原文传递
ARTIFICIAL NEURAL NETWORK MODELING FOR EVALUATING ROUGHNESS PARAMETERS BASED ON PAVEMENT DISTRESS VALUES
| 题名： | ARTIFICIAL NEURAL NETWORK MODELING FOR EVALUATING ROUGHNESS PARAMETERS BASED ON PAVEMENT DISTRESS VALUES |
| 正文语种： | 英文 |
| 作者： | RUMI SUTRADHAR |
| 关键词： | representation；international；patching；collection；parameter；correlation；measur；rough；stress；integra |
| 摘要： | Pavement roughness is a characteristic of pavement unevenness which can be measured by various pavement measuring devices like MERLIN (Machine for Evaluating Roughness using Low-Cost Instrumentation), Bump Integrator etc. This roughness value may be expressed in terms of IRI (International Roughness Index), an international parameter used to measure pavement roughness conditions. But the data collection tasks of measuring IRI for enormous road networks consume substantial cost and time. In this paper, an attempt has been taken to evaluate roughness value in terms of the IRI and BI (Bump Integrator), which may be treated as a critical representation index of pavement performance. The analysis is done using artificial neural network methodology, and a model is developed. The model, is based on different pavement distresses (Alligators, Potholes, Segregation, Edge cracking, Corrugation and Patching) found in local low volume roads in Tripura, India. Using this model roughness values can be easily evaluated analytically without going to field. The model is verified comparing the field values and the predicted values of IRI & BI and coefficient of correlation (R2) are 0.8943 and 0.9414 respectively. Moreover it is found that out of all the distresses, patching works has more effect (25.46％) on road roughness in low volume roads of Tripura. |
| 出版年： | 2015 |
| 论文唯一标识： | I-75Y2015V43N05002 |
| 英文栏目名称： | Technical Papers |
| 期刊名称： | Indian Highways |
| 拼音刊名(出版物代码)： | I-75 |
| 卷： | 43 |
| 期： | 05 |
| 页码： | 11-14 |
相关文献
- Calibration and Application of Treatment Performance Models in a Pavement Management System in Tennessee
- Framework for Development and Comprehensive Comparison of Empirical Pavement Performance Models
- Development of Rehabilitation Strategies Based on Structural Capacity for Composite and Flexible Pavements
- Artificial Neural Network–Based Predictive Tool for Modeling of Self-Centering Endplate Connections with SMA Bolts
- Improved Method of Determination of Basic Wind Speed with Terrain Effects Using Graph Neural Network
- Prediction of pedestrian dynamics in complex architectures with artificial neural networks
检索历史
应用推荐
