基于混合遗传算法的四线钢桁斜拉桥索力优化研究

Study on Cable Force Optimization of Steel Truss Cable Stayed Bridge Carrying Four Rail Tracks Based on Hybrid Genetic Algorithm

基于结构应变能最小目标,针对不同子结构的内力分量应变能实施加权,将应力、索力、位移三种约束条件以罚函数形式纳入优化模型,提出适用于高速铁路四线钢桁斜拉桥成桥索力的无约束优化模型。针对该模型,设计一种混合遗传算法,该算法将模拟退火机制并引入到传统遗传算法的代际更替中,增强了算法在迭代初期的广泛化搜索能力和末期的精细化搜索能力,从而保证优化的效率和全局最优性。以杭绍台铁路椒江特大桥480m主跨钢桁斜拉桥为例,对其进行索力优化,优化后结构整体受力更加均匀,局部受力改善明显。研究表明,所提出的优化算法具有较高的效率和较强的全局收敛性,该算法在斜拉桥索力优化方面具有较高的实用价值。

Based on the objective of minimizing the strain energy of the main bridge structure,the stress,cable force and displacement constraints are incorporated into the optimization model in the form of penalty function,and an unconstrained optimization model is proposed for cable forces of steel truss cable-stayed bridge carrying four rail tracks on high-speed railway.For this model,a hybrid genetic algorithm is designed to simulate annealing mechanism and adapt it to the inter-generational replacement of the traditional genetic algorithm,which enhances the algorithm’s extensive searching ability at the beginning of the iteration and fine searching ability at the end of the iteration.Thus,the efficiency of the optimization and global optimality can be guaranteed.With reference to the 480 main span steel truss cable-stayed bridge of Jiaojiang Super Major Bridge on Hangzhou-Shaoxing-Taizhou Railway,the cable forces of the bridge are optimized.After optimization,the overall force distribution is more uniform and the local force distribution is improved obviously.The optimization results show that the proposed optimization algorithm has higher efficiency and stronger global convergence,and the algorithm can be widely used in cable force optimization of steel truss cable stayed bridge.