PID Parameter Tuning for Buck Controllers Based on an Improved Differential Evolution Algorithm

Abstract

With the wide application of Buck converters, it is of great significance to select the optimal PID controller parameters to ensure the dynamic and steady-state performance of the Buck converters. On this basis, a PID parameter tuning method based on an improved differential evolution algorithm is proposed. Also, a crossover approach with population generation across orders of magnitude, elastic boundary absorption, and retention of some combined features is put forward. In addition, an adaptive mutation factor, and a cost function that suppresses overshoot and is sensitive to the tuning time are designed. The results of 10 tests show that the differential evolution algorithm fails in the case of parameters in a range cross orders of magnitude, while the algorithm proposed in this paper enables a better combination of PID parameters selected in terms of speed and stability, with a wide search space and stable convergence results.