Development and experimental verification of the adaptive cable-strut antenna array

Abstract

This study introduces a machine learning-based framework aimed at enhancing the precision of cable-strut antenna arrays, addressing a crucial need in advanced antenna deployment. The framework employs the force density method for shape sensing and incorporates genetic algorithm optimization, markedly improving the control of antenna shape and effectively reducing the time lag in actuators' response for real-time control. Through experiment, the framework's effectiveness was validated, demonstrating substantial improvements in the shape accuracy of the antenna arrays. The study further explores the results of various cases on a prototype, revealing the control system's capacity to activate actuators for optimal control. Digital Image Correlation (DIC) is taken for analyzing shape deformation, thereby substantiating the precision of control system. This research makes a contribution to the field by advancing the application of adaptive control techniques in large-scale phased antenna arrays, underscoring its potential to enhance construction and deployment efficiency in complex engineering systems.