Title

A Machine Learning-Based Algorithm For Processing Massive Data Collected From The Mechanical Components Of Movable Bridges

Keywords

Damage detection; Machine-learning; Maintenance monitoring; Massive SHM data; Movable bridge; Robust regression analysis

Abstract

This paper presents a machine learning algorithm for processing of massive data collected from the mechanical components of movable bridges. The proposed approach consists of training and monitoring phases. The training phase was focused on the extracting statistical features and conducting cross correlation analysis (CCA) and robust regression analysis (RRA). The monitoring phase included tracking of errors associated with the derived models. The main goal was to analyze the efficiency of the developed system for health monitoring of the bridge mechanical components such as gearbox, motor and rack and pinion. To this aim, Sunrise Movable Bridge in Ft. Lauderdale, Florida was selected and instrumented. A comprehensive database was collected from the sensors installed on the mechanical and structural components of the Sunrise Bridge for about 4 years. The collected data were utilized to assess the performance of the algorithm under baseline and different common damage scenarios. Based on the results, the proposed health monitoring system has a satisfactory performance for the detection of the damage scenarios caused by leakage and lack of sufficient oil in gearbox, as well as bolt removal from rack and pinion. The introduced algorithm can be regarded as a valuable tool for the management and interpretation of the massive (big) data collected for structural health monitoring (SHM) of movable bridges.

Publication Date

12-1-2016

Publication Title

Automation in Construction

Volume

72

Number of Pages

269-278

Document Type

Article

Personal Identifier

scopus

DOI Link

https://doi.org/10.1016/j.autcon.2016.02.008

Socpus ID

85003972483 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85003972483

This document is currently not available here.

Share

COinS