钢筋混凝土主梁无监督式安全性评估方法

Unsupervised safety evaluation method of reinforced concrete main girder structure

为了解决人为主观因素和指标权重不确定性对桥梁安全性评估的影响,基于模糊聚类理论,提出了一种适用于钢筋混凝土主梁结构的无监督式安全状态评估方法。首先,构建了钢筋混凝土主梁结构安全性评估指标体系,以典型桥梁的检测数据为聚类样本,基于F统计量计算,确定了聚类样本的最佳分类。其次,将同一类桥梁检测指标均值作为该类别的中心,通过待评估桥梁检测指标数据与类别中心的模糊贴近度计算,实现该桥梁的安全性评价。以长春赛德大桥为实体工程,针对评价指标对结构安全性的影响差异,通过考虑和不考虑指标权重对桥梁安全性评估进行了分析。结果表明,基于模糊聚类的钢筋混凝土主梁结构安全评估方法是有效性。研究方法对客观评价钢筋混凝土主梁安全性及研究钢筋混凝土桥梁安全性评估方法具有借鉴意义。

Aiming at the influence of subjective factors and uncertainty of index weight on bridge safety assessment, an unsupervised safety assessment method for RC bridge superstructure based on fuzzy clustering is proposed in this paper. Firstly, the index system of safety assessment of bridge superstructure is constructed, and the monitoring data of bridge are selected as clustering samples. Based on the principle of fuzzy clustering, the optimal classification of samples is determined. Secondly, average value of the same kind of bridge monitoring index is taken as the approximate center. Bridge safety is evaluated by calculating the fuzzy closeness between the unknown bridge monitoring data and corresponding approximate center. The Changchun Saide Bridge is selected as practical project, bridge safety assessment is investigated by the influence of index weight considering the impact difference of evaluation index on structure safety, and effectiveness of the proposed method in this study is verified, which provides reference for the girder safety of reinforced concrete bridge, and for safety assessment method of the reinforced concrete bridge study.