Abstract

The stress in structural steel members is the most useful and directly measurable physical quantity to evaluate the structural safety in structural health monitoring, which is also an important index to evaluate the stress distribution and force condition of structures during structural construction and service phases. Thus, it is common to set stress as a measure in steel structural monitoring. Considering the economy and the importance of the structural members, there are only a limited number of sensors that can be placed, which means that it is impossible to obtain the stresses of all members directly using sensors. This study aims to develop a stress response prediction method for locations where there are insufficent sensors, using measurements from a limited number of sensors and pattern recognition. The detailed improved aspects are: (1) a distributed computing process is proposed, where the same pattern is recognized by several subsets of measurements; and (2) the pattern recognition using the subset of measurements is carried out by considering the optimal number of sensors and number of fusion patterns. The validity and feasibility of the proposed method are verified using two examples: the finite-element simulation of a single-layer shell-like steel structure, and the structural health monitoring of the space steel roof of Shenzhen Bay Stadium; for the latter, the anti-noise performance of this method is verified by the stress measurements from a real-world project.

Highlights

  • Structural stress distribution is a critical and directly-measurable parameter to evaluate structural safety [1] in which monitoring and prediction of the stress distribution of real-world structures is key to the application of structural health monitoring

  • For such stress prediction methods, the critical problem is that the number of the sensors for stress monitoring is limited, there is some research on the optimal placement of sensors to control the number of sensors in a reasonable range and ensure that the sensors provide the greatest efficiency

  • Stress prediction for the distributed structural health monitoring using existing measurements and pattern recognition is proposed in this paper

Read more

Summary

Introduction

Structural stress distribution is a critical and directly-measurable parameter to evaluate structural safety [1] in which monitoring and prediction of the stress distribution of real-world structures is key to the application of structural health monitoring. The maximum stress of beam structures was measured during monitoring, and it was compared with the allowable stress and used to assess structural safety [3] For such stress prediction methods, the critical problem is that the number of the sensors for stress monitoring is limited, there is some research on the optimal placement of sensors to control the number of sensors in a reasonable range and ensure that the sensors provide the greatest efficiency. This is because the installation of sensor systems is expensive.

Objectives
Findings
Methods
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call