Application of an optimization tool for inverse modelling of thin-film silicon solar cells

Device simulation of thin-film silicon solar cells depends strongly on good input parameters, but many of the electronic parameters are difficult to determine by experiment. Therefore inverse modelling, i.e., the determination of simulation input parameters by adjusting simulation results to given experimental data, is very useful. We have developed a systematic procedure of inverse modelling with makes use of a particle swarm optimization algorithm. Experimentally determined external quantum efficiency, dark and illuminated current-voltage characteristics were modelled to determine the input parameters, resulting in very good agreement between the simulation and experiment.