| 基于曲率模态和BP神经网络的简支梁的损伤识别 |
| |
| 引用本文： | 李冬,曾春平,马琨,吴光敏.基于曲率模态和BP神经网络的简支梁的损伤识别[J].贵州大学学报(自然科学版),2013(6):110-113. |
| |
| 作者姓名： | 李冬 曾春平 马琨 吴光敏 |
| |
| 作者单位： | 昆明理工大学理学院,云南昆明650500 |
| |
| 基金项目： | 国家自然科学基金（KKGA201207006）;云南省高校结构健康诊断重点实验室（平台建设KKKP201207003）;云南省中青年学术和技术带头人后备人才培养项目（2008PY013） |
| |
| 摘 要： | 通过BP神经网络对简支梁的损伤位置和损伤程度进行了研究。文中首先采用有限元仿真软件ANSYS计算得到不同损伤情况下结构的前两阶固有频率并计算指定点的曲率模态值，并以此为输入参数建立用于识别简支梁损伤的BP网络，最后利用LM算法训练网络来进行损伤检测。结果表明：该方法能有效地对损伤位置及损伤程度进行识别，且对损伤程度进行识别的精确度较高。
|
| 关 键 词： | BP神经网络 曲率模态 简支梁 ANSYS |
Damage Identification Based on Curvature Mode and BP Neural Network for Simply Supported Beam |
| |
| LI Dong,ZENG Chun-Ping,MA Kun,WU Guang-min.Damage Identification Based on Curvature Mode and BP Neural Network for Simply Supported Beam[J].Journal of Guizhou University(Natural Science),2013(6):110-113. |
| |
| Authors: | LI Dong ZENG Chun-Ping MA Kun WU Guang-min |
| |
| Institution: | (Faculty of Science, Kunming University of Science and Technology, Kunming 650500, China) |
| |
| Abstract: | BP neural network was used to study the damage position and degree for simply supported beam. Firstly, the finite element simulation software was used to calculate the first two natural frequencies of the structures with different damage and the curvature mode of the specially appointed points ; then they were made as the input of a BP network for simply supported beam＇s damage identification ; finally, the trained network was used to detect the damage. The results show that： this method can effectively identify the damage position and degree; there will be a high accuracy for the identification of damage degree. |
| |
| Keywords: | BP neural network curvature mode simply supported beam ANSYS |
| 本文献已被 维普 等数据库收录！ |
