SEARCH · Search NASA
Results for “Solar Arrays”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
An in-depth field validation of “DUSST”: A novel low-maintenance soiling measurement device
This study presents indoor and field validation results for two versions of the “DUSST” optical soiling sensor, intended to be a low-cost and low-maintenance device for measuring photovoltaic soiling losses. Indoor testing covers irradiance calibration and temperature dependencies, which are necessary to achieve high accuracy, low uncertainty field measurements. Field testing includes an array of different environments including Saudi Arabia, California, Utah, and Colorado. DUSST versions include a configuration with a 530-nm light emitting diode (LED) (discussed in previous work) and a unit with seven white LEDs and a polycarbonate collimating optic. The new design increases light intensity fivefold and demonstrates a single linear calibration coefficient is effective to measure soiling losses as high as 75%. Field data from Utah and California demonstrate that daily soiling loss measurements and soiling rate calculations closely match both reference cell and full-size module measurements of soiling losses and soiling rates. Corrective methods employed on the Utah DUSST sensor suggest that it is possible to achieve measurement errors as low as ±0.1% at two standard deviations. Field data from both Colorado and Saudi Arabia demonstrate that LED lens soiling can occur and that further design optimizations are needed. The lesson learned from all the field deployment locations suggests directions for future design improvements.
Initial Characterization of Active Transitioning Centaur, P/2019 LD{sub 2} (ATLAS), Using Hubble, Spitzer, ZTF, Keck, Apache Point Observatory, and GROWTH Visible and Infrared Imaging and Spectroscopy
We present visible and mid-infrared imagery and photometry of temporary Jovian co-orbital comet P/2019 LD{sub 2} taken with Hubble Space Telescope/Wide Field Camera 3 (HST/WFC3), Spitzer Space Telescope/Infrared Array Camera (Spitzer/IRAC), and the GROWTH telescope network, visible spectroscopy from Keck/Low-Resolution Imaging Spectrometer (LRIS), and archival Zwicky Transient Facility observations taken between 2019 April and 2020 August. Our observations indicate that the nucleus of LD{sub 2} has a radius between 0.2 and 1.8 km assuming a 0.08 albedo and a coma dominated by ∼100 μm-scale dust ejected at ∼1 m s{sup −1} speeds with a ∼1′ jet pointing in the southwest direction. LD{sub 2} experienced a total dust mass loss of ∼10{sup 8} kg at a loss rate of ∼6 kg s{sup −1} with Afρ/cross section varying between ∼85 cm/125 km{sup 2} and ∼200 cm/310 km{sup 2} from 2019 April 9 to 2019 November 8. If the increase in Afρ/cross section remained constant, it implies LD{sub 2}'s activity began ∼2018 November when within 4.8 au of the Sun, implying the onset of H{sub 2}O sublimation. We measure CO/CO{sub 2} gas production of ≲10{sup 27} mol s{sup −1}/≲10{sup 26} mol s{sup −1} from our 4.5 μm Spitzer observations; g–r = 0.59 ± 0.03, r–i = 0.18 ± 0.05, and i–z = 0.01 ± 0.07 from GROWTH observations; and H{sub 2}O gas production of ≲80 kg s{sup −1} scaling from our estimated C {sub 2} production of Q{sub C{sub 2}}≲7.5×10{sup 24} mol s{sup −1} from Keck/LRIS spectroscopy. We determine that the long-term orbit of LD{sub 2} is similar to Jupiter-family comets having close encounters with Jupiter within ∼0.5 Hill radius in the last ∼3 y and within 0.8 Hill radius in ∼9 y. Additionally, 78.8% of our orbital clones are ejected from the solar system within 1 × 10{sup 6} yr, having a dynamical half-life of 3.4 × 10{sup 5} yr.
hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware
We present hls4ml, a free and open-source platform that translates machine learning (ML) models from modern deep learning frameworks into high-level synthesis (HLS) code that can be integrated into full designs for field-programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs). With its flexible and modular design, hls4ml supports a large number of deep learning frameworks and can target HLS compilers from several vendors, including Vitis HLS, Intel oneAPI and Catapult HLS. Together with a wider eco-system for software-hardware co-design, hls4ml has enabled the acceleration of ML inference in a wide range of commercial and scientific applications where low latency, resource usage, and power consumption are critical. In this paper, we describe the structure and functionality of the hls4ml platform. The overarching design considerations for the generated HLS code are discussed, together with selected performance results.
Connector system for photovoltaic array
A photovoltaic assembly comprising; (a) at least two photovoltaic components that are adjacent to each other in a first direction, each photovoltaic component comprising (i) a partial recess in communication with the partial recess in an adjacent photovoltaic component and (ii) one or more connector receptors aligned in a second direction which is non-parallel to the first direction; (b) a connector located at feast partially in the partial recess of the photovoltaic component and at least partially in the partial recess of the adjacent photovoltaic component so that the connector connects the photovoltaic component to the adjacent photovoltaic component, the connector comprising: (i) a flexible housing having a first end and a second end; (ii) one or more connection ports at the first end; (iii) one or more connection ports at the second end; and (iv) one more flexible electrical conductors that extend from the one or more connection ports at the first end to the one or more connection ports at the second end; wherein the connector is flexible so that the first end and the second end are movable relative to each other in a plane, out of the plane, or both; wherein the one or more connection ports at the first end and the one or more connection ports at the second end form a connection with the one or more connector receptors of the photovoltaic component and the adjacent photovoltaic component so that the connector electrically connects the photovoltaic component to the adjacent photovoltaic component.
Grassland productivity responds unexpectedly to dynamic light and soil water environments induced by photovoltaic arrays
Abstract Agrivoltaic (AV) systems are designed to coproduce photovoltaic (PV) energy on lands simultaneously supporting agriculture (food/forage production). PV infrastructure in agroecosystems alters resources critical for plant growth, and water‐limited agroecosystems such as grasslands are likely to be particularly sensitive to the unique spatial and temporal patterns of incident sunlight and soil water inherent within AV systems. However, the impact of resource alteration on forage production, the primary ecosystem service from managed grasslands, is poorly resolved. Here, we evaluated seasonal patterns of soil moisture (SM) and diurnal variation in incident sunlight (photosynthetic photon flux density [PPFD]) in a single‐axis‐tracking AV system established in a formerly managed semiarid C 3 grassland in Colorado. Our goals were to (1) quantify dynamic patterns of PPFD and SM within a 1.2 MW PV array in a perennial grassland, and (2) determine how aboveground net primary production (ANPP) and photosynthetic parameters responded to the resource patterns created by the PV array. We hypothesized that spatial variability in ANPP would be strongly related to SM patterns, typical of most grasslands. We measured significant reductions in ANPP directly beneath PV panels, where SM and PPFD were both low. However, in locations with significantly increased SM from the shedding and redistribution of precipitation by PV panels, ANPP was not increased. Instead, ANPP was greatest in locations where plants were shaded in the afternoon but received high levels of PPFD in the morning hours, when air temperatures and vapor pressure deficits were relatively low. Thus, contrary to expectations, we found relatively weak relationships between SM and ANPP despite significant spatial variability in both. Further, there was little evidence that light‐saturated photosynthesis ( A sat ) and quantum yield of CO 2 assimilation (ϕCO 2 ) differed for plants growing directly beneath (lowest PPFD) versus between (highest PPFD) PV panels. Overall, the AV system established in this semiarid managed grassland did not alter patterns of ANPP in ways predictable from past studies of controls of ANPP in open grasslands. However, our results suggest that the diurnal timing of low versus high periods of PPFD incident on plants is an important determinant of productivity patterns in grasslands.
Long-term impact of light- and elevated temperature-induced degradation on photovoltaic arrays
Low levelized cost of electricity (LCOE) has been identified as critical for widespread adoption of photovoltaics (PV) without subsidies. Maintaining decades-long high-energy production is often an under-recognized opportunity in meeting cost goals because component lifetimes are not fully quantified at the time of manufacture. Whereas certain standardized tests minimize risk of early failure, there is little guidance to quantitatively predict degradation (or lack thereof) over decades, based on accelerated tests. In this article, we move toward bridging the understanding between indoor accelerated tests and outdoor performance data, with the goal of predicting energy yield with enough accuracy to inform financial decisions. Light- and elevated temperature-induced degradation (LETID) in p-type Si modules is analyzed in terms of impact on long-term module performance and thus LCOE. A method to predict the progression of LETID, using fixed kinetic constants and a numerical solution to the basic reaction rate equations, is detailed. Predictions are compared against both published data and that new to this study. These data include both indoor accelerated tests and fielded modules. We use the results in financial models to derive LCOE of modules in different climates with varying amounts of LETID, including uncertainty. Cost models based on the predictions indicate that LETID has a significant and climate-dependent impact on LCOE. Finally, we show that - even given the uncertainties identified in the study - these financial calculations can provide useful guidance to quantify risk based on accelerated test results. The analysis serves as an example of developing a predictive approach to PV reliability using physics of failure.
Disk-integrated Thermal Properties of Ceres Measured at Millimeter Wavelengths
We observed Ceres at three epochs in 2015 November and 2017 September and October with Atacama Large Millimeter/submillimeter Array (ALMA) 12 m array and in 2017 October with the ALMA Compact Array (ACA), all at ∼265 GHz continuum (wavelengths of ∼1.1 mm) to map the temperatures of Ceres over a full rotation at each epoch. We also used 2017 October ACA observations to search for HCN. The disk-averaged brightness temperature of Ceres is measured to be between 170 and 180 K during our 2017 observations. The rotational light curve of Ceres shows a double-peaked shape with an amplitude of about 4%. Our HCN search returns a negative result with an upper limit production rate of ∼2 × 10{sup 24} molecules s{sup −1}, assuming globally uniform production and a Haser model. A thermophysical model suggests that Ceres’s top layer has higher dielectric absorption than lunar-like materials at a wavelength of 1 mm. However, previous observations showed that the dielectric absorption of Ceres decreases toward longer wavelengths. Such distinct dielectric properties might be related to the hydrated phyllosilicate composition of Ceres and possibly abundant micrometer-sized grains on its surface. The thermal inertia of Ceres is constrained by our modeling as likely being between 40 and 160 thermal inertia units, much higher than previous measurements at infrared wavelengths. Modeling also suggests that Ceres’s light curve is likely dominated by spatial variations in its physical or compositional properties that cause changes in Ceres’s observed thermal properties and dielectric absorption as it rotates.
Open‐source photovoltaic model pipeline validation against well‐characterized system data
Abstract All freely available plane‐of‐array (POA) transposition models and photovoltaic (PV) temperature and performance models in pvlib‐python and pvpltools‐python were examined against multiyear field data from Albuquerque, New Mexico. The data include different PV systems composed of crystalline silicon modules that vary in cell type, module construction, and materials. These systems have been characterized via IEC 61853‐1 and 61853‐2 testing, and the input data for each model were sourced from these system‐specific test results, rather than considering any generic input data (e.g., manufacturer's specification [spec] sheets or generic Panneau Solaire [PAN] files). Six POA transposition models, 7 temperature models, and 12 performance models are included in this comparative analysis. These freely available models were proven effective across many different types of technologies. The POA transposition models exhibited average normalized mean bias errors (NMBEs) within ±3%. Most PV temperature models underestimated temperature exhibiting mean and median residuals ranging from −6.5°C to 2.7°C; all temperature models saw a reduction in root mean square error when using transient assumptions over steady state. The performance models demonstrated similar behavior with a first and third interquartile NMBEs within ±4.2% and an overall average NMBE within ±2.3%. Although differences among models were observed at different times of the day/year, this study shows that the availability of system‐specific input data is more important than model selection. For example, using spec sheet or generic PAN file data with a complex PV performance model does not guarantee a better accuracy than a simpler PV performance model that uses system‐specific data.
Determining circuit model parameters from operation data for PV system degradation analysis: $\mathrm{PVPRO}$
Physics-based circuit parameters like series and shunt resistance are essential to provide insights into the degradation status of photovoltaic (PV) arrays. However, calculating these parameters typically requires a full current-voltage characteristic (I-V curve), the acquisition of which involves specific measurement devices and costly methods. Thus, I-V curves of the PV system level are often not available. Here this paper proposes a methodology (PVPRO) to estimate these I-V curve parameters using only operation (string-level DC voltage and current) and weather data (irradiance and temperature). PVPRO first performs multi-stage data pre-processing to remove noisy data. Next, the time-series DC data are used to fit an equivalent circuit single-diode model (SDM) to estimate the circuit parameters by minimizing the differences between the measured and estimated values. In this way, the time evolutions of the SDM parameters are obtained. We evaluate PVPRO on synthetic datasets and find an excellent estimation of both SDM and the key I-V parameters (e.g., open-circuit voltage, short-circuit current, maximum power, etc.) with an average relative error of 0.55%. The performance, especially the extracted degradation rate of parameters, is robust to various measurement noises and the presence of faults. In addition, PVPRO is applied to a 271 kW PV field system. The relative error between the real and estimated operation voltage and current is less than 1%, suggesting that degradation trends are well captured. PVPRO represents a promising open-source tool to extract the time-series degradation trends of key PV parameters from routine operation data.
Feature review of photovoltaic modeling software utilizing blind performance assessment
While confidence in photovoltaic (PV) modeling software has always been essential, the rapid pace of new PV plant developments makes accuracy and credibility more critical than ever. Independent assessments, particularly through blind modeling comparisons, are therefore necessary to ensure unbiased benchmarking across PV modeling software. Previous studies have been limited by a narrow range of models compared, anonymized results, or system size. This study presents results from the first-ever onymous blind modeling comparison, evaluated using both lab- and utility-scale fixed-tilt, monofacial, south-facing systems at sub-hourly time intervals. Seven commercially used PV software tools were compared: 3E SynaptiQ, PlantPredict, PVsyst, RatedPower, SAM, SolarFarmer, and Solargis Evaluate. Predictions were submitted directly by software representatives, providing unique insights into each software’s implementation and resulting prediction behavior. Notable features, including plane-of-array (POA) transposition model, module temperature model, shading model, and performance model were analyzed and compared. Four summary tables compile these features of the software, serving as a resource to help users understand the methodological differences and select the most suitable software for their applications. The software tools show deviations from mean error in annual yield up to 2.5 % in the lab-scale system, increasing to 6.0 % for the utility-scale system. These differences arise from a combination of user decisions and the inherent behavior of the software, indicating the need for continuous and rigorous validation of modeling methods using these software tools against complex, real-world systems.
Effect of Redox-Inactive Metal Ion–Nickel(III) Interactions on the Redox Properties and Proton-Coupled Electron Transfer Reactivity
We report Mononuclear nickel(II) and nickel(III) complexes of a bisamidate-bisalkoxide ligand, (NMe 4 ) 2 [Ni- II (HMPAB)] (1) and (NMe 4 )[Ni III (HMPAB)] (2), respectively, have been synthesized and characterized by various spectroscopic techniques including X-ray crystallography. The reaction of redox-inactive metal ions (M n+ = Ca 2+ , Mg 2+ , Zn 2+ , Y 3+ , and Sc 3+ ) with 2 resulted in 2-M n+ adducts, which was assessed by an array of spectroscopic techniques including X-ray absorption spectroscopy (XAS), electron paramagnetic resonance (EPR), and reactivity studies. The X-ray structure of Ca 2+ coordinated to Ni(III) complexes, 2Ca 2+ T, was determined and exhibited an average Ni-Ca distance of 3.1253 angstrom, close to the metal ions' covalent radius. XAS analysis of 2-Ca 2+ and 2-Y 3+ in solution further revealed an additional coordination to Ca and Y in the 2-M n+ adducts with shortened Ni-M distances of 2.15 and 2.11 Å, respectively, implying direct bonding interactions between Ni and Lewis acids (LAs). Such a short interatomic distance between Ni(III) and M is unprecedented and was not observed before. EPR analysis of 2 and 2-M n+ species, moreover, displayed rhombic signals with g av > 2.12 for all complexes, supporting the +III oxidation state of Ni. The Ni III /Ni II redox potential of 2 and 2-M n+ species was determined, and a plot of E 1/ 2 of 2-M n+ versus pK a of [M(H 2 O) n ] m+ exhibited a linear relationship, implying that the Ni III /Ni II potential of 2 can be tuned with different redox-inactive metal ions. Reactivity studies of 2 and 2-M n+ with different 4-X-2,6-ditert-butylphenol (4-XDTBP) and other phenol derivatives were performed, and based on kinetic studies, we propose the involvement of a proton-coupled electron transfer (PCET) pathway. Analysis of the reaction products after the reaction of 2 with 4-OMe-DTBP showed the formation of a Ni(II) complex (1a) where one of the alkoxide arms of the ligand is protonated. A pK a value of 24.2 was estimated for 1a. The reaction of 2-M n+ species was examined with 4-OMe-DTBP, and it was observed that the k 2 values of 2-M n+ species increase by increasing the Lewis acidity of redox-inactive metal ions. However, the obtained k 2 values for 2-M n+ species are much lower compared to the k 2 value for 2. Such a variation of PCET reactivity between 2 and 2-M n+ species may be attributed to the interactions between Ni(III) and LAs. Our findings show the significance of the secondary coordination sphere effect on the PCET reactivity of Ni(III) complexes and furnish important insights into the reaction mechanism involving high-valent nickel species, which are frequently invoked as key intermediates in Ni-mediated enzymatic reactions, solar-fuel catalysis, and biomimetic/synthetic transformation reactions.
Updated sustainability status of cadmium telluride thin‐film photovoltaic systems and projections
This paper provides a comprehensive assessment of the up-to-date life-cycle sustainability status of cadmium-telluride based photovoltaic (PV) systems. Current production modules (Series 6 and Series 7) are analyzed in terms of their energy performance and environmental footprint and compared with the older series 4 module production and current single-crystalline Silicon (sc-Si) module production. For fixed-tilt systems with Series 6 modules operating under average US irradiation of 1800 kWh/m 2 /year, the global warming potential (GWP) is reduced from 16 g CO 2eq /kWh in Series 4 systems to 10 CO 2 eq /kWh in Series 6 systems. For operation in US-SW irradiation of 2300 kWh/m 2 /year, the GWP is reduced from 11 to 8 CO 2eq /kWh and for 1-axis tracking systems operating in Phoenix, Arizona, with point-of array irradiation of 3051 kWh/m 2 /year the GWP is reduced to 6.5 CO 2eq /kWh. Similar reductions have happened in all environmental indicators. Energy payback times (EPBT) of currently installed systems range from 0.6 years for fixed–tilt ground–mounted installations at average US irradiation at latitude tilt installations to 0.3 years for one-axis trackers at high US-SW irradiation, considering average fossil-fuel dominated electricity grids with fuel to electricity conversion efficiency of 0.3. The resulting energy return on energy investment (EROI) also depends on the conversion efficiency of the electricity grid and on the operation life expectance. For a 30-year operational life and grid conversion efficiency of 0.3, EROI ranges from 50 (at US average irradiation) to 70 for US-SW irradiation. The EROI declines with increased grid conversion efficiency; for CdTe PV operating in south California with grid conversion efficiency of 49%, the EROI is about 50 and is projected to fall to 30 when the state's 2030 target of 80% renewable energy penetration materializes. Material alternatives that show a potential of further reductions in degradation rates and materials for enhanced encapsulation that would enable longer operation lives have also been investigated. A degradation rate of 0.3%/year, which has been verified by accelerated testing, is assumed in 30-year scenarios; this is projected to be reduced to 0.2%/year in the near-term and potentially to 0.1%/year in the longer term. With such low degradation rates and enhanced edge-sealing, modules can last 40- to 50-years. Consequently, all impact indicators will be proportionally reduced while EROI will increase. This detailed LCA was conducted according to ISO standards and IEA PVPS Task 12 guidelines. Furthermore, the study revealed that the choices of system models, methods and temporal system boundaries can significantly impact the results and points out to the need to include assumptions regarding these choices in the “transparency in reporting” requirements listed in the IEA PVPS Task 12 Guidelines.
hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware
We present hls4ml, a free and open-source platform that translates machine learning (ML) models from modern deep learning frameworks into high-level synthesis (HLS) code that can be integrated into full designs for field-programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs). With its flexible and modular design, hls4ml supports a large number of deep learning frameworks and can target HLS compilers from several vendors, including Vitis HLS, Intel oneAPI and Catapult HLS. Together with a wider eco-system for software-hardware co-design, hls4ml has enabled the acceleration of ML inference in a wide range of commercial and scientific applications where low latency, resource usage, and power consumption are critical. In this paper, we describe the structure and functionality of the hls4ml platform. The overarching design considerations for the generated HLS code are discussed, together with selected performance results.
Combined dark matter search towards dwarf spheroidal galaxies with Fermi -LAT, HAWC, H.E.S.S., MAGIC, and VERITAS
Dwarf spheroidal galaxies (dSphs) are excellent targets for indirect dark matter (DM) searches using gamma-ray telescopes because they are thought to have high DM content and a low astrophysical background. The sensitivity of these searches is improved by combining the observations of dSphs made by different gamma-ray telescopes. We present the results of a combined search by the most sensitive currently operating gamma-ray telescopes, namely: the satellite-borne Fermi -LAT telescope; the ground-based imaging atmospheric Cherenkov telescope arrays H.E.S.S., MAGIC, and VERITAS; and the HAWC water Cherenkov detector. Individual datasets were analyzed using a common statistical approach. Results were subsequently combined via a global joint likelihood analysis. We obtain constraints on the velocity-weighted cross section 〈σv〉 for DM self-annihilation as a function of the DM particle mass. This five-instrument combination allows the derivation of up to 2-3 times more constraining upper limits on 〈σv〉 than the individual results over a wide mass range spanning from 5 GeV to 100 TeV. Depending on the DM content modeling, the 95% confidence level observed limits reach 1.5×10 -24 cm 3 s -1 and 3.2×10 -25 cm 3 s -1 , respectively, in the τ + τ - annihilation channel for a DM mass of 2 TeV.