File(s) not publicly available

A novel neural network based method for analysis of pavement deflection data

conference contribution
posted on 29.04.2020, 00:00 by J Lee, Muhammad Zohaib JanMuhammad Zohaib Jan, Brijesh VermaBrijesh Verma
Efficient management of road infrastructure involves planning, construction, maintenance, operation and disposal of road assets. Knowledge of current conditions and deterioration of road pavements are essential to enable effective road asset management. This paper presents a novel neural network based method for the analysis and prediction of Falling Weight Deflectometer (FWD) parameters based on Traffic Speed Deflectometer (TSD) parameters. A neural network based method was designed and applied to analyse the correlation between FWD and TSD data. The method used a feed-forward neural network that was trained with TSD data as an input and FWD data as an output. The proposed method was evaluated on TSD and FWD data provided by the industry partner Australian Road Research Board (ARRB). The prediction results are very promising and within an acceptable range set by the industry partner. A detailed results analysis is included in this paper.

Funding

Category 3 - Industry and Other Research Income

History

Start Page

2506

End Page

2513

Number of Pages

8

Start Date

06/12/2019

Finish Date

09/12/2019

ISBN-13

9781728124858

Location

Xiamen, China

Publisher

IEEE

Place of Publication

Piscataway, NJ

Peer Reviewed

Yes

Open Access

No

External Author Affiliations

Australian Road Research Board

Author Research Institute

Centre for Intelligent Systems

Era Eligible

Yes

Name of Conference

IEEE Symposium Series on Computational Intelligence (SSCI 2019)