Rui Teixeira

 Inspections and maintenance of infrastructure are expensive. In some cases, overdue or insufficient maintenance/monitoring can lead to an unacceptable risk of collapse and to a tragic failure as the Morandi bridge in Genoa, Italy, on 14th August 2018. An accurate assessment of the safety of a structure is a difficult task due to uncertainties associated with the aging and response of the structure, with the operational and environmental loads, and with their interaction. During the period from 2015 to 2019, the project TRUSS (Training in Reducing Uncertainty in Structural Safety) ITN (Innovative Training Network), funded by the EU H2020 Marie Curie-Skłodowska Action (MSCA) programme, has worked towards improving the structural assessment of buildings, energy, marine, and transport infrastructure. Fourteen Early Stage Researchers (ESRs) have been recruited to carry out related research on new materials, testing methods, improved and more efficient modelling methods and management strategies, and sensor and algorithm development for Structural Health Monitoring (SHM) purposes. This research has been enhanced by an advanced program of scientific and professional training delivered via a collaboration between 6 Universities, 1 research institute and 11 companies from 5 European countries. The high proportion of companies participating in TRUSS ITN has ensured significant industrial expertise and has introduced a diverse range of perspectives to the consortium on the activities necessary to do business in the structural safety sector. -> Link to full text in repository

Analysis of Offshore Wind Turbine (OWT) fatigue damage is an intense, resource demanding task. While the current methodologies to design OWT to fatigue are quite limited in the way and amount of uncertainty they can account for, they still represent a relevant share of the total effort needed in the OWT design process. The robustness achieved in the design process is usually limited. To enable OWT to be more robust, an innovative methodology that tackles current limitations using a balanced amount of designing effort was developed. It consists of generating a short-term fatigue damage (DSH ) using a Kriging surrogate model that accurately accounts for uncertainty using an adaptive approach. The current paper discusses the application of a reinterpolation convergence to build a Kriging surrogate model that replicates DSH in OWT tower components. Different variables involved in the convergence are discussed. The discussion extends then to how the design could be improved by using different convergence scenarios for the Kriging surface. Cross-validation is used to train and validate the surrogate surface. The main goal is to give the designer a rationale on the trade-off between computational time and accuracy using the mentioned approach to design robust OWT towers. Results show that on a design basis two levels of approach may be efficient. In the first, if a very high computational cost is expected, a trade-off between accuracy and computational time must be considered and then, if the intention is to check how robust the current design is, a full convergence of the surface should be pursued.

The present work researches on the definition of the load spectra used for offshore wind turbine low SN slope materials’ fatigue design. Uncertainty in the sample sized used to scale fatigue life is analyzed for the tower component. Damage density is investigated for different environmental conditions in order to understand the importance of the different regimes of operation. Damage density is identified to be a heterogeneous function of the loading environmental conditions. In some cases, even for low SN slope materials, most of the damage occurs due to high load ranges. To study on the influence of this heterogeneity, different tail fits are used to compare the influence of accurately defining the tail region on a reference design time (?). Results show that OWT fatigue is highly dependent on the ? shorter that ? time used to approximate ?. This is mainly related to the fact that fatigue design depends not only on scaling stress ranges, but also cycle counts. Effort on the design phase should be applied in the definition of the uncertainty of the load spectra due to the limitation imposed by using low sample sizes to cover the extensive joint distribution of environmental parameters.

The current paper discusses the applicability of Gaussian process regressions, also known as Kriging models, in the context of structural and reliability analysis. Due to their flexibility these models appear in the field of structural analysis in many forms. Applications to approximate limit state functions, replace the computational expensive codes that solves the dynamic of complex systems, or replicate stochastic fields can be identified. Due to this fact, a discussion on the different parameters that depend on the implementation procedure chose to use these model is presented in the current paper. Design of experiments, polynomial approximation, correlation function, hyperparameters convergence and estimation function are the main global variables analysed. When implementing a Gaussian regression or Kriging model, the user is faced with the choice of these before any further progress. The discussion presented complements previous works on the implementation of such models in the sense that it focus on the structural analysis application and on how these parameters influence the accuracy. It is shown that depending on the approximation, significant advantage can be taken from understanding these major variables. Different examples are presented to support the understanding of the problem and the main conclusions on the applicability of the Gaussian regression models as surrogates for structural analysis are drawn.

For complex systems, the applicability of surrogate models has shown the potential to enable accurate assessments using a reduced batch of data and to compile information about large datasets. These behave as black-box functions that replace a series of inputs/outputs. In the present work, a Kriging surrogate is used to predict confidence intervals in an offshore wind turbine tower fatigue design. Uncertainty in fatigue due to loading is highly connected to the mean. One year operational fatigue results is used to validate the results. The Kriging is applied to replicate the yearly states of operation, and successfully predicts intervals of confidence for the long-term fatigue design. Regarding the interest of data analysis, the approach implemented is characterized by its flexibility and capability of approaching any problem that can be characterized by a single variable. Being therefore an interesting tool in decision schemes where large datasets are available or prediction of unknown outputs is required.

There is multitude of models available to assess structural safety based on a set of input parameters. As the degree of complexity of the models increases, the uncertainty of their output tends to decrease. However, more complex models typically require more input parameters, which may contain a higher degree of uncertainty. Therefore, it becomes necessary to find the balance that, for a particular scenario, will reduce the overall uncertainty (model + parameters) in structural safety. The latter is the objective of the Marie Skłodowska-Curie Innovative Training Network titled TRUSS (Training in Reducing Uncertainty in Structural Safety) funded by the EU Horizon 2020 research and innovation programme (http://trussitn.eu). This paper describes how TRUSS addresses uncertainty in: (a) structural reliability of materials such as basalt fiber reinforced polymer, (b) testing techniques in the assessment of concrete strength in buildings, (c) numerical methods in computing the non-linear response of submerged nuclear components subjected to an earthquake, (d) estimation of life of wind turbines, (e) the optimal inspection times and management strategies for ships, (f) characterization of the dynamic response of ship unloaders and (g) the relationship between vehicles fuel consumption and pavement condition.-> Link to full text in repository

The fatigue design of Offshore Wind Turbines (OWT) is one of the most resource demanding tasks in the OWT design process. Techniques have been developed recently to simplify the amount of effort needed to design to structural fatigue. This is the example of the usage of Kriging surrogate models. These may be used in OWTs design not only, to reduce the computational effort needed to analyse an OWT, but also to allow their design to be robust. Due to the stress variability and its non-linear character, the short-term fatigue damage variability is high, and converging the stochastic field approached by the surrogate model in relation to the real observations is challenging. A thorough analysis of the different components that load an OWT and are more critical for the tower component fatigue life is required, and therefore, presented and discussed in the current paper. The tower, jointly with the foundation, are particular components of the OWT regarding the fatigue analysis process. Statistical assessments of the extrapolation of fatigue loads for the tower and the influence of the environmental parameters in the short-term damage are presented in this paper. This sets a support analysis for the creation of the Kriging response surfaces for fatigue analysis. NREL’s 5MW monopile turbine is used due to its state of the art character. Five environmental variables are considered in the analysis. A sensitivity analysis is conducted to identify which variables are most prominent in the quantification of the short-term damage uncertainty in the tower. The decoupling of the different external contributions for the fatigue life is a major contribution of the work presented. Preliminary guidelines are drawn for the creation of surrogate models to analyse fatigue of OWT towers and the most relevant conclusions are presented in an industry-oriented design outline regarding the most critical random variables that influence OWT short-term fatigue calculation. -> Link to full text in repository

The probabilistic analysis of Offshore Wind Turbines (OWT) is not a new practice. The standards for designing OWT (IEC 61400 class) emphasizes that assessing uncertainty is of major importance inside the design chain. Still, major challenges related to the uncertainty and the probabilistic assessment pose to the sector and its development. The analysis of operational loads is one them. The problem of analyzing extreme responses or cumulated damage in operation during the design phase is significantly related to its high computational cost. As we progressively add complexity to the system to account for its uncertainties, the computational effort increases and a perceptive design becomes a heavy task. If an optimization process is then sought, the designing effort grows even further. In the particular case of fatigue analysis, it is frequent to not be able to cover a full lifetime of simulations due to computational cost restrictions. The mentioned difficulties fomented the utilization of surrogate models in the reliability analysis of OWT. From these surrogate approximations the ones based on Kriging models gained a special emphasis recently for structural reliability. It was shown that, for several applications, these models can be efficient and accurate to approximate the response of the system or the limit state surfaces. The presented paper tackles some of the issues related to their applicability to OWT, in a case specific scenario of the tower component subjected to operational fatigue loads. A methodology to assess the reliability of the tower component to fatigue damage is presented. This methodology combines a Kriging model with the theory of extreme values. A one-dimensional Kriging case using the state of art NREL’s monopile turbine is presented. The reliability of the OWT tower is calculated for 20 years. The results show that the usage of a Kriging model to calculate the long term damage variation shows a high potential to assess the reliability of OWT towers to fatigue failure. -> Link to full text in repository

Offshore wind energy experienced an exponential growth in installed power since the beginning of the current century. While this growing trend is expected to continue, further growth of the sector imposes more demanding engineering methods. It is then envisaged that enhanced technical competitiveness can be achieved through a progressively less deterministic design process. Under the described context, a comparative study on the applicability of different probabilistic methods to estimate the probability of failure (Pf) of offshore wind turbine (OWT) towers under extreme events is presented here. Depending on the complexity introduced in the analysis of the OWT towers the applicability of different probabilistic approaches may be limited. FORM, SORM, Monte Carlo Simulation are examples of well-established methodologies to estimate Pf. Nevertheless, alternative methodologies such as the directional simulation can be an even more efficient solution for the problem. This preliminary assessment of the probabilistic approaches enables further developments in reliability methodologies for the specific case of OWT towers. -> Link to full text in repository