| 混凝土中钢筋直径雷达检测的神经网络方法 |
| |
| 引用本文： | 叶爱文,;谢慧才.混凝土中钢筋直径雷达检测的神经网络方法[J].长安大学学报(建筑与环境科学版),2008(4):105-110. |
| |
| 作者姓名： | 叶爱文 ;谢慧才 |
| |
| 作者单位： | [1]深圳大学土木工程学院,广东深圳518060; [2]深圳大学深圳市土木工程耐久性重点实验室,广东深圳518060 |
| |
| 基金项目： | 广东省自然科学基金项目（032023） |
| |
| 摘 要： | 介绍了神经网络的基本理论，提出了一种利用神经网络处理探地雷达信号对钢筋混凝土中钢筋直径进行测算的新方法；采用BP和Elman两种神经网络模型，并利用小波包分析提取了雷达波的特征参数；针对神经网络应用时实际训练数据不足的问题，提出利用多项式拟合的方法来补充训练数据。试验结果表明：该方法对钢筋直径的测算结果是令人满意的。
|
| 关 键 词： | 探地雷达 钢筋直径检测 神经网络 小波包分析 |
Neural Network Method of Diameter Detection of Rebar in Concrete by Using GPR |
| |
| Affiliation: | YE Ai-wen, XIE Hui-cai (1. School of Civil Engineering, Shenzhen University, Shenzhen 518060, Guangdong, China; 2. Shenzhen Municipal Key Lab on Durability of Civil Engineering, Shenzhen University, Shenzhen 518060, Guangdong, China) |
| |
| Abstract: | The basic theory of neural network was introduced, and a new method using neural network to estimate the diameter of rebar in concrete by processing ground penetrating radar （GPR） signal was presented. The models of back propagating （BP） and Elman neural network were employed, and feature parameters were extracted using wavelet packet. Authors detailed alternative collection. synthetic data generation by The results show that the meth ing polynomial, which could reduce the volume of data od to estimate diameter of rebar is satisfactory. |
| |
| Keywords: | GPR diameter detection of rebar neural network wavelet packet analysis |
| 本文献已被 维普 等数据库收录！ |
|
