Multi-objective Antenna Design Based on Improved Sparrow Search Algorithm to Optimize BP Neural Network Surrogate Model

Abstract

In response to the problems of time-consuming and low potency of traditional antenna design methods, a multi-objective antenna design method based on MISSA-BP surrogate model is proposed. Meanwhile, for the problems of sparrow search algorithm, such as single population and difficulty in reaching global optimum, proposes a multi-strategy improved sparrow search algorithm (MISSA), three strategies of reverse learning, inertial weight and adaptive t-distribution are introduced for improvement to enrich the population diversity and strengthen the flexibility of the algorithm to reach the optimum result. The experimental results show that the improved sparrow search algorithm has higher improvement performance, and the proposed MISSA-BP surrogate model can be bound to the multi-objective optimization algorithm to achieve fast multi-objective antenna design.