ABSTRACT
Vibration-based monitoring approaches have in recent years gained in popularity within the context of monitoring and life cycle assessment of engineered systems. Dynamic response data comprise a rich information content, which may further be exploited for inferring structural properties as well as their evolution under varying operational conditions. To this end, it is often useful to maintain access to the full response signals, rather than condensed information in the form of modal properties. This task is oftentimes prohibitive for the case of Wireless Sensor Networks (WSNs), a solution which increasingly gains in popularity for large scale infrastructure, due to its low cost and ease of deployment. In this work, validation of a spectro-temporal compressive sensing approach introduced in earlier work of the authors is carried out using both a synthetic data-set of a 4-storey shear frame building, as well as experimental data obtained from an actual bridge deployment. The spectro-temporal information present in the signal is used for recovery purposes. Next, the recovered signals are processed to extract damage sensitive features.
The demonstrated in the work examples indicate the ability of the proposed framework in providing reliable signal reconstructions for the purpose of structural identification. The quality of extracted modal shapes is attested via comparison with the reference counterpart, extracted using the complete time series. In both the numerical example and the experimental validation high, i.e. above 0.88 MAC values are obtained. The results reveal the promise of the proposed spector-temporal compressive sensing framework, relying on re-weighted Basis Pursuit De-Noising (Becker 2011), for enhancement of the quality of information extracted from partially transmitted data. In the demonstrated examples a transmission percentage (rate) as low as 36% of the complete time series is employed. Modal shape extracted using via NeXT-ERA full time series (top) and recovered time series (bottom). https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/fig31_1.tif"/>
The advantages delivered by this approach are highly promising for reducing transmission costs, and therefore extending the sustainability of the nodes of WNS deployments, effectively rendering these a valid and likely superior alternative to tethered technologies.
