Conventional optimization algorithms face challenges such as lengthy computation times, premature termination at non-convergent points, and the generation of local optima when addressing multi-objective optimization. A multi-objective optimization method based on the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is proposed for optimizing Doherty power amplifier circuits. The pre-layout simulation results show that, compared to traditional design methods, the optimized Doherty power amplifier circuit achieves a 6.4% increase in saturation efficiency, a 3.3% increase in 6 dB roll-off efficiency, and a 1 dB increase in saturation output power at 2.63 GHz. This approach enables multi-objective optimization design for more complex PA circuits and enhances the overall circuit performance.
Qu et al. (Thu,) studied this question.