Improving the performance of photovoltaic laser power converters using automatic global optimization techniques

To develop the potential of the High Power Laser Transmission technology, the efficiencies of the Laser Power Converters (LPCs) need to improve. New optimization methods are required to find the optimal configurations of a new generation of ultra-efficient LPCs. In this work we explore the suitability of the SHGO algorithm, a recently published blackbox, global optimization algorithm, to optimize the TCAD modeling of a state-of-the-art GaAs-based Laser Power Converter. We found various configurations that improve the efficiency by a 7.3% of the original device structure in computational times under 17h.