基于故障树-模糊贝叶斯网络的装配式建筑施工质量风险分析

Quality Risk Analysis of Prefabricated Building Construction Basedon Fault Tree Analysis-Fuzzy Bayesian Network

装配式建筑施工质量直接影响装配式建筑的安全性和耐久性,以提升装配式建筑的质量管理水平为目标,针对如何有效地控制施工质量风险,提出了基于故障树-模糊贝叶斯网络方法的风险评价模型。结合装配式建筑的特点进行风险源识别、分类,采用故障树对风险因素的因果关系进行建模,将其映射成贝叶斯网络,引入模糊集理论和改进的相似聚合方法,将专家意见转化为节点事件的先验概率和条件概率,构建了基于故障树-模糊贝叶斯网络的装配式建筑质量风险分析模型。结合案例进行分析:将收集的数据导入GeNIe软件中,进行参数学习得到了各风险等级发生概率;反向推理获得后验概率确定了关键性因素,进行致因分析得到了最大可能致因链,最后,利用贝叶斯网络的快速预测功能进行快速预测与推断。

The quality of prefabricated building construction directly affects the safety and durability of prefabricated construction,with the goal of improving the quality management level of prefabricated construction,a risk evaluation model based on the fault tree-fuzzy Bayesian network method was proposed for how to effectively control the construction quality risk.Risk sources were identified and classified by combining the characteristics of prefabricated construction,the causal relationship of risk factors was modeled by using fault tree,which was mapped into a Bayesian network,fuzzy set theory and the improved similar aggregation method were introduced,and the expert opinion was transformed into the a priori probability and conditional probability of node events,and the quality risk analysis model of prefabricated construction based on fault tree-fuzzy Bayesian network was constructed.Combined with the case study:the collected data are imported into GeNIe software,parameter learning is performed to obtain the occurrence probability of each risk level.The posteriori probability is obtained by reverse inference to determine the critical factors,and the causal analysis is performed to obtain the maximum probable causal chain,and finally,the fast prediction and inference are performed by utilizing the fast prediction function of Bayesian network.