基于时间序列分析的桥梁结构安全评价研究

发布时间：2018-01-29 02:17

本文关键词： 桥梁结构结构安全评价时间序列分析灰色时序组合预测模型 ARMA控制图　出处：《重庆交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：桥梁结构在长期运营中，由于受到各种不确定因素的影响，结构超限、超期服役、疲劳、腐蚀等危害性事件，必将导致桥梁结构出现严重安全隐患，并可能由此带来巨大的生命财产损失。这就要求安全评估系统能及时发现损伤并给出损伤预警提示。然而，作为桥梁健康监测系统核心的安全评价，依然采用基于精确的结构模型的安全方法。事实上，结构在长期服役中其结构参数不断发生变化，为此建立精确结构模型的安全评价方法不能取得令人满意的结果。鉴于桥梁健康监测系统采集了海量包含反应结构损伤特征信息的时序数据，可从中挖掘出反应结构系统性能的演变规律。这种基于时序分析的方法既能够克服了建立精确结构模型的困难，又具有通用性，但它仍处于研究初期，对其进行深入研究极具现实意义。基于此，本文从时间序列分析的角度出发，对桥梁结构安全评价方法进行了探索和研究。为此，本文主要从以下几个方面展开研究： (1)研究桥梁结构安全评价的重要性、归纳总结现有桥梁结构安全评价方法，并分析其中的不足。突出基于时序分析法的重要性。 (2)将时序分析法、灰色预测模型引入到桥梁监测数据分析中，提出基于灰色时序组合模型（ARMA-GM）算法的数据处理方法在桥梁监测数据中的应用；并以实际桥梁监测数据为例，通过有效预测桥梁结构未来状态趋势，证明了ARMA-GM模型在桥梁未来状态趋势预测中的可行性。 (3)提出一种基于ARMA控制图的桥梁结构安全评价方法，该方法考虑了桥梁结构的实际情况，将控制图的控制限与桥梁结构可靠度指标相结合，设置了初级预警限和安全告警限，通过观测统计量在二级控制限内外的分布情况来判断桥梁结构的安全情况。这使得控制图的控制限与桥梁结构的安全设计规范紧密相连。 (4)以重庆马桑溪长江大桥监测数据为实例，针对本文提出的灰色时序组合预测模型和基于ARMA控制图的安全评价方法进行了实际应用。结果表明：时序模型和灰色模型能够对桥梁健康监测系统中的监测数据进行有效处理，对桥梁结构状态参数预报精度较高，在预测步长相同情况下，本文提出的灰色时序组合模型的预报精度比单一时序模型、灰色模型更高，能够对桥梁结构趋势进行中长期预报，提前给出预警提示；基于ARMA控制图的安全评价方法，所确定的安全评价指标可有效识别结构损伤，克服了虚发报警和漏发报警的缺陷。
[Abstract]:In the long-term operation of bridge structure, due to the influence of various uncertain factors, the structure overrun, excessive service, fatigue, corrosion and other harmful events, will inevitably lead to serious hidden dangers of bridge structure. And it may bring huge loss of life and property. This requires the safety assessment system to detect the damage and give the warning of damage in time. However, as the core of the bridge health monitoring system, the safety evaluation. The security method based on precise structural model is still adopted. In fact, the structural parameters of the structure are constantly changing during the long service. Therefore, the safety evaluation method based on accurate structural model can not obtain satisfactory results. In view of the bridge health monitoring system to collect a large number of response structure damage information containing time series data. This method based on time series analysis can not only overcome the difficulty of establishing accurate structure model, but also be universal, but it is still in the early stage of study. It is of great practical significance to study it in depth. Based on this, this paper explores and studies the method of bridge structure safety evaluation from the point of view of time series analysis. 1) the importance of safety assessment of bridge structure is studied, the existing methods of safety assessment of bridge structure are summarized, and the shortcomings are analyzed, and the importance of time series analysis is highlighted. 2) introducing the time series analysis method and the grey prediction model into the bridge monitoring data analysis. The application of data processing method based on grey time series combination model (ARMA-GM) algorithm in bridge monitoring data is presented. Taking the actual bridge monitoring data as an example, the feasibility of the ARMA-GM model in predicting the future state trend of the bridge structure is proved by effectively predicting the future state trend of the bridge structure. (3) A method of bridge structure safety evaluation based on ARMA control chart is proposed. The control limit of the control chart is combined with the reliability index of the bridge structure, taking into account the actual situation of the bridge structure. The primary warning limit and the safety warning limit are set. The safety situation of bridge structure is judged by observing the distribution of statistics inside and outside the two-stage control limit, which makes the control limit of control chart closely related to the safety design specification of bridge structure. Taking the monitoring data of Mausangxi Yangtze River Bridge in Chongqing as an example, the grey time series combination prediction model and the safety evaluation method based on ARMA control chart are applied in this paper. The results show that the time series model and the grey model can effectively deal with the monitoring data in the bridge health monitoring system, and the prediction accuracy of the bridge structure state parameters is higher, and the prediction step is the same. The prediction accuracy of the grey time series combination model proposed in this paper is higher than that of the single time series model. The grey model can forecast the bridge structure trend in the medium and long term and give the warning warning in advance. Based on the safety evaluation method of ARMA control chart, the safety evaluation index can effectively identify the structural damage and overcome the defects of false alarm and false alarm.
【学位授予单位】：重庆交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U441;U447

【参考文献】