Search NASA⌕ Search

Engineering topics

Smith, Sarah E.

Publications and source records attributed to Smith, Sarah E..

Particle transport-driven flow dynamics and heat transfer modulation in solar photovoltaic modules: Implications on soiling

Wind-swept debris and excessive heating reduce solar photovoltaic efficiency and accelerate panel degradation. For many solar plants in open environments, particle deposition is well-understood to exacerbate efficiency losses by blocking useable light and insulating panel surfaces. However, little is known about particle behavior in multi-panel systems, where complex wakes dominate and alter trajectories. This study examines debris-laden flow within a scaled array of two model solar panels, observing convective cooling and focusing on turbulent inter-panel flow dynamics. Convection is compared between the first and second panels (P 1 and P 2 ) and for a single inflow velocity at four particle concentrations (Φ v ), including a neutrally buoyant particle (i.e. tracer) case and three water droplet cases. Tracers and Φ v > 0 concentrations show increased convective heat transfer for P 2 compared to P 1 , while inertial particle presence and increased concentration reduces this improvement from +170% to +90%. Particle image velocimetry surrounding P 2 reveals dampened inter-panel wake features for particle-laden flow compared to the tracer case, where diminished mean flow and mean kinetic energy breakdown are observed for the particles due to their inertia. Together, the reduction in comparative convection and dampened wake effects accompanying inertial particle concentration motivates further studies on coupled array physics in debris-filled flow.

14 SOLAR ENERGY↗

Viewing convection as a solar farm phenomenon broadens modern power predictions for solar photovoltaics

We report heat mitigation for large-scale solar photovoltaic (PV) arrays is crucial to extend lifetime and energy harvesting capacity. PV module temperature is dependent on site-specific farm geometry, yet common predictions consider panel-scale and environmental factors only. Here, we characterize convective cooling in diverse PV array designs, capturing combined effects of spatial and atmospheric variation on panel temperature and production. Parameters, including row spacing, panel inclination, module height, and wind velocity, are explored through wind tunnel experiments, high-resolution numerical simulations, and operating field data. A length scale based on fractal lacunarity encapsulates all aspects of arrangement (angle, height, etc.) in a single value. When applied to the Reynolds number Re within the canonical Nusselt number heat transfer correlation, lacunarity reveals a relationship between convection and farm-specific geometry. This correlation can be applied to existing and forthcoming array designs to optimize convective cooling, ultimately increasing production and PV cell life.

14 SOLAR ENERGY↗

Increased panel height enhances cooling for photovoltaic solar farms

We report solar photovoltaic (PV) systems suffer substantial efficiency loss due to environmental and internal heating. However, increasing the canopy height of these systems promotes surface heat transfer and boosts production. This work represents the first wind tunnel experiments to explore this concept in terms of array flow behavior and relative convective heat transfer, comparing model solar arrays of varied height arrangements - a nominal height, extended height, and a staggered height configuration. Analyses of surface thermocouple data show average Nusselt number (Nu) to increase with array elevation, where panel convection at double height improved up to 1.88 times that of the nominal case. This behavior is an effect of sub-array entrainment of high velocity flow and panel interactions as evidenced through flow statistics and mean kinetic energy budgets on particle image velocimetry (PIV) data. The staggered height arrangement encourages faster sub-panel flow than in the nominal array. Despite sub-array blockage due to the lower panel interaction, heat shedding at panel surfaces promotes improvements on over 1.3 times that of the nominal height case.

14 SOLAR ENERGY↗

An Automated Algorithm to Screen Massive Training Samples for a Global Impervious Surface Classification

An algorithm is developed to automatically screen the outliers from massive training samples for Global Land Survey - Imperviousness Mapping Project (GLS-IMP). GLS-IMP is to produce a global 30 m spatial resolution impervious cover data set for years 2000 and 2010 based on the Landsat Global Land Survey (GLS) data set. This unprecedented high resolution impervious cover data set is not only significant to the urbanization studies but also desired by the global carbon, hydrology, and energy balance researches. A supervised classification method, regression tree, is applied in this project. A set of accurate training samples is the key to the supervised classifications. Here we developed the global scale training samples from 1 m or so resolution fine resolution satellite data (Quickbird and Worldview2), and then aggregate the fine resolution impervious cover map to 30 m resolution. In order to improve the classification accuracy, the training samples should be screened before used to train the regression tree. It is impossible to manually screen 30 m resolution training samples collected globally. For example, in Europe only, there are 174 training sites. The size of the sites ranges from 4.5 km by 4.5 km to 8.1 km by 3.6 km. The amount training samples are over six millions. Therefore, we develop this automated statistic based algorithm to screen the training samples in two levels: site and scene level. At the site level, all the training samples are divided to 10 groups according to the percentage of the impervious surface within a sample pixel. The samples following in each 10% forms one group. For each group, both univariate and multivariate outliers are detected and removed. Then the screen process escalates to the scene level. A similar screen process but with a looser threshold is applied on the scene level considering the possible variance due to the site difference. We do not perform the screen process across the scenes because the scenes might vary due to the phenology, solar-view geometry, and atmospheric condition etc. factors but not actual landcover difference. Finally, we will compare the classification results from screened and unscreened training samples to assess the improvement achieved by cleaning up the training samples. Keywords:

Tan, Bin↗