[1]冯微,张兆津,邵海鹏.一种基于深度置信网络的车撞桥墩损伤等级判别方法[J].长安大学学报(自然科学版),2018,38(06):135-145.
　FENG Wei,ZHANG Zhao jin,SHAO Hai peng.A method for discriminating damage level in vehiclebridge piercollision based on deep belief network[J].Journal of Chang’an University (Natural Science Edition),2018,38(06):135-145.
点击复制

一种基于深度置信网络的车撞桥墩损伤等级判别方法()
分享到：

长安大学学报（自然科学版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
第38卷
期数:
2018年06期
页码:
135-145
栏目:
桥梁与隧道工程
出版日期:
2018-12-01

文章信息/Info

Title:
A method for discriminating damage level in vehiclebridge piercollision based on deep belief network
作者:
冯微; 张兆津; 邵海鹏
(1. 长安大学 公路学院，陕西 西安 710064； 2. 武威职业学院 电子信息工程系，甘肃 武威 733000)
Author(s):
FENG Wei1;  ZHANG Zhaojin2;  SHAO Haipeng1
(1. School of Highway, Changan University, Xian 710064, Shaanxi, China； 2. Department of ElectronicInformation Engineering, Wuwei Occupational College, Wuwei 733000, Gansu, China)
关键词:
桥梁工程; 桥墩安全; 结构损伤等级; 车辆碰撞; 深度学习
Keywords:
bridge engineering;  bridge pier safety;  structural damage level;  vehicle collision;  deep learning
文献标志码:
A
摘要:
为了解决车撞桥墩过程中桥墩受损等级不易被实时监测和提前预警的问题，在深度学习理论框架下，对车撞桥墩损伤等级判别方法进行研究。在探讨车撞桥墩过程及其影响因素的基础上，引入场论，分析碰撞过程动力学和能量转换关系，并构建桥墩结构损伤场势计算公式；借鉴风险评估概念，对不同场势值所属的损伤等级进行定义，建立车撞桥墩损伤等级判别模型。利用深度置信网络能够准确提取数据本质特征和Softmax分类器可有效判别多等级概率的优势，构建基于深度置信网络的车撞桥墩损伤等级判别模型。将已有研究中相关试验和仿真数据输入构建的网络模型中，对网络结构中参数进行调试，并将该模型判别结果与传统机器学习判别方法结果进行对比验证。研究结果表明：针对该模型设计的3种网络结构，网络深度为7层时，模型判别结果的精度最高，平均绝对误差和平均相对误差分别为0.37和14.8%；相比人工神经网络模型，提出的模型采用深层神经网络结构，判别结果更为稳定，平均绝对误差和平均相对误差分别降低了1.34和11.5%；相比随机森林模型，提出的模型可自动提取数据的特征属性，所得判别等级结果的偏差更低，不超过0.5级，且其平均绝对误差和平均相对误差分别降低了0.46和4.7%，在多损伤等级判别问题中效率和准确度更高。
Abstract:
In order to solve the problem that the damage level of bridge pier was not easy to be monitored in real time and early warning, under the deep learning theory framework, the method of discrimination the damage level in the bridge pier was studied. The field theory was used to analyze the dynamics and energy conversion relationship during the collision of a car with a bridge pier, and a formula to calculate the damage field potential of the bridge pier was derived. The concept of risk assessment was used to define the damage levels based on different damage field potential values, and the damage level discrimination model for the vehiclebridge pier collision was established. The DBN can accurately extract the essential characteristics of the data and the Softmax classifier can effectively discriminate the multilevel probability. Hence, these two were combined to establish the damage level discrimination model for vehiclebridge pier collision. The relevant experimental and simulation data from existing research were input to the network model formulated, the parameters of the network structure were debugged, and then the results were compared with those obtained from the traditional discrimination methods based on machine learning. The results show that among the three network structures designed in this study, the deep belief network (DBN) model has the highest accuracy when the network depth is seven layers. The mean absolute error and the mean relative error are 0.37 and 14.8%, respectively. Compared to the artificial neural network model, the DBN model adopts the deep neural network structure, the discrimination level result is more stable, and the mean absolute error and the mean relative error are reduced by 1.34 and 11.5%, respectively. Compared to the random forest model, the DBN model can automatically extract the characteristic attributes of the data, the deviation of the obtained discrimination level result is lower, not exceeding 0.5 level, and the mean absolute error and the mean relative error are reduced by 0.46 and 4.7%, respectively. Hence, the DBN model is more efficient and accurate than the random forest model for the multidamage level discrimination problem. 6 tabs, 8 figs, 36 refs.

相似文献/References:

[1]李宇,朱晞,杨庆山,等.高墩大跨桥梁结构的脆弱性分析[J].长安大学学报(自然科学版),2012,32(01):0.
[2]高亮,刘健新,张丹,等.桁架桥主梁三分力系数试验[J].长安大学学报(自然科学版),2012,32(01):0.
[3]刘旭政,王丰平,黄平明,等.斜拉桥各构件校验系数的常值范围[J].长安大学学报(自然科学版),2012,32(01):0.
[4]尚维波,张春宁.高墩刚构桥系梁抗震分析[J].长安大学学报(自然科学版),2012,32(01):0.
[5]邬晓光,李冀弘,宋伟伟.基于改进响应面法的在役PC桥梁承载力可靠性[J].长安大学学报(自然科学版),2012,32(03):53.
　WU Xiao-guang,LI Ji-hong,SONG Wei-wei.Reliability of existing PC bridge based on improved response surface method[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):53.
[6]石雄伟,袁卓亚,马毓泉,等.钢板-混凝土组合加固预应力混凝土箱梁[J].长安大学学报(自然科学版),2012,32(03):58.
　SHI Xiong-wei,YUAN Zhuo-ya,MA Yu-quan,et al.Prestressed concrete box girder strengthened with comsposition of steel plate and concrete[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):58.
[7]李传习,陶 伟,董创文.斜交墩截面刚度与弯曲正应力[J].长安大学学报(自然科学版),2012,32(03):63.
　LI Chuan-xi,TAO Wei,DONG Chuang-wen.Sectional stiffness and bending normal stress of oblique pier[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):63.
[8]邓继华,邵旭东.带铰平面梁元几何非线性有限元分析[J].长安大学学报(自然科学版),2012,32(03):68.
　DENG Ji-hua,SHAO Xu-dong.Geometric nonlinear finite element analysis of plane beam element with hinge[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):68.
[9]蒲广宁,赵 煜,宋一凡.减梁增肋法加固空心板桥的力学性能[J].长安大学学报(自然科学版),2012,32(06):38.
　PU Guang-ning,ZHAO Yu,SONG Yi-fan.Mechanical properties of strengthening hollow slab bridge based on beam-reduction and rib-addition method[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):38.
[10]党 栋,贺拴海,周勇军,等.基于车辆统计数据的汽车荷载标准值取值与评估[J].长安大学学报(自然科学版),2012,32(06):44.
　DANG Dong,HE Shuan-hai,ZHOU Yong-jun,et al.Choosing and assessment for the standard of vehicle load based on vehicle statistical data[J].Journal of Chang’an University (Natural Science Edition),2012,32(06):44.

更新日期/Last Update: 2018-12-18