Different procedures have been used for the averaging of series resistance images. Trupke et al. (2007) suggested using the harmonic average of the RseriI – values, Ramspeck et al. (2007) used the arithmetic average, and Michl et al. (2008) utilized the independent series resistance model similar to the LIA approach as shown above to calculate the current-voltage characteristics and subsequently extract from these characteristics the global series resistance value. In order to find a procedure to calculate the global series resistance from series resistance images, Michl evaluated all three averaging
procedures. The harmonic-averaging approach is dominated by low Revalues. However, this is not the case for an illuminated solar cell, where every part of the solar cell is equally important due to the homogeneous current generation. In contrast, the arithmetic-averaging procedure weights every pixel equally for the global series resistance value. This results in reasonable values because in a solar cell with low series resistance at the maximum power point, the series resistance is much smaller than the diode resistance. Thus, the local diodes are ofminor importance and the current of the local element is approximately constant over the cell area. Nevertheless, for high local series resistances, the local diodes are not negligible any more. In this case, Michl recommends use of the independent series resistance approach to predict the global series resistance as determined from standard IV measurements.
