Impact force localization for civil infrastructure using augmented Kalman Filter optimization

Muhammad M. Saleem, Hongki Jo

Research output: Contribution to journalArticle

Abstract

Impact forces induced by external object collisions can cause serious damages to civil engineering structures. While accurate and prompt identification of such impact forces is a critical task in structural health monitoring, it is not readily feasible for civil structures because the force measurement is extremely challenging and the force location is unpredictable for full-scale field structures. This study proposes a novel approach for identification of impact force including its location and time history using a small number of multi-metric observations. The method combines an augmented Kalman filter (AKF) and Genetic algorithm for accurate identification of impact force. The location of impact force is statistically determined in the way to minimize the AKF response estimate error at measured locations and then time history of the impact force is accurately constructed by optimizing the error co-variances of AKF using Genetic algorithm. The efficacy of proposed approach is numerically demonstrated using a truss and a plate model considering the presence of modelling error and measurement noises.

Original languageEnglish (US)
Pages (from-to)123-139
Number of pages17
JournalSmart Structures and Systems
Volume23
Issue number2
DOIs
StatePublished - Feb 1 2019

Fingerprint

Kalman filters
Genetic algorithms
Force measurement
Structural health monitoring
Civil engineering

Keywords

  • Accelerometers
  • Augmented Kalman filter
  • Genetic algorithm
  • Impact force
  • Strain gauges

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Impact force localization for civil infrastructure using augmented Kalman Filter optimization. / Saleem, Muhammad M.; Jo, Hongki.

In: Smart Structures and Systems, Vol. 23, No. 2, 01.02.2019, p. 123-139.

Research output: Contribution to journalArticle

@article{87cca65427844bc086f6a1f80a5ba3db,
title = "Impact force localization for civil infrastructure using augmented Kalman Filter optimization",
abstract = "Impact forces induced by external object collisions can cause serious damages to civil engineering structures. While accurate and prompt identification of such impact forces is a critical task in structural health monitoring, it is not readily feasible for civil structures because the force measurement is extremely challenging and the force location is unpredictable for full-scale field structures. This study proposes a novel approach for identification of impact force including its location and time history using a small number of multi-metric observations. The method combines an augmented Kalman filter (AKF) and Genetic algorithm for accurate identification of impact force. The location of impact force is statistically determined in the way to minimize the AKF response estimate error at measured locations and then time history of the impact force is accurately constructed by optimizing the error co-variances of AKF using Genetic algorithm. The efficacy of proposed approach is numerically demonstrated using a truss and a plate model considering the presence of modelling error and measurement noises.",
keywords = "Accelerometers, Augmented Kalman filter, Genetic algorithm, Impact force, Strain gauges",
author = "Saleem, {Muhammad M.} and Hongki Jo",
year = "2019",
month = "2",
day = "1",
doi = "10.12989/sss.2019.23.2.123",
language = "English (US)",
volume = "23",
pages = "123--139",
journal = "Smart Structures and Systems",
issn = "1738-1584",
publisher = "Techno Press",
number = "2",

}

TY - JOUR

T1 - Impact force localization for civil infrastructure using augmented Kalman Filter optimization

AU - Saleem, Muhammad M.

AU - Jo, Hongki

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Impact forces induced by external object collisions can cause serious damages to civil engineering structures. While accurate and prompt identification of such impact forces is a critical task in structural health monitoring, it is not readily feasible for civil structures because the force measurement is extremely challenging and the force location is unpredictable for full-scale field structures. This study proposes a novel approach for identification of impact force including its location and time history using a small number of multi-metric observations. The method combines an augmented Kalman filter (AKF) and Genetic algorithm for accurate identification of impact force. The location of impact force is statistically determined in the way to minimize the AKF response estimate error at measured locations and then time history of the impact force is accurately constructed by optimizing the error co-variances of AKF using Genetic algorithm. The efficacy of proposed approach is numerically demonstrated using a truss and a plate model considering the presence of modelling error and measurement noises.

AB - Impact forces induced by external object collisions can cause serious damages to civil engineering structures. While accurate and prompt identification of such impact forces is a critical task in structural health monitoring, it is not readily feasible for civil structures because the force measurement is extremely challenging and the force location is unpredictable for full-scale field structures. This study proposes a novel approach for identification of impact force including its location and time history using a small number of multi-metric observations. The method combines an augmented Kalman filter (AKF) and Genetic algorithm for accurate identification of impact force. The location of impact force is statistically determined in the way to minimize the AKF response estimate error at measured locations and then time history of the impact force is accurately constructed by optimizing the error co-variances of AKF using Genetic algorithm. The efficacy of proposed approach is numerically demonstrated using a truss and a plate model considering the presence of modelling error and measurement noises.

KW - Accelerometers

KW - Augmented Kalman filter

KW - Genetic algorithm

KW - Impact force

KW - Strain gauges

UR - http://www.scopus.com/inward/record.url?scp=85063446137&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063446137&partnerID=8YFLogxK

U2 - 10.12989/sss.2019.23.2.123

DO - 10.12989/sss.2019.23.2.123

M3 - Article

VL - 23

SP - 123

EP - 139

JO - Smart Structures and Systems

JF - Smart Structures and Systems

SN - 1738-1584

IS - 2

ER -