• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International PhD Symposium in Civil Engineering vol.2; 20040616-19; Delft(NL) >Using an Artificial Neural Network for interpreting ultrasonic pulse velocity of concrete
【24h】

Using an Artificial Neural Network for interpreting ultrasonic pulse velocity of concrete

机译：用人工神经网络解释混凝土的超声脉冲速度

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This work is based on the idea that ultrasonic pulse velocity (UPV) tests are a useful way to determine the quality of concrete and that the use of this Nondestructive Testing (NDT) technique might supply the basis for the verification of the condition state of a concrete element. In fact, it is largely accepted that it is feasible to associate certain ultrasonic results with material characteristics such as density, homogeneity or even strength. However, the interpretation of ultrasonic test data can become quite complex, since various factors might affect the results. Therefore, the diagnosis is usually based on the sensitivity of experts. Nonetheless, when dealing with real-life structures, the amount of data to be analyzed can be very large, creating difficulties for the expert to consider it. The working hypothesis of this research is that the process of analyzing ultrasonic test data for determining concrete condition can be facilitated and standardized using Artificial Neural Networks (ANN). This Artificial Intelligence (AI) tool is capable of processing a large amount of unstructured data, generating a non-linear correlation model that can provide very good results. The net is created using knowledge extracted from specialists or information available from previous cases where the relationship between input data and outcome is known. By using the ANN it is possible to search for the most efficient solutions without prior knowledge of the relationships between these input variables and outcomes. The research aims to collect data from various types of concretes and use ANN to establish models that correlate concrete properties and ultrasonic readings. The models will be tested to determine their accuracy and sensitivity to the topology of the net. A special model to estimate compressive strength from ultrasonic readings and concrete characteristics is also being tested. The preliminary results are very encouraging, since they give indication that the models are robust and more accurate that traditional regression models. The major aim of the work is to test, explore and demonstrate the potential of ANN as an interesting tool for diagnosis, training and storage of non-structured knowledge in the civil engineering field. The working hypothesis of this research is that the process of analyzing ultrasonic test data for determining concrete condition can be facilitated and standardized using Artificial Neural Networks (ANN).
机译：这项工作基于这样的思想，即超声脉冲速度（UPV）测试是确定混凝土质量的一种有用方法，并且这种无损检测（NDT）技术的使用可能为验证混凝土的状态提供基础。具体元素。实际上，人们普遍认为将某些超声结果与材料特性（例如密度，均质性甚至强度）相关联是可行的。但是，超声测试数据的解释可能会变得非常复杂，因为各种因素都可能影响结果。因此，诊断通常基于专家的敏感性。但是，在处理现实生活中的结构时，要分析的数据量可能非常大，给专家考虑带来了困难。这项研究的工作假设是，可以使用人工神经网络（ANN）来促进和标准化分析超声测试数据以确定具体条件的过程。该人工智能（AI）工具能够处理大量非结构化数据，生成可以提供非常好的结果的非线性相关模型。网络是使用从专家那里提取的知识或从以前输入数据与结果之间的关系已知的情况下可获得的信息创建的。通过使用ANN，无需事先知道这些输入变量和结果之间的关系，就可以搜索最有效的解决方案。该研究旨在从各种类型的混凝土中收集数据，并使用人工神经网络来建立与混凝土性能和超声读数相关的模型。将对模型进行测试，以确定其对网络拓扑的准确性和敏感性。还正在测试一种特殊的模型，该模型可以根据超声波读数和混凝土特性估算抗压强度。初步结果令人鼓舞，因为它们表明模型比传统回归模型更健壮和更准确。这项工作的主要目的是测试，探索和证明ANN作为土木工程领域诊断，训练和存储非结构化知识的有趣工具的潜力。这项研究的工作假设是，可以使用人工神经网络（ANN）来促进和标准化分析超声测试数据以确定具体条件的过程。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号