Integrated Front–Rear-Grid Optimization of Free-Form Solar Cells

Abstract

Free-form solar cells expand solar power beyond traditional rectangular geometries. With the flexibility of being installed on objects of daily use, they allow making better use of available space and are expected to bring in new possibilities of generating solar power in the coming future. In addition, their customizable shape can add to the aesthetics of the surroundings. Evidently, free-form solar cells need to be efficient as well. One way to improve their performance is to optimize the metallization patterns for these cells. This work introduces an optimization strategy to optimize the metallization designs of a solar cell such that its performance can be maximized. For the purpose of optimization, we model an existing transparent free-form solar cell design, including front and rear electrode patterns, to validate it against previously published experimental results. The front and rear metallizations of this transparent freeform solar cell are subsequently redesigned using topology optimization. More than 50% improvement in output power is achieved by using topology optimization.

CPV solar cell modeling and metallization optimization

“…However, the proposed approach is general and also applicable to other solar cell types. In the past, variants of the method have been used for other cell types such as thin films (Gupta et al, 2014), organic cells (Gupta et al, 2017).…”

“…While the discussion is restricted to modeling only the front metallization pattern, the rear side can as well be modeled with slight modifications. To adapt the model for the rear side metallization design, see (Gupta et al, 2017).…”

“…For the sake of simplicity, the resistive components associated with the rear side as well as the edges, i.e., R c r , and R edge , are ignored. However, the inclusion of resistive components associated with the rear side should be evident from the model description and from the study presented in Gupta et al (2017). With these simplifications, R s can be restated as…”

“…TO does not impose any restriction on the design of the metal grids and is capable of generating metallization patterns that cannot be obtained with any of the previously existing methods (Gupta et al, 2015). An application where the advantage of TO has been particularly clear is the design of metallization patterns for freeform solar cells, where the traditional patterns are not suited and intuition based designs are far from optimal (Gupta et al, 2016(Gupta et al, , 2017. Under higher illumination intensity (more than one sun), the photoillumination current density is increased, which in turn leads to a larger voltage drop on the front side of the cell.…”

Improving Solar Cell Metallization Designs using Convolutional Neural Networks

“…Recently, Gupta et al (2015) presented the application of topology optimization methods (TO) capable of optimizing the front metallization designs with no apriori assumption on the shapes. Due to its flexibility, TO allows to design more efficient solar cells for any given variable set of environmental conditions, arbitrary choice of cell geometries and busbar locations (Gupta et al, 2016a;b) and concentrated photovoltaics (CPV) with non-uniform illuminations (Gupta et al, 2018).…”