在役桥梁质量主控因素与损伤识别研究  

Quality Control Factors and Damage Identification of Bridges in Service

作　　者：郑妍 李鸿钊 秦祖品 薛凤蕊 王鲁鑫 朱灿 ZHENG Yan;LI Hongzhao;QIN Zupin;XUE Fengrui;WANG Luxin;ZHU Can(Heze Transportation Management Center,Shandong High-Speed Co.,Ltd.,Heze Shandong 274000,China;Shandong Transportation Institute,Jinan Shandong 250031,China)

机构地区：[1]山东高速股份有限公司菏泽运管中心,山东菏泽274000 [2]山东省交通科学研究院,山东济南250031

出　　处：《工程质量》2024年第S01期66-69,共4页Construction Quality

基　　金：山东省交通运输厅科技计划项目(2021B40)。

摘　　要：论文从影响桥梁安全的自然因素和人为因素两方面研究了在役桥梁质量主控因素。在自然因素方面,着重讨论了环境因素和材料性能因素。在人为因素方面,重点分析了施工因素、荷载因素。最后使用随机森林模型和Adaboost模型对桥梁收集到的各种类型数据进行异常状态的判定,研究表明随机森林算法具有较高的精度,精确度均超过了0.86,为桥梁结构耐久性养护工作提供了参考。In this paper,the main control factors of bridge quality in service are studied from the aspects of natural factors and human factors that affect bridge safety.In the aspect of natural factors,the environmental factors and material properties factors are emphatically discussed.In the aspect of human factors,the construction factors and load factors are analyzed.Finally,random forest model and Adaboost model are used to determine the abnormal state of various types of data collected by the bridge.The research has shown that the random forest algorithm has high accuracy,with accuracy exceeding 0.86,providing a reference for the durability maintenance of bridge structures.

关 键 词：在役桥梁质量 影响因素 机器学习 桥梁损伤 

分 类 号：U447[建筑科学—桥梁与隧道工程]