This paper discusses how techniques coming from Computational Intelligence (CI) have contributed substantially to the understanding and solution of problems in structural health monitoring (SHM) The paper begins by giving some background to SHM into a number of eras, with the most recent being the Computational Intelligence Era The paper then outlines three case studies regarding novelty detection for three different systems and illustrates alternative machine learning and statistics tools in terms of progressive complexity

Aspects of computational intelligence in structural dynamics: Structural health monitoring

This paper describes a new software package for various structural health monitoring (SHM) applications. The software is a set of standardized MATLAB routines covering three main stages of SHM: data acquisition, feature extraction, and feature classification for damage identification. A subset of the software in SHMTools is embeddable, which consists of Matlab functions that can be cross-compiled into generic 'C' programs to be run on a target hardware. The software is also designed to accommodate multiple sensing modalities, including piezoelectric active-sensing, which has been widely used in SHM practice. The software package, including standardized datasets, are publicly available for use by the SHM community. The details of this embeddable software will be discussed, along with several example processes that can be used for guidelines for future use of the software.

Tajana S Rosin

Papers

SHMTools: a general-purpose software tool for SHM applications