Multi-objective particle swarm optimization on ultra-thin silicon solar cells
- PDF / 1,582,978 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 113 Downloads / 177 Views
RESEARCH ARTICLE
Multi-objective particle swarm optimization on ultra-thin silicon solar cells Ipek Anil Atalay1 Hamza Kurt1
•
Hasan Alper Gunes1 • Ahmet Mesut Alpkilic1
•
Received: 9 January 2020 / Accepted: 16 September 2020 Ó The Optical Society of India 2020
Abstract Finding optimized parameters for any photonic device is a challenging problem, because as the search space enlarges the computation time and design complexity increase. For higher performance solar cells, various studies have been carried out to procure optimized parameters, to attain better performance and low cost as well. In this study, we used a multi-objective particle swarm optimization approach to search design space effectively and obtain fixed parameters for enhanced solar spectrum absorption. Numerical investigations are conducted for pyramid surface pattern, to find proper solar cell parameters for minimum reflection and maximum light trapping which give rise to enhanced absorption of photons. For the ultra-thin-film silicon solar cell having a thickness of 1 lm, a designed double-sided pyramid structure provides an ideal short-circuit photocurrent of 34.23 mA/cm2. In this regard, the proposed approach can be applied to different film thicknesses of semiconductors for different photonic applications by manipulating the reflection/transmission coefficient and light trapping mechanism. Keywords Solar cells Anti-reflection Absorption enhancement Surface texturing Light trapping Multiobjective particle swarm optimization
& Ipek Anil Atalay [email protected] 1
Department of Electrical and Electronics Engineering, TOBB University of Economics and Technology, 06560 Ankara, Turkey
Introduction Achieving improved performance for any photonic device is drawing a great deal of interest. However, as the devices become more sophisticated in the course of time, a limited number of parameters remain insufficient to entirely search the solution space and find an optimum solution. In order to solve this dilemma, parameter space search methods like genetic algorithm, particle swarm optimization, and inverse design methods are used in nanophotonic devices [1–3]. There is a great deal of research conducted on solar cells to reach better performance and low cost. Hence, ultra-thinfilm solar cells are found to be not only by the scientific communities but also by the commercial communities. Due to the use of less silicon (Si), thin-film photovoltaic technology has the potential for a higher performance to cost ratio, which means it is one of the candidates to take over in the long term from the current commercial-dominant Si wafer-based technology point of view [4]. To achieve high performance in thin-film solar cells, effective light trapping and adequate coupling of solar light are required, since the very fine thickness is not able to completely absorb the solar spectrum. In other words, to increase the absorption performance, additional features like surface texturing or usage of nanoplasmonics is needed. Yet, finding a structure
Data Loading...
