The growing number of existing infrastructure with decreased or indeterminate reliability, and on the other hand, constant design of lighter, albeit more productive structures facilitate the adoption of automated Structural Health Monitoring (SHM) identification tools capable of unprejudiced diagnosis of in-service structures. In comparison with customarily exploited simulation-based approaches, databased schemes or hybrid analysis (data/finite element model) can often introduce a more objective perspective on the behavior of operating structures, and as a result can enable long-term, automated and even online assessment.

Recent trends for energy conservation have placed the focus on Wind Turbine (WT) structures, which represent systems characterized with complex dynamics and alternating operating nature. We propose a diagnostic framework able to describe the variability of such monitored systems. The novel approach combines the Smoothness Priors Time Varying Autoregressive Moving Average (SP-TARMA) method for modeling the non-stationary structural response, and a Polynomial Chaos Expansion (PCE) probabilistic model for modeling the propagation of response uncertainty.

The computational tool is applied on long-term data, collected from an active sensing system installed for four years on a real operating WT structure located in Dortmund, Germany. The long-term tracking of the obtained PCESPTARMA Diagnostic Indicator (DI) is further assessed by means of a statistical analysis. The results demonstrate that the proposed treatment yields a DI sensitive to unfamiliar fluctuations in measured Environmental and Operational Parameters (EOP).