【摘要】 针对大跨斜拉桥索力优化问题的非线性特点以及简单遗传算法易早熟收敛的缺陷,提出了一种基于多种群遗传算法的斜拉桥索力优化方法。以结构的最小加权弯曲应变能作为目标函数,并加以多约束条件,建立索力优化模型;通过改进多种群遗传算法的编码方式,将其应用于大跨斜拉桥的索力优化中;在此基础上综合运用MATLAB及APDL语言编制了大跨斜拉桥索力优化程序。以南京长江三桥为工程实例进行计算分析,结果显示采用多种群遗传算法优化斜拉桥索力是可行且有效的,优化后全桥结构内力均匀合理,该方法操作简单,具有较高的实用价值。更多还原

【Abstract】 A cable force optimization method based on multi-population genetic algorithm(MPGA)is put forward,taking into account the non-linear problem concerning cable force optimization of long-span cable-stayed bridge and the defects of premature convergence of the simple genetic algorithm(SGA).The cable force optimization model is established by taking the minimum weighted bending strain energy of the structure as the target function and with multiple restrictions.The MPGA is employed to the optimization of long-span cable-stayed bridge based on the upgradation of the coding mode of the MPGA,and the MATLAB and APDL language were used to compile the cable force optimization program of long-span cable-stayed bridge.The Third Nanjing Changjiang River Bridge was taken as the engineering example to conduct calculation and analysis.The results indicate that it is feasible to use MPGA to optimize the cable forces of the cablestayed bridge and the internal force of the overall bridge structure is uniform and reasonable,proving that the method is easy to practice and has wide applicability.更多还原