Parameter extraction for photodiode equivalent circuit model based on hybrid genetic algorithm

Photodiode equivalent circuit models often involve numerous parameters, making the parameter extraction process complex. Therefore, this work presents a hybrid genetic algorithm (HGA) to extract photodiode parameters. Firstly, we deduce the calculation formula of the scattering parameters of the photodiode equivalent circuit model. Then, to address the limitations of the standard genetic algorithm (SGA) that its slow convergence and extended search times, we incorporate the elite strategy and acceleration operator, enhancing the local search ability of HGA. We compare the performance of the HGA with the other four optimization algorithms under multiple sets of measured data. Results show that HGA has better performance in parameter fitting, convergence speed, and accuracy from 10 MHz to 40 GHz, highlighting its superiority. The algorithm significantly simplifies photodiode parameter extraction, and provides a feasible solution for the parameter extraction of optoelectronic devices.

Automated summary

This paper proposes a hybrid genetic algorithm (HGA) to simplify the process of photodiode parameter extraction and compares its performance with other optimization algorithms. The results show that HGA has better performance in parameter fitting, convergence speed, and accuracy.