Big Data Analytics in Online Structural Health Monitoring

Guowei Cai,Sankaran Mahadevan

doi:10.36001/ijphm.2016.v7i4.2462

Abstract

This manuscript explores the application of big data analytics in online structural health monitoring. As smart sensor technology is making progress and low cost online monitoring is increasingly possible, large quantities of highly heterogeneous data can be acquired during the monitoring, thus exceeding the capacity of traditional data analytics techniques. This paper investigates big data techniques to handle the highvolume data obtained in structural health monitoring. In particular, we investigate the analysis of infrared thermal images for structural damage diagnosis. We explore the MapReduce technique to parallelize the data analytics and efficiently handle the high volume, high velocity and high variety of information. In our study, MapReduce is implemented with the Spark platform, and image processing functions such as uniform filter and Sobel filter are wrapped in the mappers. The methodology is illustrated with concrete slabs, using actual experimental data with induced damage

Highlights

During the span of a structures service life, conditions such as wear, overload, environmental degradation, and natural disasters may accelerate the degradation of the material and the structure
The score increases by almost 40% (i.e., 100% × (0.723 − 0.523)/0.523), as the sampling interval decreases from 2 mins to 1 second. This indicates that by increasing the sample rate, the damage detection performance can be greatly improved. This increases the demand on the data analytics computation, which is resolved by the MapReduce technique
This paper developed a framework for applying a big data technique to online structural health monitoring

Summary

Introduction

During the span of a structures service life, conditions such as wear, overload, environmental degradation, and natural disasters may accelerate the degradation of the material and the structure. SHM techniques can be either data-driven or model-based In both cases, the data is often obtained using non-destructive evaluation (NDE) techniques, which can be divided into active and passive techniques. The purpose of online structural health monitoring is to detect the damage in the structure, analyze future risk, predict the remaining useful life, and guide the maintenance/repair actions if needed. In the context of pattern recognition (Farrar, Doebling, W., & Nix, 2001), a four-step procedure is described: (1) Operational evaluation, (2) Data acquisition and cleansing, (3) Feature selection, and (4) Statistical model development. Only deterministic structural health monitoring is considered in the context of big data, so step (4) will develop a deterministic model which will be used only to detect the damage, without considering any uncertainty sources

Methods

Findings

Discussion

Conclusion