中文摘要：

目前工程实践中的桥梁传感器布置多根据经验确定,现行规范对传感器优化布置也没有明确的理论与方法,基于此,提出基于二重结构编码遗传算法的传感器优化布置方法,即采用二重结构编码进行种群的选择、交叉和变异,在选择时采用最优保存策略,交叉时采用自适应部分匹配交叉,变异时采用自适应逆位变异的算子转化。该方法弥补了计算机应用于大型结构时数据冗长、储存空间不足的缺陷,并且搜索到最优解。通过对一个连续刚构桥工程的实例分析,证明了遗传算法在搜索能力、计算效率和可靠性方面明显优于有效独立法,但收敛速度仍需进一步完善。

英文摘要：

At present bridge sensors are mostly arranged based on past experience, and current specifications have no specific theory and methods for optimizing sensors placement, based on which, the authors propose an optimization method based on dual-structure code genetic algorithm method, which uses dual-structure code to select populations, crossover and mutation, and when choosing the optimal preservation strategy will be used, matching cross will be made for automatic parts, and adaptive inverse transformation of alleles when the operator variability will be adopted. This method makes up the structure of the data used in large computer lengthy, lack of storage space deficiencies, and the search to the optimal solution. Through the study of a continuous rigid frame bridge engineering case, it is proved that the genetic algorithm method is significantly better than he effective independence method on search capability, computing efficiency and reliability, but still need further improvement on the convergence rate.