Computer-aided analysis of multijunction solar concentrator cells and systems

Alexander W Haas, Purdue University

Abstract

Multijunction concentrator solar cells are a promising technology in the effort to alleviate the growing energy demand with renewable sources, as they are capable of high conversion efficiencies while reducing system cost by utilizing small cell areas. Yet, these cells and systems have many unique issues that must be resolved in order for them to be viable for large-scale energy production. Moreover, many of the power loss mechanisms are internal phenomenon and therefore must be modeled. Consequentially, this work is directed towards the computer-aided analysis of multijunction solar concentrator cells and systems. A curve-fit model, for use as an embedded system tool is developed. It is shown that the series resistance extracted from this model for a GaAs concentrator solar cell is within 10% of the expected value. A quasi-3D distributed emitter model is developed to simulate effects related to the lateral flow of current in a solar cell emitter layer. The quasi-3D model is applied to the analysis loss mechanisms and grid optimization for single- and dual-junction solar cells under a non-uniform illumination profile. It is shown that highly non-uniform illumination may result in a greater than 5% power loss, relative to the uniform illumination case. Further, it is shown that the bias-point loss, which results from the potential gradient across the emitter forcing most of the cell area to operate away from the local maximum power condition, plays a very significant role in determining cell performance and the optimal grid electrode pattern, particularly under highly non-uniform illumination. The quasi-3D model is also applied to the analysis of solar tracker error. It is shown that for a systematic tracker error that varies over the course of a day, only small performance degradation is found unless the illumination pattern begins to wander off the cell area. Detailed numerical models are utilized in the design of a GaInP/GaAs concentrator solar cell for maximum yearly energy production. It is found that designs optimized for AM1.5d produce nearly maximum yearly energy. This result is largely independent of geographic location and the optical concentration.

Degree

Ph.D.

Advisors

Gray, Purdue University.

Subject Area

Alternative Energy|Electrical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS