Multi-objective EM-driven design of integrated spiral inductors by Pareto front exploration

In this paper, multi-objective design of integrated spiral inductors is investigated. The method aims at finding the best possible trade-offs between inductor size and its quality factor. We adopt a penalty function approach to achieve a required inductance at a given operating frequency. The design process exploits a Pareto front exploration technique with trust-region-embedded gradient search as the optimization engine. Our considerations are illustrated using an example of a 3.5-nH 3.5-turn spiral inductor in a single-ended configuration.