基于支持向量机代理模型的地铁施工诱发临近建筑扰动的参数全局敏感性分析

A global sensitivity analysis of building movements induced by tunnel excavating based on PSO-LSSVM meta-model

运用傅里叶幅度灵敏度检验扩展法(EFAST)对在隧道建筑模型中影响建筑物安全的各项输入指标的全局敏感性进行评价.以武汉地铁二号线越江段为案例进行分析,验证了所提方法的实用性.运用确定性和随机有限元,得到了能精确描述隧道与建筑物之间相互关系的180组数据,因为进行全局敏感性分析需要大量的数据,采用支持向量机构建代理模型,将数据扩展成了有10 000组数据的数据集,用以进行全局敏感性分析.敏感性分析有助于增加管理人员的经验知识,以方便其采用正确的措施对类似的地铁隧道临近建筑施工的风险安全进行应对管理.

The paper proposes a novel global sensitivity analysis method, namely extended Fourier ampli- tude sensitivity test（EFAST）algorithm, to explore the building disturbance induced by tunnel excavating. The Wuhan Yangtze River Metro Tunnel （WYRMT） is taken as a case study to verify the applicability of the proposed approachl The deterministic and stochastic FE model are used to develop the approximate re- lationship between the input and Output parameters with high level of accuracy; and then based on the sim- ulated data from FE model, a meta-model is constructed by support vector machine （SVM）; 10,000 groups of data are generated by the meta-model, using for sensitivity analysis. Sensitivity analysis can help enhancing the knowledge of designers and aiding them to optimize the design or management scheme of the tunnel excavating when they are confronting the similar tunnels.