摘要
本文以CRTS Ⅰ型板式无砟轨道为研究对象,采用ABAQUS建立CA砂浆充填层完好及各脱空损伤工况下的板式无砟轨道有限元模型,通过模态分析发现,相比于频率,振型对CA砂浆充填层脱空损伤更敏感。以参数化处理的振型作为BP神经网络的输入对各损伤工况下板式轨道CA充填层损伤位置进行识别发现,在两种识别模式(单损伤和双损伤)下,本文提出的方法可以准确地识别出所有工况中CA充填层的损伤位置。本文提出的基于BP神经网络的CRTS I型板式无砟轨道CA充填层损伤位置识别的方法是可行的,有望为无砟轨道的安全诊断提供有效的技术支撑。
This research was focused on the CRTS I slab track. A finite element model was established using the software ABAQUS to analyze the modal parameters of the track structure under the intact and void damaged conditions of the CA mortar layer. The results showed that the mode shape was more sensitive to the damage compared with the frequeney. Taking the normalized mode shape as the input parameter of the BP neural network, the damage location of the CA mortar layer of the slab track under different damage conditions could be identified. The results indicated that all damage locations of the CA mortar layer were identified accurately by the proposed method under two identification patterns （single-damage identification and double-damage identification）. The proposed methodology to identify the damage location of the CA mortar layer of the CRTS I slab track based on BP neural network is feasible and it is expected to provide effective technical support for the safety diagnosis of the slab track.
作者
胡琴
徐巍
高飞
朱宏平
HU Qin;XU Wei;GAO Fei;ZHU Hong-ping(School of Civil Engineering and Mechanics,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《土木工程与管理学报》
北大核心
2018年第5期87-93,共7页
Journal of Civil Engineering and Management
基金
国家重点研发计划(2016YFC0802002)
国家自然科学基金(51708242
51578260
51629801)
中央高校基本科研基金(2017KFYXJJ137)
关键词
损伤识别
模态分析
BP神经网络
CA砂浆
无砟轨道
damage identification
modal analysis
BP neural network
CA mortar layer
slab track
