基于改进粒子群算法的天线索网预应力优化

为了解决天线反射面索网张拉时预应力求解问题,利用索网结构空间几何特性对其进行简化处理,从整体中提取部分索段并加设相应的边界条件.在平衡矩阵法的基础上,采用与禁忌搜索法相结合的混合粒子群算法,将简化后结构由形找力的问题转化为数学上的约束优化问题,同时改进粒子群进化算法来解这一优化问题.采用该算法对三向网格形式的天线抛物面索网结构进行了预应力的优化确定,并将优化的预应力值作为初始力导入索网结构有限元模型中进行非线性有限元分析,索网结构的节点位移量和相应的物理模型型面测量的精度均方根为0.219 mm，验证了算法是可行和正确的. 关键词：

Improved PSO based pretension optimum of cable net in antenna

In order to resolve the pretension solution when the cable net of antenna was straining, the space geometry characteristic of cable net structure was utilized to simplify the problem through extracting part structure from the whole and adding a corresponding boundary. Based on the equilibrium matrix method, the problem of force finding by form of the simplified the structure was translated into a mathematically constrained optimization problem. The particle swarm optimization (PSO) was improved and combined with the taboo search (TS) to solve the optimization problem. This method was used on a specific example about pretension optimum of parabolic cable net structure of triclinic mesh in antenna, and the results was put as the initial force into the cable net structure finite element model for nonlinear finite element analysis. The root-mean-square (RMS) of the displacement of cable net nodes and the surface accuracy of the corresponding physical model was 0.219mm, which proved the feasibility and correctness of the improved algorithm.