基于二级神经网络集成的桥梁健康状态的融合评估方法

Fusion assessment method for bridge healthy state based on integrated two-level neural network

实现对桥梁的监测并对桥梁健康状态进行评估可以有效提高桥梁的安全性。但在对桥梁进行监测时,参数众多,数据繁杂,难以通过简单传统的方法对桥梁健康状态进行准确评估,为了融合不同类型监测参数和同一类型参数的多点异步数据,获得对桥梁的健康状态的一致性评估,提出了一种针对桥梁健康状态评估的基于二级神经网络集成的融合评估方法,以降低监测数据的多源融合过程的复杂度,提高桥梁健康评估的准确性。

The security of the bridge could be effectively improved by monitoring and evaluation on its healthy status. Since there are too many parameters in the bridge＇s monitoring, which are complex and difficult,it is actually hard to obtain a relatively accurate assessment of the bridge health status using the simple and traditional method. To integrate different types of parameters and asynchronous data, and get conformity assessment of the bridge health status, we proposed an assessment method of bridge health status, based on two-level neural network integration. Our method was verified to reduce the complexity of multisource monitoring data fusion process and improve the accuracy of the bridge health assessment by the simulation calculations.