Fast electrical modelling for spatially-resolved characterization of amorphous silicon photovoltaic cells

Authors: X. Wu, M. Bliss, J. Zhu, TR. Betts, R. Gottschalg

40th IEEE Photovoltaic Specialists Conference (PVSC-40), Denver, USA, pp. 2620-2625, June 2014.

A fast spatially-resolved electrical modelling approach for amorphous silicon (a-Si) photovoltaic (PV) cells is introduced. An a-Si PV cell can be represented by a discrete circuit network of equivalent diode units and series resistances. The local diode unit is based on Merten’s variant of single-diode model, including an additional loss term an effective mobility lifetime. This approach considers the lateral resistances and uses a PV-oriented nodal analysis (PVONA) algorithm as the simulation engine. PVONA allows the simulation of an operating point of a cell with 1000×1000 subcells in 15 minutes on a standard desktop PV. Case studies show that it is feasible to be applied in simulating spatially-resolved characterization measurement e.g. electroluminescence (EL) imaging, or in predicting the performance of a cell under different operating conditions, for quantitatively analysis of the global and local electrical properties. The method has been applied to simulate a-Si PV cells operating under different conditions, finding that it is capable of accessing local operating points and the overall I–V characteristics.

Paper: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6925467

Draft: https://www.researchgate.net/publication/267155029

 

elecMod_wu_1

Customized sparse data structure for the nodal equation system. Top: data organization of a “node” object in the circuit network; Middle: the i-th node represents the i-th row in the G matrix; Bottom: the indexing system of the sparse data structure for matrix-vector multiplication in CG.