Automatic Design of Planar Spiral Inductors Based on Improved Artificial Bee Colony Algorithm

Abstract

Aiming at the low design efficiency of planar spiral inductors in Integrated Passive Devices (IPD), a design scheme for automatically searching the optimal parameters of planar spiral inductors is presented based on the Improved Artificial Bee Colony (IABC) algorithm. By improving the single π model, the universality of the inductor equivalent model is improved, and the extracted model is convenient for the parameter optimization of the inductor. The artificial bee colony (ABC) algorithm is improved by introducing multi-neighborhood topology (MNT) and modification rate mechanism, which greatly improves the optimization efficiency of the original algorithm. The optimization parameters of the improved algorithm are simulated in ADS, and the results are consistent with the multiple design cases of planar spiral inductors, which verifies the effectiveness of the algorithm. The improved algorithm significantly improves the design efficiency of the planar spiral inductor, making the design time of a single inductor only in seconds, and the accuracy rate is more than 90%.