Optimization design of composite rod lay-up for deployable truss structure

In order to improve the dynamic performance of deployable truss structure, a two-level optimization method for composite rods is proposed. The first level optimization takes the number of standard lay-ups as design variable, and the weighted sum of the fundamental frequencies of the truss structure in the folded and deployed states as the objective function. The simulated annealing and Bayesian regularization are employed to obtain the optimal standard spread layers of BP neural network, where the optimal number of standard lay-ups is obtained by the neural network mapping. Based on the results of the first level optimization, the second level optimization takes the orientation and thickness of each layup as the design variables, and a multi-objective optimization is formulated with the objective is to maximize the fundamental frequency and minimize the structural mass. The multi-objective optimization problem is solved by using the NSGA-II algorithm. The results demonstrate that the proposed method can effectively increase the fundamental frequency of the deployable truss structure in both the folded and deployed states, and at the same time realize the lightweight design.