Sensor optimization using a genetic algorithm for structural health monitoring in harsh environments

Abstract

Structural health monitoring (SHM) has become a powerful tool for engineering fields to make decisions for resource allocation in harsh environments such as fire, earthquake, and flood. To effectively make a decision based on the monitoring data, the SHM system requires a large number of sensors for different data resource measurements, for example, strain, temperature, and vibration, resulting in a need to determine the tradeoff between the number of sensors of each type and the associated cost of the system. This paper introduces a sensor optimization approach based on genetic algorithm for the multiple objective sensor placement of structural health monitoring in harsh environments. The derived theoretic multi-task objective function of the genetic function is validated by a single-bay steel frame in a harsh environment of simultaneous high temperature and large strain. The variance between the theoretical and the experimental analysis was within 5 %, indicating an effective sensor placement optimization using the developed genetic algorithm, which can be further applied to general sensor optimization for SHM system applications in harsh environments.