| 基于PSO-SVM的城市桥梁群体震害预测模型研究 |
| |
| 引用本文： | 王二涛,高惠瑛,孙海,王俊杰.基于PSO-SVM的城市桥梁群体震害预测模型研究[J].震灾防御技术,2017,12(1):185-193. |
| |
| 作者姓名： | 王二涛 高惠瑛 孙海 王俊杰 |
| |
| 作者单位： | 中国海洋大学工程学院土木工程系, 山东青岛 266100 |
| |
| 摘 要： | 本文根据城市桥梁群体的实际震害资料数据,采用粒子群算法（PSO）来优化支持向量机（SVM）参数,选择影响桥梁震害等级的8个因素作为特征输入向量,充分用2种算法的优点建立PSO-SVM的桥梁震害预测模型。通过比较PSO-SVM和SVM模型对桥梁震害的预测能力,发现PSO-SVM模型具有较高预测精度和较高的推广价值。本文的研究成果对桥梁震害等级的预测具有一定的参考价值和指导意义。
|
| 关 键 词： | 粒子群-支持向量机 支持向量机 桥梁 震害预测 |
| 收稿时间： | 2016/4/25 0:00:00 |
Study on Seismic Damage Prediction Model of Urban Bridges Group Based on PSO-SVM |
| |
| Wang Ertao,Gao Huiying,Sun Hai and Wang Junjie.Study on Seismic Damage Prediction Model of Urban Bridges Group Based on PSO-SVM[J].Technology for Earthquake Disaster Prevention,2017,12(1):185-193. |
| |
| Authors: | Wang Ertao Gao Huiying Sun Hai and Wang Junjie |
| |
| Institution: | Department of Civil Engineering, School of Engineering, Ocean University of China, Qingdao 266100, China |
| |
| Abstract: | According to the observed urban bridge damage data, the particle swarm optimization (PSO) was used to optimize the input parameters of support vector machine (SVM) model. Eight factors that effect bridge seismic damage level are chosen as the input vector. By making full use of the advantages of PSO and SVM, we establish the PSO-SVM model. By comparing the urban bridge damage prediction ability of the SVM model and PSO-SVM model, we conclude that the PSO-SVM model has relatively high accuracy and strong generalization capability, which is of important reference and guide value. Key worlds:PSO-SVM; SVM; Bridge; Seismic damage prediction |
| |
| Keywords: | PSO-SVM SVM Bridge Seismic damage prediction |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《震灾防御技术》浏览原始摘要信息 |
| 点击此处可从《震灾防御技术》下载免费的PDF全文 |
|
