Experimental evaluation of environmental effects on a polymer-coated aluminium structure: a time-series analysis and pattern recognition approach

Temperature variation is an important issue that needs to be considered when trying to develop a reliable Structural Health Monitoring (SHM) strategy. In the case that a data-based approach is chosen for damage detection, environmental fluctuations could be erroneously regarded as an abnormal condition of the structure and could mask the presence of damage. One of the objectives of the current work is to examine a statistical pattern recognition approach for novelty detection under different temperature conditions. A second important issue that could hinder the reliability of a SHM strategy is any kind of nonlinear behaviour, not associated with damage, in a system. For the purposes of this paper, the dynamic behaviour of a polymer-coated aluminium structure with ribs fixed with bolts is examined. The autoregressive parameters are the damage sensitive features and later, it is performed Principal Component Analysis (PCA) for robust novelty detection that takes into account the temperature variation.