论文部分内容阅读
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。
At present, most bridge sensors in engineering practice are determined empirically. However, there is no clear theory or method for sensor optimization in current codes. Based on this, a sensor optimization layout method based on dual structure coding genetic algorithm is proposed, that is, The selection, crossover and mutation of the population were carried out. The optimal preservation strategy was adopted in the selection. The adaptive partial matching crossover was used in the crossover, and the operator using the adaptive back-mutation was used in the mutation. This method makes up for the shortcomings of the computer when the data is applied to the large-scale structure, the shortage of storage space, and searches for the optimal solution. Through the analysis of an example of a continuous rigid frame bridge project, it is proved that the genetic algorithm is obviously superior to the effective independent method in terms of search ability, computational efficiency and reliability, but the convergence speed still needs to be further improved.