• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文OA文献 >Nondestructive Pavement Evaluation Using ILLI-PAVE Based Artificial Neural Network Models
【2h】

Nondestructive Pavement Evaluation Using ILLI-PAVE Based Artificial Neural Network Models

机译：基于ILLI-paVE的人工神经网络模型无损检测评价

页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The overall objective in this research project is to develop advanced pavementstructural analysis models for more accurate solutions with fast computation schemes. Softcomputing and modeling approaches, specifically the Artificial Neural Network (ANN) andGenetic Algorithm (GA) techniques, have been implemented to develop forward andbackcalculation type pavement analysis models based on the validated nonlinear ILLI-PAVEfinite element solutions of the most commonly found/constructed flexible pavements in theState of Illinois. The developed pavement evaluation toolbox can be used for rapidly andmore accurately backcalculating field or in-service pavement layer properties andthicknesses; predicting critical stress, strain, and deformation responses of these in-servicepavements in real time from the measured Falling Weight Deflectometer (FWD) deflectiondata; and incorporating these predicted critical pavement responses, such as tensile strainfor asphalt fatigue, directly into the Illinois Department of Transportation???s (IDOT???s)mechanistic pavement analysis and design with emphasis on extended life asphaltpavement design concepts. The outcome of the project???s successful research efforts nowprovides IDOT with a field validated nondestructive pavement evaluation professional ANN(ANN-Pro) software package to assess pavement condition through FWD backcalculationand eventually help assess pavement rehabilitation strategies. In addition, a secondsoftware package also developed in the project provides the framework SOFTSYS, SoftComputing Based Pavement and Geomaterial System Analyzer, which estimates full-depthasphalt pavement thickness when there is no thickness data available for the pavementsection where FWD testing is performed.
机译：该研究项目的总体目标是开发先进的路面结构分析模型，以便通过快速计算方案获得更准确的解决方案。已实施软计算和建模方法，特别是人工神经网络（ANN）和遗传算法（GA）技术，以基于经过验证的最常见/构造的柔性路面非线性ILLI-PAVE有限元解决方案来开发前后计算类型的路面分析模型在伊利诺伊州。开发的路面评估工具箱可用于快速，更准确地反算野外或在役路面层的性质和厚度；根据测得的落锤挠度计（FWD）挠度数据实时实时预测这些在用路面的临界应力，应变和变形响应；并将这些预测的关键路面响应，例如沥青疲劳的拉伸应变，直接结合到伊利诺伊州交通部（IDOT）的机械路面分析和设计中，重点是延长使用寿命的沥青路面设计概念。该项目的成功研究成果为IDOT提供了经过现场验证的无损路面评估专业ANN（ANN-Pro）软件包，可通过FWD反算来评估路面状况，并最终帮助评估路面修复策略。此外，该项目中还开发了第二个软件包，该软件包提供了SOFTSYS框架，基于SoftComputing的路面和土工材料系统分析器，当进行FWD测试的路面断面没有可用的厚度数据时，它可以估算全深度沥青路面的厚度。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利

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