基于响应面法和粒子群算法的桥梁高耸临时提升支架优化

Optimization of Bridge High Temporary Lifting Support Based on Response Surface Method and Particle Swarm Algorithm

为获得大跨钢箱拱桥整体临时提升支架的最优结构参数,提出基于响应面法和粒子群算法的多目标优化方法。该方法选择提升支架钢管内混凝土填充高度、钢管外径和钢管壁厚为优化设计参数,以拱肋提升过程中支架的最大拉应力、最大压应力、屈曲系数和工程造价为优化目标参数,基于数值模拟试验的计算结果分别拟合出各优化目标参数的响应面函数;通过归一化处理、引入安全系数和成本系数,构建包含4个优化目标参数的单目标优化函数,采用粒子群优化算法进行结构参数优化。结果表明:当安全系数和成本系数分别取0.3和0.7时,提升支架目标优化函数得到较优解,此时在支架最大压应力和屈曲系数变化不大的情况下,结构最大拉应力明显降低,降幅达23%,安全度得到较大提升,且工程造价略有降低。

In order to obtain the optimal structural parameters of integral temporary lifting support of long-span steel box arch bridge,a multi-objective optimization method based on response surface method and particle swarm algorithm was proposed.For this method,concrete filling height,outer diameter and wall-thickness of the steel pipe of the lifting support were selected as the optimal design parameters.The maximum tensile stress,maximum compressive stress,buckling coefficient and project cost of the support during the process of arch rib lifting were taken as optimization target parameters.Based on the calculation results of the numerical simulation test,the response surface function of each optimization target parameter was fitted individually;and then,through normalization,the safety factor and the cost factor were introduced to construct a single objective optimization function containing four optimization objective parameters.And then,the structural parameters were optimized by using the particle swarm optimization algorithm.The results show that when the safety factor and the cost factor are 0.3 and 0.7,respectively,the objective optimization function of the lifting support obtains a better solution;meanwhile,the maximum compressive stress and the buckling coefficient of the support only reveal minor changes,the maximum tensile stress of the structure is significantly reduced with the decrease reaching 23%,the safety is greatly improved,and the project cost is slightly reduced.