摘要
随着交通领域的迅速发展,保障桥梁安全、降低维护费用成为普遍关注的问题。以振动分析为基础、广泛借助信息技术的理论和方法,进行桥梁状态监测成为当前的研究热点。考虑到的桥梁损伤样本很难获得、个体差异使得损伤数据难以共享的实际问题,本文将桥梁状态监测归结为异常监测问题,引入基于SVM的一类学习算法从长期监测数据中获取正常状态的模式,实现异常状态的精确报警。文中采用香港汀九桥400小时实测数据,验证了这种方法的实际效果。
As the long span suspension and cable-stayed bridges are widely used in the world, the highway department has become increasingly interested in developing the bridge condition monitoring system to enhance the bridge security and reduce the cost for maintenance. The current researches are mainly focused on vibration analysis based monitoring technologies, associated with wide use of information theories and means. Due to the fact that the signal of the damage case is difficult to get, the bridge condition monitoring is regarded as a 'one class learning' pattern recognition problem in this paper. To performing exact alarm when abnormal circs occurs, it is proposed to model bridge normality from its long-term data distribution by SVM based novelty detection algorithm. 400 hours real world data from Hong Kong Ting Kau Bridge are used to examine the method.
出处
《公路交通科技》
CAS
CSCD
北大核心
2004年第1期67-70,共4页
Journal of Highway and Transportation Research and Development
关键词
桥梁
状态监测
一类学习
支持矢量机
Bridge
Condition monitoring
One class learning
Support vector machine