地铁施工变形预测与控制的智能方法

Intelligent Prediction and Control Method of Deformation in Metro Construction

地铁施工变形关系到工程安全和周边土工环境维护和稳定。采用人工智能神经网络和模糊逻辑法则等手段,对地铁车站深大基坑和地铁盾构施工中的土体变形以及土工环境问题进行智能预测,在变形达到规定限值之前,通过变形控制的智能方法,调整地铁施工的参数,达到合理控制施工变形的目的。介绍了人工智能神经网络、多步滚动预测、模糊逻辑智能控制方法的原理及工程实践。经多处地铁施工证明,变形预测与控制的智能方法是可行的,效果显著。

Deformation in metro construction relates to safety of projects and maintenance and stability of surrounding geotechnical environment. With artificial intelligence neural network and fuzzy logical rule, some chief construction parameters are predicted, such as soil deformation during construction of deep excavation in metro-station work and shield driving, as well as corresponding geotechnical environmental problems. Based on the prediction, intelligent deformation control measures are suggested, thus construction parameters in metro construction are adjusted in order to control the deformation reasonably. Principle of artificial intelligent neural network, multistep roUing prediction, and fuzzy logical intelligent control are introduced in this paper, as well as its engineering applications. It is proved by some construction practice that intelligent method for deformation prediction and control is feasible and has obvious benefit.