Surrogate-assisted evolutionary optimization and microgravity experimental validation of a planar deployable mechanism

This paper proposes a surrogate-assisted evolutionary framework to optimize the design parameters of a novel planar deployable mechanism, with experimental validation conducted through microgravity tests. The mechanism features an elastic outer frame integrated with an inner rope net, utilizing stored strain energy for passive deployment to overcome the transient operational time limitations of traditional designs. The highly nonlinear deployment dynamics and computationally expensive simulations pose significant challenges for conventional optimization methods. To address this, a Genetic Algorithm (GA) is integrated with Response Surface Methodology (RSM) and Kriging surrogate models to efficiently minimize the maximum backward penetration length during deployment. A parallel evaluation strategy is implemented to significantly reduce the optimization time compared to conventional approaches. Comparative analysis demonstrates the superior performance of the Kriging model, which shows a substantial reduction in maximum residual error compared to the polynomial RSM. Experimental validation under microgravity conditions confirmed the accuracy of the mechanism simulation, and the optimal design successfully deployed with intrusion depths within required limits. The proposed methodology provides an efficient and reliable approach for optimizing complex deployable structures involving computationally expensive simulations.