MS17.15 - Structural Health Monitoring

The current need for sustained renewable energy is pushing the boundaries of the size and availability of wind turbines. As a result, there is an urge to develop a well-functioning Structural Health Monitoring (SHM) systems to keep these systems online. This need is also shared with other pieces of critical infrastructure and structures in general. Among different issues, the effect Environmental and Operational Variability (EOV) in Damage Sensitivity Features (DSF) (and in the dynamics of the structure in general) is probably still one of the most relevant challenges facing SHM. While numerous methods have been developed to tackle this issue, these may be designated either as implicit or explicit. The first class of methods use characteristics of the features themselves, while in the latter, information from environmental and operational conditions is also introduced. In paper, explicit methods bear the advantage, as these directly correlate the information of the variables affecting the DSFs, and may potentially lead to increased damage diagnosis sensitivity. On the other hand, implicit methods do not make assumptions on cause-effect relations and will not introduce a bias from wrongly-assumed correlations. In this work, we postulate that the combined application of both types of methods can lead to a more robust approach that will also lessen their individual shortcomings. More precisely, we combine Principal Component Analysis (PCA) with Bayesian non-linear regression for EOV compensation in damage detection of a wind turbine blades. We evaluate the methodology on data from a laboratory-scale wind turbine blade in controlled temperature and humidity conditions, as presented in [1]. On each vibration test, the temperature was set to a fixed value within the range [-20◦C,+20◦C]. Furthermore, the humidity was increased for some of the samples by spraying water onto the blade. Tests were performed on a reference blade, and two blades with different damage severity. DSFs originate from Vector Auto-Regressive (VAR) models, and include the VAR model coefficients and the respective natural frequency estimates. The raw DSFs are initially PCA-transformed, and, following the PCA method for EOV compensation, PCs with largest eigenvalues would be removed, as these bear most of the EOV influence. In our approach, we use Bayesian non-linear regression on all the PCs and use the F-statistic to assess which PCs accept the regression model and which not. The PCs rejecting the regression model would be left as they are, while the remainder are adjusted with the help of the obtained regression model. We compare the performance of the combined method, with that obtained using only PCA, and using the regression model only, in terms of Receiver Operating Characteristic (ROC) analysis. Results point at the combined method providing a more robust performance among the considered variants. [1] A. Gómez-González and S.D. Fassois, “A supervised vibration-based statistical methodology for damage detection under varying environmental conditions & its laboratory assessment with a scale wind turbine blade”, Journal of Sound and Vibration, 366, pp. 484-500, 2016.

The challenge that is at the forefront of data-driven vibration-based structural health monitoring (VSHM) is the detrimental effect caused by environmental and operational variations (EOVs). Therefore, action must be taken in order to mitigate the effects of the EOVs without affecting the influence of damage. A number of regression-based approaches have been applied in VSHM, using measured environmental and operational parameters to model damage sensitive features. However, action is infrequently taken to address the problem of the inherent variabilities in the data, as well as the effect of biases in the creation of the regression models. This work aims to combat these issues with a demonstration on a multivariate nonlinear regression model, using data taken from an operational wind turbine blade. A number of metrics were used to assess the stability of the regression coefficient matrices. This included assessing the effect that was observed in the distribution of the outlier analysis, distribution of main group and number of outliers, as well as the variance of the covariance of the input variables. These metrics were tested by using a number of different sets of observations for training the regression models. The results of the analysis showed that by addressing the aforementioned issues, the damage detection actually decreased. However, the parasitic nuances cannot be left untreated. Ultimately, whilst the damage detection does decrease, the robustness of the system increases and so too does the confidence that can be given to each observation.

Ferry docks together with bridges are part of the critical infrastructure that makes it possible to establish connections between two destination points separated by a large body of water. For coastal nations with the majority of the population living by the sea, ferries and bridges are an indispensable part of almost any travel. One of such nations is Norway with its famous landscapes that often present a serious challenge when it comes to planning, building, and maintaining infrastructure. A potential failure or closure of one ferry dock can lead to substantial economic losses and massive delays, as one has to often drive far inland along the Norwegian fjords, or put a strain on smaller, surrounding road systems that were designed to accommodate traffic expected on the national highway. Therefore improved understanding of dynamic behavior of ferry docks and loads they are exposed can provide economical benefits in form of better fatigue design or more cost-effective maintenance. Norwegian standard N400 instructs that the following loads induced by the ferry should be consider in the design of the ferry dock bridges in Norway: • Impact load • Longitudinal pressure • Forced deformations The two latter ones should be considered to occur simultaneously with the traffic loads, while the impact load appears separately during berthing operation (vessel sway motion) according to N400. Longitudinal pressure is defined as the force along the bridge deck center line induced by the vessel pushing towards or away from the land (vessel sway and yaw motions), while forced deformations correspond to a rotation of the front of the ferry deck bridge that is in contact with a ferry. Those rotations arise from the fact that a ferry is not able to keep itself perfectly aligned with the horizontal plane as it is affected by wind and waves when attached to the ferry dock bridge, and it will transfer its own tilt to a ferry deck bridge (vessel roll motion). The ability to determine above mentioned loads under real-life conditions together with traffic loads is crucial for a proper estimation of fatigue service life, tracking deterioration process and finally, as a result of that, a cost-effective planning of inspections and maintenance schedule. If ferry-induced and traffic loads can be estimated accurately from measurements, then long-term monitoring will provide valuable data that can be used to validate the design approach and recommendations in N400. In this study we try to examine whether the impact load can be identified by means of kalman filter using measured acceleration time series and an updated finite element model.