摘要： 将盾构推进的过程划分成7个阶段,每个阶段用一个神经网络进行模拟,在此基础上构造成多级神经网络,拟合盾构法隧道施工中施工参数与地面沉降之间关系的数学模型。最后根据保证质量,兼顾效益和效率的目标,采用遗传算法进行施工参数的优化匹配,提出相应的控制方案。该方法在多项工程项目中使用,结果表明,此方法对工程实际施工有指导作用。

关键词: 盾构法隧道, 施工参数, 多级神经网络, 遗传算法

Abstract: This paper divides whole shield drive process into seven stages, each stage is simulated by a artificial neural network. The multi-level artificial neural network, which is composed by the seven networks, builds the relation model between construction parameters and the ground settlement. According this model, it optimizes the control scheme to assure quality, increases speed and decreases cost by genetic algorithms. This approach is applied to many tunneling project and the result shows good prospect.

中图分类号: 

• U455