Photovoltaic Systems


Modeling Annual Energy Yield of CPV Systems:
Concentrator photovoltaic (CPV) systems use one or more optics to concentrate a large area of sunlight onto a small-area and highly-efficient solar cell. Accurate modeling of the power and energy production of CPV systems requires detailed knowledge of the local solar resource, behavior of the optics, and performance of the solar cells. Combining these factors together into a single coherent model provides accurate estimates for expected instantaneous power and annual energy yields under a variety of system and environmental conditions. This modeling provides feedback for design optimization of the solar cells to maximize the energy output, and has demonstrated the utility of designing a cell for a representative spectrum that is different from the traditional target AM1.5D standard spectrum.

 

Modeling Annual Energy Yield of Bifacial PV Systems:
A bifacial panel is a photovoltaic panel that can absorb light on both its front and rear face. This increases the available solar resource by accepting diffuse light from the sky, ground, and other objects. We have developed a new computationally-efficient algorithm for the evaluation of annual energy yields from bifacial panels. This model includes detailed anisotropic sky dome and albedo ray tracing with directional reflection, self-shading, and rack shading for the illumination of both front and rear faces. The model is being used to investigate the influence of environmental conditions and system configurations on the instantaneous power production and annual energy yield for specific deployment locations

Simulation computational zones and irradiance on the back face of a bifacial panel.

 

Modeling CPV Optics using Ray Tracing:
A typical solar concentrator design employs a Fresnel lens as the primary optical element (POE) used for concentration. Efficiency of the overall CPV system is determined by the efficiency of the solar cell as well as the concentrating optics. Irradiance uniformity is also an important factor for designing concentrators since non-uniform illumination leads to an increase in series resistance within the cell and causes a reduction in efficiency. Therefore, a homogenizer is often employed as a secondary optical element to improve irradiance uniformity.

Ray trace with a Fresnel lens and a homogenizer (secondary optics of various shapes)

In this project, a comparative evaluation of the three types of secondary optics (see picture) has been done using ray-tracing.

Incoherent irradiance at 632nm with a (a) CPC, (b) truncated pyramid, (c) truncated cone, and (d) no secondary optical element

 

Modeling Non-Uniform Illumination:
A two-dimensional distributed circuit model constructed in SPICE was used to simulate lattice-matched triple- and quadruple-junction solar cells (using 2-diode models for each subcell) for understanding cell performance under non-uniform illumination. The overall cell performance is governed not only by the selection of materials and device structures, but also by layout of the front metal contact grid design. Often the front contact grid has been optimized for uniform illumination, despite the delivery of potentially highly non-uniform illumination by the concentrating optics. Therefore this project sought to optimize grid design under non-uniform illumination from a variety of concentrating optics and conditions.

2D distributed circuit model for a 3-junction solar cell and detailed SPICE equivalent circuit model of the boxed region

Various Gaussian profiles were used to simulate non-uniform illumination effects on device performance. Designs based on optimal spacing for non-uniform illumination show an efficiency increase of more than 0.5% (absolute) at concentrations greater than 500 suns.

Efficiency as a function of concentration at different spacings under a Gaussian illumination profile with peak to average ratio of (a) 2.6 and (b) 5.3

 

Chromatic Aberration:
The non-uniform illumination produced by concentrating optics is also spectrally-dependent due to chromatic aberration. As a result, the illumination profile incident on each subcell is different and consequently affects their current profiles, forcing lateral current spreading to maintain the current-matching condition. To quantify these effects, we have employed our 2D distributed circuit model in triple- and quadruple-junction solar cells to determine the impact of chromatic aberration when using a Fresnel lens and truncated pyramid based concentrating optical system. Neglecting chromatic aberration (an effect that cannot be incorporated into lumped circuit models) was found to over-state cell efficiency by 3.2% at 1000 suns.

A functional block in a 2-D distributed circuit model for a 4-junction solar cell and the corresponding irradiance distributions

 

PV System Performance on the Grid:
How do PV systems perform in Ottawa? Are they affected by snow? How do we optimize system orientation for non-FIT revenues? The following two reports were completed in collaboration with Hydro Ottawa Limited:
Report 1 – Energy Yield Analysis of Installed Systems
Report 2 – Matching of PV to Grid Pricing and Grid Peaks

 

PV and CPV System Cost Modeling:
Technological advances in CPV cells and CPV systems are realizing very high efficiency outputs, and a high rate of improvement year over year. We have studied the systems costs trends versus time and applied learning curve analysis to the results, and CPV looks very favorable in the long term compared with PV system costs. On-going work includes extensions to complete project return on investment analysis, and sensitivities to numerous factors.

More information on this study can be found Professor David Wright’s home page.

 

Wind Load Studies on Solar Structures:
The interactions of solar panels and solar trackers with wind are studied using outdoor experiments and simulations.
More information on this study can be found on Dr. Elena Dragomirescu’s home page.