摘要
建筑工程项目随着施工进度推进,结构变形程度增大,超过变形临界值时将出现安全事故,且在工程活动和外界气候环境共同作用下,易诱发滑坡等地质灾害.为及时有效发现并解决建设过程中的变形问题,以及可能诱发的地质灾害,将Bagging算法、Boosting算法、Stacking集成学习方法应用于土木工程预测中,以便更加合理地应用智能预测技术.并指出目前存在无法预测前期结构变形、集成学习种类繁多、滞后效应时效性明显的问题,对该技术应用前景进行展望.
Abstract
With the progress of the construction project,the degree of structural deformation increases,and safety accidents will occur when the deformation threshold is exceeded,under the combined action of engineering activities and external climate and environment,it is easy to induce geological disasters such as landslides.In order to find and solve the deformation problems in the construction process in time and effectively,as well as the geological disasters that may be induced,the Bagging,Boosting and Stacking ensemble learning methods is applied to civil engineering prediction,in order to apply intelligent prediction technology more reasonably.It is pointed out that there are some problems,such as the inability to predict the early structural deformation,the wide variety of ensemble learning,and the obvious timeliness of lag effect,finally,the application prospect is prospected.
基金项目
青岛市科技惠民示范专项项目计划(23-2-8-cspz-3-nsh)
NETL
NSTL
维普
万方数据