75 unique perovskite-inspired compositions within a 2-month period, with 87% exhibiting band gaps between 1.2 and 2.4 eV, which are of interest for energy-harvesting applications. We utilize a fully connected deep neural network to classify compounds based on experimental X-ray diffraction data into 0D, 2D, and 3D structures, more than 10 times faster than human analysis and with 90% accuracy. We validate our methods using lead-halide perovskites and extend the application to lead-free compositions. The wider synthesis window and faster cycle of learning enables the realization of a multi-site lead-free alloy series, Cs3(Bi1-xSbx)2(I1-xBrx)9. We reveal the non-linear band-gap behavior and transition in dimensionality upon simultaneous alloying on the B-site and X-site of Cs3Bi2I9 with Sb and Br.

Summary What role have batteries to play in the transition toward terawatt levels of photovoltaics? In this perspective, we attempt to answer this question by looking at technical, economic, and ecological features of PV-battery systems. We argue that the window of opportunity for batteries lies in the capacities of a few kWh/kWP, the exact amount depending on various factors including battery cost, degradation rate, location, load profile, diversification of renewable energy sources, and interconnections. Using a simple PV plus battery model, we illustrate that such storage capacities efficiently reduce fluctuations in electricity generation, enabling higher PV adoption rates at competitive costs, and with a carbon footprint that is at least five times lower than that of the current energy mixes. Using sensible capacities, batteries are a powerful companion for solar energy, yet technical, economic, and policy innovations are needed to expand adoption. We see longer battery lifetimes and low capacity degradation rates are the most impactful technological parameters. Economic efforts should aim to reduce balance of plant costs and create better market opportunities for stationary storage, whereas policies should provide a strong regulatory framework to facilitate multipurpose usage and sector coupling.

The daily Sun supplies the continents of the Earth with four times more energy than humanity consumes in a year. This enormous potential of solar energy to generate clean energy is therefore driving great efforts to replace conventional and unsustainable fossil fuel consumption that damages our climate and our environment. Solar photovoltaic (PV) is emerging as the fastest growing renewable energy technology in the world, yet its share to the electricity production currently is less than 3%. While coal and gas remain key to electricity production, the climate crisis demands a fast transition to a carbon-neutral energy system. In the year 2019, the PV industry produced solar panels with a capacity of about 115 GW. To reach multi-TW levels by 2030, PV production capacity needs to scale at high rates, and continued innovation is necessary. One way to increase production is to reduce silicon wafer thickness, making more cells from the amount of silicon produced. Thinning the cell will not only allow making more of them in less time but will also reduce the cost for a module, and for the electricity it generates. A thinner absorber reduces the absorption of sunlight. To counter this loss, optical concepts are needed. In this chapter, we review the most promising approaches in photonics that have made physically thinner but optically thicker solar cells possible. By enabling thinner solar cells, photonic solutions offer the possibility of efficiency enhancements at lower costs. We will explain their applications, discuss their challenges and estimate their impact on energy yield. Finally, we outline how non-silicon thin-film PV is opening up a plethora of consumer-oriented applications. Since the higher flexibility/bendability of thinner cells broadens their applicability range, thin-film PV will empower our transition to a 100% green society in many aspects of our lives.

Automated inspection plays an important role in monitoring large‐scale photovoltaic power plants. Commonly, electroluminescense measurements are used to identify various types of defects on solar modules, but have not been used to determine the power of a module. However, knowledge of the power at maximum power point is important as well, since drops in the power of a single module can affect the performance of an entire string. By now, this is commonly determined by measurements that require to discontact or even dismount the module, rendering a regular inspection of individual modules infeasible. In this work, we bridge the gap between electroluminescense measurements and the power determination of a module. We compile a large dataset of 719 electroluminescense measurements of modules at various stages of degradation, especially cell cracks and fractures, and the corresponding power at maximum power point. Here, we focus on inactive regions and cracks as the predominant type of defect. We set up a baseline regression model to predict the power from electroluminescense measurements with a mean absolute error (MAE) of 9.0 ± 8.4WP (4.0 ± 3.7%). Then, we show that deep learning can be used to train a model that performs significantly better (7.3 ± 6.5WP or 3.2 ± 2.7%) and propose a variant of class activation maps to obtain the per cell power loss, as predicted by the model. With this work, we aim to open a new research topic. Therefore, we publicly release the dataset, the code, and trained models to empower other researchers to compare against our results. Finally, we present a thorough evaluation of certain boundary conditions like the dataset size and an automated preprocessing pipeline for on‐site measurements showing multiple modules at once.

Spectral differences affect solar cell performance, an effect that is especially visible when comparing different solar cell technologies. To reproduce the impact of varying spectra on solar cell performance in the lab, a unique classification of spectra is needed, which is currently missing in literature. The most commonly used classification, average photon energy (APE), is not unique, and a single APE value may represent various spectra depending on location. In this work, we propose a classification method based on an iterative use of the k-means clustering algorithm. We call this method RISE (Representative Identification of Spectra and the Environment). We define a set of 18 spectra using RISE and reproduce the spectral impact on energy yield for various solar cell technologies and locations. We explore effects on yield for commercially available solar cell technologies (Si, CdTe) in four locations: Singapore (fully humid equatorial climate), Colorado (cold arid), Brazil (warm, humid, subtropical), and Denmark (fully humid warm temperature).We then reduce our findings to practice by implementing the spectrum set into an LED current-voltage (IV)tester. We verify our performance predictions using our set of representative spectra to reproduce energyyield differences between Si solar cells and CdTe solar cells with an average error of less than 1.5 ± 0.5% ascompared to over 5% when using standard testing conditions.

Visual inspection of solar modules is an important monitoring facility in photovoltaic power plants. Since a single measurement of fast CMOS sensors is limited in spatial resolution and often not sufficient to reliably detect small defects, we apply multiframe superresolution (MFSR) to a sequence of low-resolution measurements. In addition, the rectification and removal of lens distortion simplifies subsequent analysis. Therefore, we propose to fuse this preprocessing with standard MFSR algorithms. This is advantageous, because we omit a separate processing step, the motion estimation becomes more stable and the spacing of high-resolution pixels on the rectified module image becomes uniform w.r.t. the module plane, regardless of perspective distortion. We present a comprehensive user study showing that MFSR is beneficial for defect recognition by human experts and that the proposed method performs better than the state-of-the-art. Furthermore, we apply automated crack segmentation and show that the proposed method performs $3\times$ better than bicubic upsampling and $2\times$ better than the state-of-the-art for automated inspection.

The field of energy forecasting has attracted many researchers from different fields (e.g., meteorology, data sciences, mechanical or electrical engineering) over the last decade. Solar forecasting is a fast-growing subdomain of energy forecasting. Despite several previous attempts, the methods and measures used for verification of deterministic (also known as single-valued or point) solar forecasts are still far from being standardized, making forecast analysis and comparison difficult. To analyze and compare solar forecasts, the well-established Murphy–Winkler framework for distribution-oriented forecast verification is recommended as a standard practice. This framework examines aspects of forecast quality, such as reliability, resolution, association, or discrimination, and analyzes the joint distribution of forecasts and observations, which contains all time-independent information relevant to verification. To verify forecasts, one can use any graphical display or mathematical/statistical measure to provide insights and summarize the aspects of forecast quality. The majority of graphical methods and accuracy measures known to solar forecasters are specific methods under this general framework. Additionally, measuring the overall skillfulness of forecasters is also of general interest. The use of the root mean square error (RMSE) skill score based on the optimal convex combination of climatology and persistence methods is highly recommended. By standardizing the accuracy measure and reference forecasting method, the RMSE skill score allows—with appropriate caveats—comparison of forecasts made using different models, across different locations and time periods. Refereed/Peer-reviewed

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Solar spectra change depending on location, weather, and time of day around the world. These spectral variations affect the performance of photovoltaic (PV) cells and modules and are not captured by standard testing procedures. We introduce a method to classify spectra and use this classification to develop a testing procedure to reproduce spectral conditions of locations within various climate zones. With LED solar simulators becoming commercially available and representative sets of outdoor spectra found, the gap between real world outdoor testing and indoor testing is closing. With LED-based testing of CdTe, Si, and GaAs PV cells, we show that the effects of spectral shifts on short circuit current are captured, demonstrating the potential use of this method for more accurate testing of solar modules indoors.

Energy yield is a key metric for evaluating the performance of photovoltaic systems. It describes the total amount of energy generated by a photovoltaic (PV) installation over a given period, typically a year, and depends on physical properties of the solar cell like efficiency, band gap and temperature coefficient, as well as the operating conditions in a given location. Because the response of a solar cell to these conditions varies, two photovoltaic technologies may have a different energy yield, even if their lab efficiency is identical. Predicting energy yield accurately is important to system operators and installers to estimate the technical and economic performance of a PV installation. In this paper, we summarize our findings about satellite based energy yield predictions of solar cells with various technologies.

Summary Silicon-based multi-junction (tandem) technology is one potential route to the next breakthrough for terrestrial photovoltaic conversion. Significant progress has been made in tandem solar cells. To move forward, development of tandem module technology is essential. Here, we theoretically compare five possible tandem module architectures with multiple material combinations by modeling their outdoor performance ratios around the globe. This framework aids in evaluating future module designs and technology pathways by relating lab-based efficiency to field performance and even project financing. We find that tandem performance ratios depend notably on climate conditions. In general, three-terminal and mechanically voltage-matched modules show exceptionally good outdoor performance. Furthermore, we investigate the implications of technical performance on manufacturing cost globally and find that tandems are promising in high-value markets in arid climates, where a three to five times increase in the current silicon module manufacturing cost is allowed under an optimistic scenario.

This study provides experimental validation of a previously published optimal design theory for photovoltaic (PV)-powered electrodialysis reversal (EDR) desalination systems. The prior work describes the co-optimization of PV and EDR subsystems, and flexible operation to accommodate daily and annual solar irradiance variability, significantly reducing water cost. This study presents the fabrication of a PV-EDR pilot system designed using the co-optimization theory and field testing results from the rural village of Chelluru, India. Testing in the field enabled observation and evaluation of real-world factors on system performance, resulting in updates to the previous theory to include unaccounted factors that affect costs, including: filling and draining of water tanks, salt and water accumulation in tanks from prior batches, unexpected energy losses due to locally purchased converters, and scaling in the ED stack. Therefore, water cost in the PV-EDR pilot system was updated from previous estimates based on field performance. The estimated capital cost and lifetime cost of the Chelluru system are 34% and 45% lower, respectively, than the corresponding costs if the PV-EDR system was designed using conventional design practice. The theory and experimental insights presented in this paper will enable desalination engineers to better design and optimize PV-EDR systems.

Process optimization of photovoltaic devices is a time-intensive, trial-and-error endeavor, which lacks full transparency of the underlying physics and relies on user-imposed constraints that may or may not lead to a global optimum. Herein, we demonstrate that embedding physics domain knowledge into a Bayesian network enables an optimization approach for gallium arsenide (GaAs) solar cells that identifies the root cause(s) of underperformance with layer-by-layer resolution and reveals alternative optimal process windows beyond traditional black-box optimization. Our Bayesian network approach links a key GaAs process variable (growth temperature) to material descriptors (bulk and interface properties, e.g., bulk lifetime, doping, and surface recombination) and device performance parameters (e.g., cell efficiency). For this purpose, we combine a Bayesian inference framework with a neural network surrogate device-physics model that is 100× faster than numerical solvers. With the trained surrogate model and only a small number of experimental samples, our approach reduces significantly the time-consuming intervention and characterization required by the experimentalist. As a demonstration of our method, in only five metal organic chemical vapor depositions, we identify a superior growth temperature profile for the window, bulk, and back surface field layer of a GaAs solar cell, without any secondary measurements, and demonstrate a 6.5% relative AM1.5G efficiency improvement above traditional grid search methods.

Summary This work presents a worldwide analysis on the yield potential and cost effectiveness of photovoltaic farms composed of monofacial fixed-tilt and single/dual (1T/2T) tracker installations, as well as their bifacial counterparts. Our approach starts by estimating the irradiance reaching both the front and rear surfaces of the modules for the different system designs (validated based on data from real photovoltaic systems and results from the literature) to estimate their energy production. Subsequently, the overall system cost during their 25-year lifetime is factored in, and the levelized cost of electricity (LCOE) is obtained. The results reveal that bifacial-1T installations increase energy yield by 35% and reach the lowest LCOE for the majority of the world (93.1% of the land area). Although dual-axis trackers achieve the highest energy generation, their costs are still too high and are therefore not as cost effective. Sensitivity analyses are also provided to show the general robustness of our findings.

In this paper, we discuss the value of efficiency (VOE)-a metric to assess the maximum allowable cost for innovation in photovoltaic technology. We derive VOE by analyzing the co-dependence of levelized cost of electricity (LCOE) on manufacturing cost and efficiency and use the tool to explore different scenarios and technologies. Exploring historical trends indicates that a stronger market diversification between utility and residential applications could develop. Regional trends indicate that another diversification could also happen across different countries. Finally, we also explore detrimental economic effects of degradation and show that competing for lower LCOEs demands achieving similar degradation rates as established technologies do.