摘要
大跨径悬浇连续梁桥的变形预测是相关工程实施过程中的重点之一。为充分挖掘大量已建悬浇施工桥梁变形监控数据的价值,使其为新建类似桥梁的变形预测提供校核数据参考,提出了采用模型树机器学习算法预测悬浇施工各节段预抬高值来预测桥梁变形的方法,并通过编写计算程序对该方法进行了测试与验证。结果表明:采用模型树机器学习算法获取悬浇施工桥梁节段预抬高值来预测桥梁变形的方法是可行的,该方法可广泛应用于桥梁工程的变形预测中,以充分挖掘已有工程数据的价值。
Pre-camber prediction of long-span PC continuous girder bridge is one of the key tasks in the process of construction stage. In order to fully exploit the value of the geometry data of a large number of as-built cantilever constructed bridges, and to provide reference review data for the geometry prediction of new bridges with similar features, this paper innovatively proposes to use the model tree algorithm to predict pre-camber data of each segment of cantilever construction bridges. This method is tested and verified by a computer program. The results show that this method is feasible by using the model tree algorithm to obtain the reference review data of pre-camber value of cantilever construction bridge segments. In addition, the algorithm is expected to be widely used in many data predictions of bridge engineering in order to fully exploit the value of existing engineering data.
作者
王培金
WANG Peijin(Shandong Provincial Communications Planning and Design Institute,Ji’nan 250031,China)
出处
《现代交通技术》
2019年第4期49-53,共5页
Modern Transportation Technology
关键词
连续梁桥
悬浇施工
模型树
机器学习
continuous girder bridge
cantilever construction
model tree
machine learning
