Solar panels on spacecraft | EPFL Graph Search

Spacecraft operating in the inner Solar System usually rely on the use of power electronics-managed photovoltaic solar panels to derive electricity from sunlight. Outside the orbit of Jupiter, solar radiation is too weak to produce sufficient power within current solar technology and spacecraft mass limitations, so radioisotope thermoelectric generators (RTGs) are instead used as a power source. History The first practical silicon-based solar cells were introduced by Russell Shoemaker Ohl, a researcher at Bell Labs in 1940. It was only 1% efficient. In April 25, 1954 in Murray Hill, New Jersey. They demonstrated their solar panel by using it to power a small toy Ferris wheel and a solar powered radio transmitter. They were initially about 6% efficient, but improvements began to raise this number almost immediately. Bell had been interested in the idea as a system to provide power at remote telephone repeater stations, but the cost of the devices was far too high to be p

Deep learning in the built environment: automatic detection of rooftop solar panels using Convolutional Neural Networks

Roberto Castello, Martin Esguerra, Adrian Guerra, Jean-Louis Scartezzini

Mapping the location and size of solar installations in urban areas can be a valuable input for policymakers and for investing in distributed energy infrastructures. Machine Learning techniques, combined with satellite and aerial imagery, allow to overcome the limitations of surveys and sparse databases in providing this mapping at large scale. In this paper we apply a supervised method based on convolutional neural networks to delineate rooftop solar panels and to detect their sizes by means of pixel-wise image segmentation. As input to the algorithm, we rely on high resolution aerial photos provided by the Swiss Federal Office of Topography. We explore different data augmentation and we vary network parameters in order to maximize model performance. Preliminary results show that we are able to automatically detect in test images the area of a set of solar panels at pixel level with an accuracy of about 0.94 and an Intersection over Union index of up to 0.64. The scalability of the trained model allows to predict the existing solar panels deployment at the Swiss national scale. The correlation with local environmental and socio-economic variables would allow to extract predictive models to foster future adoption of solar technology in urban areas.

IOP Science2019

Reactively Sputtered Nano-Structured Multilayer Coatings on Architectural Glazing for Active Solar Energy Façades

Stefan Mertin

Improving the aesthetic of solar thermal collectors would grant architects a huge potential for a perfect integration of solar panels into building façades: this can be achieved by coloured interference coatings on the cover glass. These coatings also open the possibility to match the colours of the collectors to those of other architectural components and design elements. An important boundary condition for the coloured filter production is however, to provide a sufficiently high solar transmittance. To deal with the latter, sophisticated numerical simulations are needed to improve the coloured filter design. In addition, a large variety of coating materials is important, ideally with huge differences in their refractive indices, while the absorption of all materials needs to be close to zero. This doctoral thesis focuses on the optical film design and the deposition of new coloured thin-film multilayers with an optimised energetic performance and an only little varying coloured reflection at different viewing angles. To access more directly the correlation between solar transmittance, light spectra and visual and energetic performance of the coatings, a thin-film simulation framework has been developed on the basis of Mathematica. New coating designs have reduced the number of parameters necessary to be adjusted and tuned during the coating-development process: this could help to shorten the prototype development phase on industrial coaters significantly as well as the whole coating-development cycle. Based on the new developed coloured filter designs, multilayer stacks were deposited via reactive magnetron sputtering. The deposited coatings exhibit a solar transmittance Tsol > 87%. For greenish colour hues even a Tsol up to 91% was reached, which corresponds to less than 1% loss in Tsol (respectively solar energy performance) in comparison to the uncoated glass substrate. Furthermore, the research deals with the deposition of low-refractive magnesium fluoride (MgF2) and MgF2 containing composites. Hereby, a novel deposition process by reactive magnetron sputtering was developed for MgF2 using a metallic magnesium target and a reactive gas mixture of oxygen (O2) and carbon tetrafluoride (CF4) diluted with argon. The obtained MgF2 films exhibit a very low refractive index of n = 1.382 at 550nm, which is congruent with the polycrystalline bulk, and only a very weak absorption in the whole solar range (300-2500nm). In addition, a novel approach to deposit Mg-F-Si-O composite materials by magnetron co- sputtering from compound MgF2 and silicon dioxide (SiO2) targets was performed. The obtained films show a two-phase system with nanocrystals of MgF2 embedded in a SiO2-rich amorphous matrix. The lowest achieved refractive index amounts to n = 1.424 at 550nm combined with a negligible extinction coefficient (k < 10-9). The investigated sputtered low-index materials open-up manifold possibilities for coloured filters. For instance, by adding a supplementary MgF2 inter-layer, a colour-invariant solar transmittance of Tsol = 85-85.7% could be achieved, as shown by numerical simulations. With this novel optical filter design it will be feasible to offer a whole colour palette for solar thermal collectors exhibiting all the same solar thermal energy performance.

EPFL2015

Reactively Sputtered Nano-Structured Multilayer Coatings on Architectural Glazing for Active Solar Energy Façades

Stefan Mertin

EPFL2015