Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Structural Monitoring System Based on Sensitivity Analysis and a Neural Network

Structural Monitoring System Based on Sensitivity Analysis and a Neural Network A design procedure for a structure's monitoring system using sensitivity analysis and a neural network is developed. The monitoring system is to be used to monitor damage to members critically affecting the overall safety of structures. Recently, many techniques for evaluating the damage of isolated members and models of simple structures have been investigated. However, actual structures are large, and their complex behavior may not be based on damage of isolated members or their simple models. To monitor large structures realistically, structures' data on behavior at many points need to be monitored. Identifying the optimal locations and numbers of these monitoring points and assessing the safety of the entire structure from the limited data are the monitoring system design problem. The procedure presented for this design problem is a two‐step process. In the first step, using sensitivity analysis and damage‐assessment techniques, individual members are ranked according to their influence on the failure probability of the entire structure or according to their effect on the abnormal behavior of the structure. Based on the rank, critical members are identified. In the second step, sensitivity analysis and a neural network are used to determine the optimal locations. In addition, the optimal number of sensors for monitoring damage to the critical members selected in step 1 is also suggested. Truss and frame examples are used to show the validity and applicability of the monitoring system design procedure. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Computer-Aided Civil and Infrastructure Engineering Wiley

Structural Monitoring System Based on Sensitivity Analysis and a Neural Network

Loading next page...
 
/lp/wiley/structural-monitoring-system-based-on-sensitivity-analysis-and-a-vaPIBx1fJ6

References (19)

Publisher
Wiley
Copyright
Blackwell Publishers, Inc.
ISSN
1093-9687
eISSN
1467-8667
DOI
10.1111/0885-9507.00194
Publisher site
See Article on Publisher Site

Abstract

A design procedure for a structure's monitoring system using sensitivity analysis and a neural network is developed. The monitoring system is to be used to monitor damage to members critically affecting the overall safety of structures. Recently, many techniques for evaluating the damage of isolated members and models of simple structures have been investigated. However, actual structures are large, and their complex behavior may not be based on damage of isolated members or their simple models. To monitor large structures realistically, structures' data on behavior at many points need to be monitored. Identifying the optimal locations and numbers of these monitoring points and assessing the safety of the entire structure from the limited data are the monitoring system design problem. The procedure presented for this design problem is a two‐step process. In the first step, using sensitivity analysis and damage‐assessment techniques, individual members are ranked according to their influence on the failure probability of the entire structure or according to their effect on the abnormal behavior of the structure. Based on the rank, critical members are identified. In the second step, sensitivity analysis and a neural network are used to determine the optimal locations. In addition, the optimal number of sensors for monitoring damage to the critical members selected in step 1 is also suggested. Truss and frame examples are used to show the validity and applicability of the monitoring system design procedure.

Journal

Computer-Aided Civil and Infrastructure EngineeringWiley

Published: Jul 1, 2000

There are no references for this article.