Designing Nonlinear Optoelectronic Devices with Numerical Optimization: Lessons Learned

Photodetectors are widely used in various applications with different requirements. Choosing the right group of semiconductor materials and adjusting the parameters of the layers that form the photodetectors are crucial for optimizing their performance. This paper investigates the suitability of four numerical optimization methods - genetic algorithm, particle swarm optimization, surrogate algorithm, and the Nelder-Mead simplex algorithm - for the design of photodetectors by comparing the performance of the methods. The findings of this study can aid in selecting the right numerical optimization technique for designing photodetector-like opto-electronic devices with optimal performance for specific applications.