Safety risk analysis of bridge hanging basket construction based on fuzzy dynamic Bayesian network

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Nov.2025, Vol.48 No.11: 1564-1571

DOI: 10.3969/j.issn.1003-5060.2025.11.019

Authors

XIONG Yuxuan $ ^{1} $, HE Wenyu $ ^{1} $, LIU Guixiang $ ^{2} $, RUAN Maoqing $ ^{2} $, CHENG Huacai $ ^{3} $

(1. School of Civil and Hydraulic Engineering, Hefei University of Technology, Hefei 230009, China; 2. China Railway Beijing Engineering Group Co., Ltd., Beijing 102308, China; 3. Anhui Expressway Engineering Test and Research Center Co., Ltd., Hefei 230031, China)

Abstract and Keywords

Abstract: In view of the dynamic characteristics of safety risk evolution over time in the construction process of bridge hanging baskets, this paper puts forward a fuzzy dynamic Bayesian network (FDBN) based safety risk analysis method for bridge hanging baskets. Firstly, the hazard sources of bridge hanging basket construction are classified from four aspects: structure, behavior, environment and process. Secondly, fault tree is used to analyze the causes and risk factors of the collapse accident, and the fault tree model is transformed into a Bayesian network topology, and then combined with observation nodes, a dynamic Bayesian network (DBN) model is established. Then, fuzzy set theory and expert judgment method are used to determine the parameters of DBN model, predict the dynamic evolution of risk probability during construction, and update the dynamic risk analysis results of FDBN model in real time according to the remote monitoring information of construction site. Finally, a practical bridge basket construction is taken as an example to verify the effectiveness of the proposed method. The results show that the method combined with the real-time monitoring data of construction site can predict the probability change of the safety risk of bridge hanging baskets scientifically.

Keywords: bridge hanging basket; dynamic Bayesian network(DBN); fuzzy set theory; monitoring data; real-time updates; safety risk analysis

作者信息

熊雨轩 $ ^{1} $，贺文宇 $ ^{1} $，刘贵香 $ ^{2} $，阮茂青 $ ^{2} $，程华才 $ ^{3} $

（1. 合肥工业大学土木与水利工程学院，安徽合肥 230009；2. 中铁北京工程局集团有限公司，北京 102308；3. 安徽省高速公路试验检测科研中心有限公司，安徽合肥 230031）

摘要和关键词

摘要: 针对挂篮施工过程中安全风险随时间演变的动态特征,文章提出基于模糊动态贝叶斯网络(fuzzy dynamic Bayesian network, FDBN)的桥梁挂篮安全风险分析方法。从结构、行为、环境、流程4个方面对桥梁挂篮施工的危险源进行分类;采用故障树分析挂篮坍塌事故的成因和风险因素,并将故障树模型转化为贝叶斯网络拓扑结构,再结合观测节点建立桥梁挂篮坍塌事故风险动态贝叶斯网络(dynamic Bayesian network, DBN)模型;采用模糊集理论和专家判断法确定DBN模型参数预测施工过程中风险概率的动态演变,进而根据施工现场远程监测的信息,实时更新FDBN模型动态风险分析结果;以实际桥梁挂篮施工为例,验证该文方法的有效性。结果表明:该方法结合施工现场的实时监测数据,能够科学、准确地预测桥梁挂篮安全风险的概率变化,为桥梁挂篮施工安全风险管理提供参考。

关键词: 桥梁挂篮；动态贝叶斯网络(DBN)；模糊集理论；监测数据；实时更新；安全风险分析

Funding

基金项目：安徽省自然科学基金杰出青年基金资助项目（2208085J20）；安徽省交通控股集团有限公司科技资助项目（JKKJ-2021-01）

合肥工业大学学报(自然科学版)

Navigation