摘要
针对福建省西溪特大桥应变长期监测数据,采用相空间重构与支持向量机(Support Vector Machine,SVM)技术对其进行建模处理并预测其未来发展趋势。试验结果显示:采用SVM方法对西溪特大桥2016年7月的应变量日均值进行预测,平均预测误差为4.3με。预测结果对于SVM预测技术在工程中的应用具有一定参考价值,可为桥梁结构的安全预警提供可靠的科学依据。
Based on the long-term strain monitoring data of Xixi Bridge in Fujian Province,in this paper,phase space reconstruction and support vector machine(SVM) are used to model and predict the future development trend of Xixi Bridge. The results show: Using SVM to forecast daily average value of strain for Xixi super large bridge in July,2016,the average forecasting error is 4. 3 με. The forecast results have certain reference value for the application of SVM forecasting technology in projects,it can provide reliable scientific basis for safe advance warning of bridge structure.
出处
《公路交通技术》
2016年第6期80-82,共3页
Technology of Highway and Transport
基金
交通运输部应用基础研究资助项目(2014319740160)
重庆市科技人才培养计划资助项目(cstc2013kjrc-qnrc30001)
关键词
西溪特大桥
支持向量机
相空间重构
预测
Xixi super-large bridge
support vector machine
phase space reconstruction
forecast