Publications
2021 |
Meijers, P C; Tsouvalas, A; Metrikine, A V Magnetomechanical response of a steel monopile during impact pile driving Journal Article Engineering Structures, 240 , pp. 112340, 2021. @article{Meijers2021a, title = {Magnetomechanical response of a steel monopile during impact pile driving}, author = {P C Meijers and A Tsouvalas and A V Metrikine}, doi = {10.1016/j.engstruct.2021.112340}, year = {2021}, date = {2021-01-01}, journal = {Engineering Structures}, volume = {240}, pages = {112340}, abstract = {This paper reports on a measurement campaign in which the magnetomechanical response of a steel monopile is recorded during installation with a hydraulic impact hammer. By comparing impact-induced changes in the magnetic stray field of the structure to the measured strain, this effect is analysed for the first time under dynamic loading conditions on such a large scale. It is shown that the measured stray field displays an excellent correspondence with the strain in terms of frequency content and amplitude ratio for hammer blows that induce compressive strain pulses of different magnitude. Using the data, a non-contact method is developed and validated to infer the hammer-induced strains using the dynamic magnetic stray field. The proposed method can be applied during pile installations when the use of conventional strain measurement devices is challenging, e.g. in the offshore environment.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This paper reports on a measurement campaign in which the magnetomechanical response of a steel monopile is recorded during installation with a hydraulic impact hammer. By comparing impact-induced changes in the magnetic stray field of the structure to the measured strain, this effect is analysed for the first time under dynamic loading conditions on such a large scale. It is shown that the measured stray field displays an excellent correspondence with the strain in terms of frequency content and amplitude ratio for hammer blows that induce compressive strain pulses of different magnitude. Using the data, a non-contact method is developed and validated to infer the hammer-induced strains using the dynamic magnetic stray field. The proposed method can be applied during pile installations when the use of conventional strain measurement devices is challenging, e.g. in the offshore environment. |
Meijers, P C Non-collocated methods to infer deformation in steel structures PhD Thesis Delft University of Technology, 2021, ISBN: 978-94-6384-217-4. @phdthesis{Meijers2021b, title = {Non-collocated methods to infer deformation in steel structures}, author = {P C Meijers}, doi = {10.4233/uuid:8d278e8f-0972-4d96-8acf-1dcd1cd0e358}, isbn = {978-94-6384-217-4}, year = {2021}, date = {2021-01-01}, address = {Delft, the Netherlands}, school = {Delft University of Technology}, keywords = {}, pubstate = {published}, tppubtype = {phdthesis} } |
Quaeghebeur, E; Bos, R; Zaaijer, M B Wind farm layout optimization using pseudo-gradients Journal Article Wind Energy Science, 6 (3), pp. 815–839, 2021. @article{wes-6-815-2021, title = {Wind farm layout optimization using pseudo-gradients}, author = {E Quaeghebeur and R Bos and M B Zaaijer}, url = {https://wes.copernicus.org/articles/6/815/2021/}, doi = {10.5194/wes-6-815-2021}, year = {2021}, date = {2021-01-01}, journal = {Wind Energy Science}, volume = {6}, number = {3}, pages = {815--839}, abstract = {This paper presents a heuristic building block for wind farm layout optimization algorithms. For each pair of wake-interacting turbines, a vector is defined. Its magnitude is proportional to the wind speed deficit of the waked turbine due to the waking turbine. Its direction is chosen from the inter-turbine, downwind, or crosswind directions. These vectors can be combined for all waking or waked turbines and averaged over the wind resource to obtain a vector, a “pseudo-gradient”, that can take the role of gradient in classical gradient-following optimization algorithms. A proof-of-concept optimization algorithm demonstrates how such vectors can be used for computationally efficient wind farm layout optimization. Results for various sites, both idealized and realistic, illustrate the types of layout generated by the proof-of-concept algorithm. These results provide a basis for a discussion of the heuristic's strong points – speed, competitive reduction in wake losses, and flexibility – and weak points – partial blindness to the objective and dependence on the starting layout. The computational speed of pseudo-gradient-based optimization is an enabler for analyses that would otherwise be computationally impractical. Pseudo-gradient-based optimization has already been used by industry in the design of large-scale (offshore) wind farms.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This paper presents a heuristic building block for wind farm layout optimization algorithms. For each pair of wake-interacting turbines, a vector is defined. Its magnitude is proportional to the wind speed deficit of the waked turbine due to the waking turbine. Its direction is chosen from the inter-turbine, downwind, or crosswind directions. These vectors can be combined for all waking or waked turbines and averaged over the wind resource to obtain a vector, a “pseudo-gradient”, that can take the role of gradient in classical gradient-following optimization algorithms. A proof-of-concept optimization algorithm demonstrates how such vectors can be used for computationally efficient wind farm layout optimization. Results for various sites, both idealized and realistic, illustrate the types of layout generated by the proof-of-concept algorithm. These results provide a basis for a discussion of the heuristic's strong points – speed, competitive reduction in wake losses, and flexibility – and weak points – partial blindness to the objective and dependence on the starting layout. The computational speed of pseudo-gradient-based optimization is an enabler for analyses that would otherwise be computationally impractical. Pseudo-gradient-based optimization has already been used by industry in the design of large-scale (offshore) wind farms. |
2020 |
Quaeghebeur, Erik; Zaaijer, Michiel B How to improve the state of the art in metocean measurement datasets Journal Article Wind Energy Science, 5 (1), pp. 285–308, 2020. @article{QuaeghebeurZ-WES20j_, title = {How to improve the state of the art in metocean measurement datasets}, author = {Erik Quaeghebeur and Michiel B. Zaaijer}, doi = {10.5194/wes-5-285-2020}, year = {2020}, date = {2020-02-28}, journal = {Wind Energy Science}, volume = {5}, number = {1}, pages = {285–308}, abstract = {We present an analysis of three datasets of 10 min metocean measurement statistics and our resulting recommendations to both producers and users of such datasets. Many of our recommendations are more generally of interest to all numerical measurement data producers. The datasets analyzed originate from offshore meteorological masts installed to support offshore wind farm planning and design: the Dutch OWEZ and MMIJ and the German FINO1. Our analysis shows that such datasets contain issues that users should look out for and whose prevalence can be reduced by producers. We also present expressions to derive uncertainty and bias values for the statistics from information typically available about sample uncertainty. We also observe that the format in which the data are disseminated is sub-optimal from the users' perspective and discuss how producers can create more immediately useful dataset files. Effectively, we advocate using an established binary format (HDF5 or netCDF4) instead of the typical text-based one (comma-separated values), as this allows for the inclusion of relevant metadata and the creation of significantly smaller directly accessible dataset files. Next to informing producers of the advantages of these formats, we also provide concrete pointers to their effective use. Our conclusion is that datasets such as the ones we analyzed can be improved substantially in usefulness and convenience with limited effort.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We present an analysis of three datasets of 10 min metocean measurement statistics and our resulting recommendations to both producers and users of such datasets. Many of our recommendations are more generally of interest to all numerical measurement data producers. The datasets analyzed originate from offshore meteorological masts installed to support offshore wind farm planning and design: the Dutch OWEZ and MMIJ and the German FINO1. Our analysis shows that such datasets contain issues that users should look out for and whose prevalence can be reduced by producers. We also present expressions to derive uncertainty and bias values for the statistics from information typically available about sample uncertainty. We also observe that the format in which the data are disseminated is sub-optimal from the users' perspective and discuss how producers can create more immediately useful dataset files. Effectively, we advocate using an established binary format (HDF5 or netCDF4) instead of the typical text-based one (comma-separated values), as this allows for the inclusion of relevant metadata and the creation of significantly smaller directly accessible dataset files. Next to informing producers of the advantages of these formats, we also provide concrete pointers to their effective use. Our conclusion is that datasets such as the ones we analyzed can be improved substantially in usefulness and convenience with limited effort. |
Quaeghebeur, Erik; Sanchez Perez-Moreno, Sebastian ; Zaaijer, Michiel B WESgraph: a graph database for the wind farm domain Journal Article Wind Energy Science, 5 (1), pp. 259–284, 2020. @article{QuaeghebeurSZ-WES20j_, title = {WESgraph: a graph database for the wind farm domain}, author = {Erik Quaeghebeur and Sebastian {Sanchez Perez-Moreno} and Michiel B. Zaaijer}, doi = {10.5194/wes-5-259-2020}, year = {2020}, date = {2020-02-27}, journal = {Wind Energy Science}, volume = {5}, number = {1}, pages = {259–284}, abstract = {The construction and management of a wind farm involve many disciplines. It is hard for a single designer or developer to keep an overview of all the relevant concepts, models, and tools. Nevertheless, this is needed when performing integrated modeling or analysis. To help researchers keep this overview, we have created WESgraph (the Wind Energy System graph), a knowledge base for the wind farm domain, implemented as a graph database. It currently contains 1222 concepts and 1725 relations between them. This paper presents the structure of this graph database – content stored in nodes and the relationships between them – as a foundational ontology, which classifies the domain's concepts. This foundational ontology partitions the domain in two: a part describing physical aspects and a part describing mathematical and computational aspects. This paper also discusses a number of generally difficult cases that exist when adding content to such a knowledge base. This paper furthermore discusses the potential applications of WESgraph and illustrates its use for computation pathway discovery – the application that triggered its creation. It also contains a description of our practical experience with its design and use as well as our thoughts about the community use and management of this tool.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The construction and management of a wind farm involve many disciplines. It is hard for a single designer or developer to keep an overview of all the relevant concepts, models, and tools. Nevertheless, this is needed when performing integrated modeling or analysis. To help researchers keep this overview, we have created WESgraph (the Wind Energy System graph), a knowledge base for the wind farm domain, implemented as a graph database. It currently contains 1222 concepts and 1725 relations between them. This paper presents the structure of this graph database – content stored in nodes and the relationships between them – as a foundational ontology, which classifies the domain's concepts. This foundational ontology partitions the domain in two: a part describing physical aspects and a part describing mathematical and computational aspects. This paper also discusses a number of generally difficult cases that exist when adding content to such a knowledge base. This paper furthermore discusses the potential applications of WESgraph and illustrates its use for computation pathway discovery – the application that triggered its creation. It also contains a description of our practical experience with its design and use as well as our thoughts about the community use and management of this tool. |
Meijers, P C; Tsouvalas, A; Metrikine, A V Monitoring Monopile Penetration through Magnetic Stray Field Measurements Inproceedings Papadrakakis, M; Fragiadakis, M; Papadimitriou, C (Ed.): Proceedings of the XI International Conference on Structural Dynamics, pp. 1272–1280, EASD Procedia, Athens, Greece, 2020. @inproceedings{Meijers2020, title = {Monitoring Monopile Penetration through Magnetic Stray Field Measurements}, author = {P C Meijers and A Tsouvalas and A V Metrikine}, editor = {M Papadrakakis and M Fragiadakis and C Papadimitriou}, doi = {10.47964/1120.9102.19534}, year = {2020}, date = {2020-01-01}, booktitle = {Proceedings of the XI International Conference on Structural Dynamics}, volume = {1}, pages = {1272--1280}, publisher = {EASD Procedia}, address = {Athens, Greece}, abstract = {Current methods to infer the penetration of a steel monopile during an offshore installation are rather inaccurate. Since a large number of foundation piles will be installed offshore in the coming years, a reliable technique to infer the penetration depth is vital. This paper proposes a non-contact method to monitor the pile progression into the seabed based on measurements of the magnetic stray field that permeates the air surrounding the structure, eliminating the necessity of a predefined pattern on the pile’s surface. A simple magnetisation model for the monopile is proposed from which the relative motion between the moving pile and a stationary magnetic field sensor can be extracted. Comparison between the measured and simulated stray field data show a promising correlation, providing the basis for the new non-contact monitoring technique that is applicable offshore.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Current methods to infer the penetration of a steel monopile during an offshore installation are rather inaccurate. Since a large number of foundation piles will be installed offshore in the coming years, a reliable technique to infer the penetration depth is vital. This paper proposes a non-contact method to monitor the pile progression into the seabed based on measurements of the magnetic stray field that permeates the air surrounding the structure, eliminating the necessity of a predefined pattern on the pile’s surface. A simple magnetisation model for the monopile is proposed from which the relative motion between the moving pile and a stationary magnetic field sensor can be extracted. Comparison between the measured and simulated stray field data show a promising correlation, providing the basis for the new non-contact monitoring technique that is applicable offshore. |
van den Bos, Laurent M M; Sanderse, Benjamin; Bierbooms, Wim A A M; van Bussel, Gerard J W Bayesian model calibration with interpolating polynomials based on adaptively weighted Leja nodes Journal Article Communications in Computational Physics, 27 (1), pp. 33–69, 2020. @article{Bos2020, title = {Bayesian model calibration with interpolating polynomials based on adaptively weighted Leja nodes}, author = {Laurent M. M. van den Bos and Benjamin Sanderse and Wim A. A. M. Bierbooms and Gerard J. W. van Bussel}, url = {https://arxiv.org/abs/1802.02035}, doi = {10.4208/cicp.oa-2018-0218}, year = {2020}, date = {2020-01-00}, journal = {Communications in Computational Physics}, volume = {27}, number = {1}, pages = {33--69}, publisher = {Global Science Press}, abstract = {An efficient algorithm is proposed for Bayesian model calibration, which is commonly used to estimate the model parameters of non-linear, computationally expensive models using measurement data. The approach is based on Bayesian statistics: using a prior distribution and a likelihood, the posterior distribution is obtained through application of Bayes’ law. Our novel algorithm to accurately determine this posterior requires significantly fewer discrete model evaluations than traditional Monte Carlo methods. The key idea is to replace the expensive model by an interpolating surrogate model and to construct the interpolating nodal set maximizing the accuracy of the posterior. To determine such a nodal set an extension to weighted Leja nodes is introduced, based on a new weighting function. We prove that the convergence of the posterior has the same rate as the convergence of the model. If the convergence of the posterior is measured in the Kullback–Leibler divergence, the rate doubles. The algorithm and its theoretical properties are verified in three different test cases: analytical cases that confirm the correctness of the theoretical findings, Burgers’ equation to show its applicability in implicit problems, and finally the calibration of the closure parameters of a turbulence model to show the effectiveness for computationally expensive problems.}, keywords = {}, pubstate = {published}, tppubtype = {article} } An efficient algorithm is proposed for Bayesian model calibration, which is commonly used to estimate the model parameters of non-linear, computationally expensive models using measurement data. The approach is based on Bayesian statistics: using a prior distribution and a likelihood, the posterior distribution is obtained through application of Bayes’ law. Our novel algorithm to accurately determine this posterior requires significantly fewer discrete model evaluations than traditional Monte Carlo methods. The key idea is to replace the expensive model by an interpolating surrogate model and to construct the interpolating nodal set maximizing the accuracy of the posterior. To determine such a nodal set an extension to weighted Leja nodes is introduced, based on a new weighting function. We prove that the convergence of the posterior has the same rate as the convergence of the model. If the convergence of the posterior is measured in the Kullback–Leibler divergence, the rate doubles. The algorithm and its theoretical properties are verified in three different test cases: analytical cases that confirm the correctness of the theoretical findings, Burgers’ equation to show its applicability in implicit problems, and finally the calibration of the closure parameters of a turbulence model to show the effectiveness for computationally expensive problems. |
2019 |
Eggels, A W Uncertainty quantification with dependent input data - including applications to offshore wind farms PhD Thesis University of Amsterdam, 2019, ISBN: 978-94-6323-848-9. @phdthesis{Egg19, title = {Uncertainty quantification with dependent input data - including applications to offshore wind farms}, author = {A. W. Eggels}, isbn = {978-94-6323-848-9}, year = {2019}, date = {2019-11-06}, school = {University of Amsterdam}, key = {WP1.3}, keywords = {}, pubstate = {published}, tppubtype = {phdthesis} } |
Quaeghebeur, Erik Robust wind farm layout optimization using pseudo-gradients Inproceedings ISIPTA 2019, SIPTA 2019. @inproceedings{Quaeghebeur-ISIPTA2019, title = {Robust wind farm layout optimization using pseudo-gradients}, author = {Erik Quaeghebeur}, url = {http://resolver.tudelft.nl/uuid:17472c3e-53aa-4f43-a684-4959ca474471}, year = {2019}, date = {2019-07-03}, booktitle = {ISIPTA 2019}, organization = {SIPTA}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Meijers, P C; Tsouvalas, A; Metrikine, A V Non-contact system for monitoring a metallic magnetic structure under dynamic load Patent N2023374, 2019. @patent{Meijers2019, title = {Non-contact system for monitoring a metallic magnetic structure under dynamic load}, author = {P C Meijers and A Tsouvalas and A V Metrikine}, year = {2019}, date = {2019-06-25}, number = {N2023374}, abstract = {The present invention is in the field of a non—contact system for monitoring a metallic magnetic structure under dynamic load for detecting an impact induced propagating stress wave, and a method of determining strain in a metallic magnetic structure under dynamic load, such as a tube-like structure, such as a monopile for a wind turbine.}, keywords = {}, pubstate = {published}, tppubtype = {patent} } The present invention is in the field of a non—contact system for monitoring a metallic magnetic structure under dynamic load for detecting an impact induced propagating stress wave, and a method of determining strain in a metallic magnetic structure under dynamic load, such as a tube-like structure, such as a monopile for a wind turbine. |
Quaeghebeur, Erik Flexible and efficient site constraint handling for wind farm layout optimization Inproceedings Wind Energy Science Conference (WESC 2019), EAWE 2019. @inproceedings{Quaeghebeur-WESC2019, title = {Flexible and efficient site constraint handling for wind farm layout optimization}, author = {Erik Quaeghebeur}, year = {2019}, date = {2019-06-20}, booktitle = {Wind Energy Science Conference (WESC 2019)}, organization = {EAWE}, abstract = {Wind farm sites can have complex, disconnected shapes and may encompass exclusion zones. Even offshore this is the case, due to sea lanes, underwater pipelines and cables, wrecks, and unidentified buried objects. An example is Borssele Wind Farm Site IV1 (BWFS IV), which is pictured in Figure 11 as the collection of greencolored parcels. Within BWFS IV, as shown in Figure 21, there is an archaeologically significant wreck—red boat—and multiple magnetic anomalies— green dots—that indicate buried objects. Furthermore, regulations require turbines to be placed a certain safety distance—one rotor radius— inside the parcels. }, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Wind farm sites can have complex, disconnected shapes and may encompass exclusion zones. Even offshore this is the case, due to sea lanes, underwater pipelines and cables, wrecks, and unidentified buried objects. An example is Borssele Wind Farm Site IV1 (BWFS IV), which is pictured in Figure 11 as the collection of greencolored parcels. Within BWFS IV, as shown in Figure 21, there is an archaeologically significant wreck—red boat—and multiple magnetic anomalies— green dots—that indicate buried objects. Furthermore, regulations require turbines to be placed a certain safety distance—one rotor radius— inside the parcels. |
Kalverla, Peter C; Jr., Duncan J B; Steeneveld, Gert-Jan; Holtslag, Albert A M Low-level jets over the North Sea based on ERA5 and observations: together they do better Journal Article Wind Energy Science, 4 (2), pp. 193–209, 2019. @article{wes-4-193-2019, title = {Low-level jets over the North Sea based on ERA5 and observations: together they do better}, author = {Peter C. Kalverla and J. B. Duncan Jr. and Gert-Jan Steeneveld and Albert A. M. Holtslag}, doi = {10.5194/wes-4-193-2019}, year = {2019}, date = {2019-04-03}, journal = {Wind Energy Science}, volume = {4}, number = {2}, pages = {193--209}, abstract = {Ten years of ERA5 reanalysis data are combined with met-mast and lidar observations from 10 offshore platforms to investigate low-level jet characteristics over the Dutch North Sea. The objective of this study is to combine the best of two worlds: (1) ERA5 data with a large spatiotemporal extent but inherent accuracy limitations due to a relatively coarse grid and an incomplete representation of physical processes and (2) observations that provide more reliable estimates of the measured quantity but are limited in both space and time. We demonstrate the effect of time and range limitations on the reconstructed wind climate, with special attention paid to the impact on low-level jets. For both measurement and model data, the representation of wind speed is biased. The limited temporal extent of observations leads to a wind speed bias on the order of ±1 m s−1 as compared to the long-term mean. In part due to data-assimilation strategies that cause abrupt discontinuities in the diurnal cycle, ERA5 also exhibits a wind speed bias of approximately 0.5 m s−1. The representation of low-level jets in ERA5 is poor in terms of a one-to-one correspondence, and the jets appear vertically displaced (“smeared out”). However, climatological characteristics such as the shape of the seasonal cycle and the affinity with certain circulation patterns are represented quite well, albeit with different magnitudes. We therefore experiment with various methods to adjust the modelled low-level jet rate to the observations or, vice versa, to correct for the erratic nature of the short observation periods using long-term ERA5 information. While quantitative uncertainty is still quite large, the presented results provide valuable insight into North Sea low-level jet characteristics. These jets occur predominantly for circulation types with an easterly component, with a clear peak in spring, and are concentrated along the coasts at heights between 50 and 200 m. Further, it is demonstrated that these characteristics can be used as predictors to infer the observed low-level jet rate from ERA5 data with reasonable accuracy.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Ten years of ERA5 reanalysis data are combined with met-mast and lidar observations from 10 offshore platforms to investigate low-level jet characteristics over the Dutch North Sea. The objective of this study is to combine the best of two worlds: (1) ERA5 data with a large spatiotemporal extent but inherent accuracy limitations due to a relatively coarse grid and an incomplete representation of physical processes and (2) observations that provide more reliable estimates of the measured quantity but are limited in both space and time. We demonstrate the effect of time and range limitations on the reconstructed wind climate, with special attention paid to the impact on low-level jets. For both measurement and model data, the representation of wind speed is biased. The limited temporal extent of observations leads to a wind speed bias on the order of ±1 m s−1 as compared to the long-term mean. In part due to data-assimilation strategies that cause abrupt discontinuities in the diurnal cycle, ERA5 also exhibits a wind speed bias of approximately 0.5 m s−1. The representation of low-level jets in ERA5 is poor in terms of a one-to-one correspondence, and the jets appear vertically displaced (“smeared out”). However, climatological characteristics such as the shape of the seasonal cycle and the affinity with certain circulation patterns are represented quite well, albeit with different magnitudes. We therefore experiment with various methods to adjust the modelled low-level jet rate to the observations or, vice versa, to correct for the erratic nature of the short observation periods using long-term ERA5 information. While quantitative uncertainty is still quite large, the presented results provide valuable insight into North Sea low-level jet characteristics. These jets occur predominantly for circulation types with an easterly component, with a clear peak in spring, and are concentrated along the coasts at heights between 50 and 200 m. Further, it is demonstrated that these characteristics can be used as predictors to infer the observed low-level jet rate from ERA5 data with reasonable accuracy. |
Eggels, Anne W; Crommelin, Daan T Quantifying Data Dependencies with Rényi Mutual Information and Minimum Spanning Trees Journal Article Entropy, 21 (2), pp. 100, 2019, ISSN: 1099-4300. @article{Eggels_2019, title = {Quantifying Data Dependencies with Rényi Mutual Information and Minimum Spanning Trees}, author = {Anne W. Eggels and Daan T. Crommelin}, doi = {10.3390/e21020100}, issn = {1099-4300}, year = {2019}, date = {2019-01-22}, journal = {Entropy}, volume = {21}, number = {2}, pages = {100}, publisher = {MDPI AG}, abstract = {In this study, we present a novel method for quantifying dependencies in multivariate datasets, based on estimating the Rényi mutual information by minimum spanning trees (MSTs). The extent to which random variables are dependent is an important question, e.g., for uncertainty quantification and sensitivity analysis. The latter is closely related to the question how strongly dependent the output of, e.g., a computer simulation, is on the individual random input variables. To estimate the Rényi mutual information from data, we use a method due to Hero et al. that relies on computing minimum spanning trees (MSTs) of the data and uses the length of the MST in an estimator for the entropy. To reduce the computational cost of constructing the exact MST for large datasets, we explore methods to compute approximations to the exact MST, and find the multilevel approach introduced recently by Zhong et al. (2015) to be the most accurate. Because the MST computation does not require knowledge (or estimation) of the distributions, our methodology is well-suited for situations where only data are available. Furthermore, we show that, in the case where only the ranking of several dependencies is required rather than their exact value, it is not necessary to compute the Rényi divergence, but only an estimator derived from it. The main contributions of this paper are the introduction of this quantifier of dependency, as well as the novel combination of using approximate methods for MSTs with estimating the Rényi mutual information via MSTs. We applied our proposed method to an artificial test case based on the Ishigami function, as well as to a real-world test case involving an El Nino dataset.}, keywords = {}, pubstate = {published}, tppubtype = {article} } In this study, we present a novel method for quantifying dependencies in multivariate datasets, based on estimating the Rényi mutual information by minimum spanning trees (MSTs). The extent to which random variables are dependent is an important question, e.g., for uncertainty quantification and sensitivity analysis. The latter is closely related to the question how strongly dependent the output of, e.g., a computer simulation, is on the individual random input variables. To estimate the Rényi mutual information from data, we use a method due to Hero et al. that relies on computing minimum spanning trees (MSTs) of the data and uses the length of the MST in an estimator for the entropy. To reduce the computational cost of constructing the exact MST for large datasets, we explore methods to compute approximations to the exact MST, and find the multilevel approach introduced recently by Zhong et al. (2015) to be the most accurate. Because the MST computation does not require knowledge (or estimation) of the distributions, our methodology is well-suited for situations where only data are available. Furthermore, we show that, in the case where only the ranking of several dependencies is required rather than their exact value, it is not necessary to compute the Rényi divergence, but only an estimator derived from it. The main contributions of this paper are the introduction of this quantifier of dependency, as well as the novel combination of using approximate methods for MSTs with estimating the Rényi mutual information via MSTs. We applied our proposed method to an artificial test case based on the Ishigami function, as well as to a real-world test case involving an El Nino dataset. |
Eggels, A W; Crommelin, D T Efficient estimation of divergence-based sensitivity indices with Gaussian process surrogates Miscellaneous 2019, (submitted). @misc{eggels2019efficient, title = {Efficient estimation of divergence-based sensitivity indices with Gaussian process surrogates}, author = {A W Eggels and D T Crommelin}, url = {https://arxiv.org/abs/1904.03859}, year = {2019}, date = {2019-01-01}, abstract = {We consider the estimation of sensitivity indices based on divergence measures such as Hellinger distance. For sensitivity analysis of complex models, these divergence-based indices can be estimated by Monte-Carlo sampling (MCS) in combination with kernel density estimation (KDE). In a direct approach, the complex model must be evaluated at every input point generated by MCS, resulting in samples in the input-output space that can be used for density estimation. However, if the computational cost of the complex model strongly limits the number of model evaluations, this direct method gives large errors. We propose to use Gaussian process (GP) surrogates to increase the number of samples in the combined input-output space. By enlarging this sample set, the KDE becomes more accurate, leading to improved estimates. To compare the GP surrogates, we use a surrogate constructed by samples obtained with stochastic collocation, combined with Lagrange interpolation. Furthermore, we propose a new estimation method for these sensitivity indices based on minimum spanning trees. Finally, we also propose a new type of sensitivity indices based on divergence measures, namely direct sensitivity indices. These are useful when the input data is dependent.}, note = {submitted}, keywords = {}, pubstate = {published}, tppubtype = {misc} } We consider the estimation of sensitivity indices based on divergence measures such as Hellinger distance. For sensitivity analysis of complex models, these divergence-based indices can be estimated by Monte-Carlo sampling (MCS) in combination with kernel density estimation (KDE). In a direct approach, the complex model must be evaluated at every input point generated by MCS, resulting in samples in the input-output space that can be used for density estimation. However, if the computational cost of the complex model strongly limits the number of model evaluations, this direct method gives large errors. We propose to use Gaussian process (GP) surrogates to increase the number of samples in the combined input-output space. By enlarging this sample set, the KDE becomes more accurate, leading to improved estimates. To compare the GP surrogates, we use a surrogate constructed by samples obtained with stochastic collocation, combined with Lagrange interpolation. Furthermore, we propose a new estimation method for these sensitivity indices based on minimum spanning trees. Finally, we also propose a new type of sensitivity indices based on divergence measures, namely direct sensitivity indices. These are useful when the input data is dependent. |
Leontaris, George; Morales Napoles, Oswaldo ; Dewan, Ashish; Wolfert, Rogier Decision support for offshore asset construction using expert judgments for supply disruptions risk Journal Article Automation in Construction, 107 , 2019, ISSN: 0926-5805, (Accepted Author Manuscript). @article{6cb62b5438b340ebb459faf9f11408da, title = {Decision support for offshore asset construction using expert judgments for supply disruptions risk}, author = {George Leontaris and Oswaldo {Morales Napoles} and Ashish Dewan and Rogier Wolfert}, doi = {10.1016/j.autcon.2019.102903}, issn = {0926-5805}, year = {2019}, date = {2019-01-01}, journal = {Automation in Construction}, volume = {107}, publisher = {Elsevier}, abstract = {Offshore asset construction is a complex and costly process that is subject to various uncertainties within the entire supply chain. Hence, both the construction management optimization and the reduction of deployment expenditures should be supported by automated decision support models which include proper representations of predominant uncertainties. One of these is the supply disruption risk that is often ignored in existing models. Therefore, this article proposes a methodology to properly take this construction risk into account. An algorithm to model this risk was developed and a study was conducted to obtain the required probability distributions of disruption delays using real data and expert judgments for an offshore wind farm construction application. The simulation of a realistic test case with an appropriately modified stochastic simulation tool showed that it is important to consider this risk in order to make optimal decisions for different offshore wind farm construction strategies.}, note = {Accepted Author Manuscript}, keywords = {}, pubstate = {published}, tppubtype = {article} } Offshore asset construction is a complex and costly process that is subject to various uncertainties within the entire supply chain. Hence, both the construction management optimization and the reduction of deployment expenditures should be supported by automated decision support models which include proper representations of predominant uncertainties. One of these is the supply disruption risk that is often ignored in existing models. Therefore, this article proposes a methodology to properly take this construction risk into account. An algorithm to model this risk was developed and a study was conducted to obtain the required probability distributions of disruption delays using real data and expert judgments for an offshore wind farm construction application. The simulation of a realistic test case with an appropriately modified stochastic simulation tool showed that it is important to consider this risk in order to make optimal decisions for different offshore wind farm construction strategies. |
2018 |
Eggels, Anne W; Crommelin, Daan T Uncertainty quantification with dependent inputs: wind and waves Inproceedings 6th European Conference on Computational Mechanics (ECCM 6) – 7th European Conference on Computational Fluid Dynamics (ECFD 7), ECCOMAS 2018. @inproceedings{4Egg18, title = {Uncertainty quantification with dependent inputs: wind and waves}, author = {Anne W. Eggels and Daan T. Crommelin}, url = {http://www.eccm-ecfd2018.org/admin/files/filePaper/p300.pdf}, year = {2018}, date = {2018-06-00}, booktitle = {6th European Conference on Computational Mechanics (ECCM 6) – 7th European Conference on Computational Fluid Dynamics (ECFD 7)}, organization = {ECCOMAS}, abstract = {A framework for performing uncertainty quantification is presented which is well-suited for systems with dependent inputs with unknown distributions. The multivariate input is given as a dataset whose variables can have strong, nonlinear dependencies. For each of the elements in the framework (dependency analysis, sample selection and sensitivity analysis), we recently developed new methods, which are here combined for the first time. The framework is tested on an example involving a wind farm simulation with offshore weather conditions as input. }, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } A framework for performing uncertainty quantification is presented which is well-suited for systems with dependent inputs with unknown distributions. The multivariate input is given as a dataset whose variables can have strong, nonlinear dependencies. For each of the elements in the framework (dependency analysis, sample selection and sensitivity analysis), we recently developed new methods, which are here combined for the first time. The framework is tested on an example involving a wind farm simulation with offshore weather conditions as input. |
Dos Santos Pereira, Ricardo Balbino ; De Oliveira Andrade, Gael ; Timmer, Nando; Quaeghebeur, Erik Probabilistic Design of Airfoils for Horizontal Axis Wind Turbines Journal Article Journal of Physics: Conference Series, 1037 , 2018, ISBN: 1742-6588. @article{Quaeghebeur-TORQUE2018, title = {Probabilistic Design of Airfoils for Horizontal Axis Wind Turbines}, author = {Dos Santos Pereira, Ricardo Balbino and De Oliveira Andrade, Gael and Nando Timmer and Erik Quaeghebeur}, doi = {10.1088/1742-6596/1037/2/022042}, isbn = {1742-6588}, year = {2018}, date = {2018-06-00}, journal = {Journal of Physics: Conference Series}, volume = {1037}, abstract = {We describe a probabilistic approach to design airfoils for wind energy applications. An analytical expression is derived for the probability of perturbations to the operational blade-section angle of attack. It includes the combined influence of wind shear, yaw-misalignment, and turbulence intensity. The theoretical fluctuations in angle of attack are validated against an aero-structural simulation of a 10 MW horizontal axis wind turbine, operating under different inflow conditions. Finally we incorporate the probabilistic approach into a multi-objective airfoil optimization problem, which is solved with a genetic algorithm. The results illustrate the compromise between airfoil performance for a specific angle of attack and robustness of airfoil performance over a large range of angle of attack fluctuations}, keywords = {}, pubstate = {published}, tppubtype = {article} } We describe a probabilistic approach to design airfoils for wind energy applications. An analytical expression is derived for the probability of perturbations to the operational blade-section angle of attack. It includes the combined influence of wind shear, yaw-misalignment, and turbulence intensity. The theoretical fluctuations in angle of attack are validated against an aero-structural simulation of a 10 MW horizontal axis wind turbine, operating under different inflow conditions. Finally we incorporate the probabilistic approach into a multi-objective airfoil optimization problem, which is solved with a genetic algorithm. The results illustrate the compromise between airfoil performance for a specific angle of attack and robustness of airfoil performance over a large range of angle of attack fluctuations |
Meijers, Peter C; Tsouvalas, Apostolos; Metrikine, Andrei V Plasticity detection and quantification in monopile support structures due to axial impact loading Inproceedings Manoach, E; Stoykov, S; Wiercigroch, M (Ed.): ICoEV 2017: International Conference on Engineering Vibration, pp. 15003, EDP Sciences, 2018. @inproceedings{Meijers2018, title = {Plasticity detection and quantification in monopile support structures due to axial impact loading}, author = {Peter C. Meijers and Apostolos Tsouvalas and Andrei V. Metrikine}, editor = {E. Manoach and S. Stoykov and M. Wiercigroch }, doi = {10.1051/matecconf/201814815003}, year = {2018}, date = {2018-02-01}, booktitle = {ICoEV 2017: International Conference on Engineering Vibration}, volume = {148}, pages = {15003}, publisher = {EDP Sciences}, series = {MATEC Web Conference}, abstract = {Recent developments in the construction of offshore wind turbines have created the need for a method to detect whether a monopile foundation is plastically deformed during the installation procedure. Since measurements at the pile head are difficult to perform, a method based on measurements at a certain distance below the pile head is proposed in this work for quantification of the amount of plasticity. By considering a onedimensional rod model with an elastic-perfectly plastic constitutive relation, it is shown that the occurrence of plastic deformation caused by an impact load can be detected from these measurements. Furthermore, this plastic deformation can be quantified by the same measurement with the help of an energy balance. The effectiveness of the proposed method is demonstrated via a numerical example.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Recent developments in the construction of offshore wind turbines have created the need for a method to detect whether a monopile foundation is plastically deformed during the installation procedure. Since measurements at the pile head are difficult to perform, a method based on measurements at a certain distance below the pile head is proposed in this work for quantification of the amount of plasticity. By considering a onedimensional rod model with an elastic-perfectly plastic constitutive relation, it is shown that the occurrence of plastic deformation caused by an impact load can be detected from these measurements. Furthermore, this plastic deformation can be quantified by the same measurement with the help of an energy balance. The effectiveness of the proposed method is demonstrated via a numerical example. |
Eggels, Anne W; Crommelin, Daan T; Witteveen, Jeroen A S Clustering-based collocation for uncertainty propagation with multivariate dependent inputs Journal Article International Journal for Uncertainty Quantification, 8 (1), pp. 43–59, 2018. @article{Eggels2018, title = {Clustering-based collocation for uncertainty propagation with multivariate dependent inputs}, author = {Anne W. Eggels and Daan T. Crommelin and Jeroen A. S. Witteveen}, url = {https://ir.cwi.nl/pub/27271 https://arxiv.org/abs/1703.06112}, doi = {10.1615/int.j.uncertaintyquantification.2018020215}, year = {2018}, date = {2018-01-01}, journal = {International Journal for Uncertainty Quantification}, volume = {8}, number = {1}, pages = {43–59}, publisher = {Begell House}, abstract = {In this paper, we propose the use of partitioning and clustering methods as an alternative to Gaussian quadrature for stochastic collocation. The key idea is to use cluster centers as the nodes for collocation. In this way, we can extend the use of collocation methods to uncertainty propagation with multivariate, dependent input, in which the output approximation is piecewise constant on the clusters. The approach is particularly useful in situations where the probability distribution of the input is unknown and only a sample from the input distribution is available. We examine several clustering methods and assess the convergence of collocation based on these methods both theoretically and numerically. We demonstrate good performance of the proposed methods, most notably for the challenging case of nonlinearly dependent inputs in higher dimensions. Numerical tests with input dimension up to 16 are included, using as benchmarks the Genz test functions and a test case from computational fluid dynamics (lid-driven cavity flow). }, keywords = {}, pubstate = {published}, tppubtype = {article} } In this paper, we propose the use of partitioning and clustering methods as an alternative to Gaussian quadrature for stochastic collocation. The key idea is to use cluster centers as the nodes for collocation. In this way, we can extend the use of collocation methods to uncertainty propagation with multivariate, dependent input, in which the output approximation is piecewise constant on the clusters. The approach is particularly useful in situations where the probability distribution of the input is unknown and only a sample from the input distribution is available. We examine several clustering methods and assess the convergence of collocation based on these methods both theoretically and numerically. We demonstrate good performance of the proposed methods, most notably for the challenging case of nonlinearly dependent inputs in higher dimensions. Numerical tests with input dimension up to 16 are included, using as benchmarks the Genz test functions and a test case from computational fluid dynamics (lid-driven cavity flow). |
Eggels, Anne W; Crommelin, Daan T Quantifying dependencies for sensitivity analysis with multivariate input sample data Unpublished 2018. @unpublished{Eggels2018b, title = {Quantifying dependencies for sensitivity analysis with multivariate input sample data}, author = {Anne W. Eggels and Daan T. Crommelin}, url = {https://arxiv.org/abs/1802.01841}, year = {2018}, date = {2018-00-00}, abstract = {We present a novel method for quantifying dependencies in multivariate datasets, based on estimating the Rényi entropy by minimum spanning trees (MSTs). The length of the MSTs can be used to order pairs of variables from strongly to weakly dependent, making it a useful tool for sensitivity analysis with dependent input variables. It is well-suited for cases where the input distribution is unknown and only a sample of the inputs is available. We introduce an estimator to quantify dependency based on the MST length, and investigate its properties with several numerical examples. To reduce the computational cost of constructing the exact MST for large datasets, we explore methods to compute approximations to the exact MST, and find the multilevel approach introduced recently by Zhong et al. (2015) to be the most accurate. We apply our proposed method to an artificial testcase based on the Ishigami function, as well as to a real-world testcase involving sediment transport in the North Sea. The results are consistent with prior knowledge and heuristic understanding, as well as with variance-based analysis using Sobol indices in the case where these indices can be computed.}, keywords = {}, pubstate = {published}, tppubtype = {unpublished} } We present a novel method for quantifying dependencies in multivariate datasets, based on estimating the Rényi entropy by minimum spanning trees (MSTs). The length of the MSTs can be used to order pairs of variables from strongly to weakly dependent, making it a useful tool for sensitivity analysis with dependent input variables. It is well-suited for cases where the input distribution is unknown and only a sample of the inputs is available. We introduce an estimator to quantify dependency based on the MST length, and investigate its properties with several numerical examples. To reduce the computational cost of constructing the exact MST for large datasets, we explore methods to compute approximations to the exact MST, and find the multilevel approach introduced recently by Zhong et al. (2015) to be the most accurate. We apply our proposed method to an artificial testcase based on the Ishigami function, as well as to a real-world testcase involving sediment transport in the North Sea. The results are consistent with prior knowledge and heuristic understanding, as well as with variance-based analysis using Sobol indices in the case where these indices can be computed. |
van den Bos, Laurent M M; Sanderse, Benjamin; Blonk, Lindert; Bierbooms, Wim A A M; van Bussel, Gerard J W Efficient ultimate load estimation for offshore wind turbines using interpolating surrogate models Journal Article Journal of Physics: Conference Series, 1037 (6), pp. 062017, 2018. @article{Bos2018a, title = {Efficient ultimate load estimation for offshore wind turbines using interpolating surrogate models}, author = {Laurent M. M. van den Bos and Benjamin Sanderse and Lindert Blonk and Wim A. A. M. Bierbooms and Gerard J. W. van Bussel}, doi = {10.1088/1742-6596/1037/6/062017}, year = {2018}, date = {2018-00-00}, journal = {Journal of Physics: Conference Series}, volume = {1037}, number = {6}, pages = {062017}, publisher = {IOP Publishing}, abstract = {During the design phase of an offshore wind turbine, it is required to assess the impact of loads on the turbine life time. Due to the varying environmental conditions, the effect of various uncertain parameters has to be studied to provide meaningful conclusions. Incorporating such uncertain parameters in this regard is often done by applying binning, where the probability density function under consideration is binned and in each bin random simulations are run to estimate the loads. A different methodology for quantifying uncertainties proposed in this work is polynomial interpolation, a more efficient technique that allows to more accurately predict the loads on the turbine for specific load cases. This efficiency is demonstrated by applying the technique to a power production test problem and to IEC Design Load Case 1.1, where the ultimate loads are determined using BLADED. The results show that the interpolating polynomial is capable of representing the load model. Our proposed surrogate modeling approach therefore has the potential to significantly speed up the design and analysis of offshore wind turbines by reducing the time required for load case assessment.}, keywords = {}, pubstate = {published}, tppubtype = {article} } During the design phase of an offshore wind turbine, it is required to assess the impact of loads on the turbine life time. Due to the varying environmental conditions, the effect of various uncertain parameters has to be studied to provide meaningful conclusions. Incorporating such uncertain parameters in this regard is often done by applying binning, where the probability density function under consideration is binned and in each bin random simulations are run to estimate the loads. A different methodology for quantifying uncertainties proposed in this work is polynomial interpolation, a more efficient technique that allows to more accurately predict the loads on the turbine for specific load cases. This efficiency is demonstrated by applying the technique to a power production test problem and to IEC Design Load Case 1.1, where the ultimate loads are determined using BLADED. The results show that the interpolating polynomial is capable of representing the load model. Our proposed surrogate modeling approach therefore has the potential to significantly speed up the design and analysis of offshore wind turbines by reducing the time required for load case assessment. |
Meijers, P C; Tsouvalas, A; Metrikine, A V A Non-Collocated Method to Quantify Plastic Deformation Caused by Impact Pile Driving Journal Article International Journal of Mechanical Sciences, 148 , pp. 1-8, 2018, ISSN: 0020-7403. @article{Meijers2018a, title = {A Non-Collocated Method to Quantify Plastic Deformation Caused by Impact Pile Driving}, author = {P. C. Meijers and A. Tsouvalas and A. V. Metrikine}, doi = {10.1016/j.ijmecsci.2018.08.013}, issn = {0020-7403}, year = {2018}, date = {2018-00-00}, journal = {International Journal of Mechanical Sciences}, volume = {148}, pages = {1-8}, abstract = {The use of bolted connections between the tower and a support structure of an offshore wind turbine has created the need for a method to detect whether a monopile foundation plastically deforms during the installation procedure. Small permanent deformations are undesirable, not only because they can accelerate fatigue of the structure; but also because they can lead to misalignment between the tower and the foundation. Since direct measurements at the pile head are difficult to perform, a method based on non-collocated strain measurements is highly desirable. This paper proposes such a method. First, a physically non-linear one-dimensional model is proposed, which accounts for wave dispersion, effects that are relevant for large-diameter piles currently used by the industry. The proposed model, combined with an energy balance principle, gives an upper bound for the amount of plastic deformation caused by an impact load based on simple strain measurements. This is verified by a lab-scale experiment with a uni-axial stress state. Second, measurement data collected during pile driving of a large-diameter pile show that the proposed one-dimensional model, while able to predict the peak stresses, fails to accurately predict the full time history of the measured stress state. In contrast, an advanced model based on shell membrane theory is able to do that, showing that a bi-axial stress state is needed for these type of structures. This requires an extension of the theory for plasticity quantification presented in this paper.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The use of bolted connections between the tower and a support structure of an offshore wind turbine has created the need for a method to detect whether a monopile foundation plastically deforms during the installation procedure. Small permanent deformations are undesirable, not only because they can accelerate fatigue of the structure; but also because they can lead to misalignment between the tower and the foundation. Since direct measurements at the pile head are difficult to perform, a method based on non-collocated strain measurements is highly desirable. This paper proposes such a method. First, a physically non-linear one-dimensional model is proposed, which accounts for wave dispersion, effects that are relevant for large-diameter piles currently used by the industry. The proposed model, combined with an energy balance principle, gives an upper bound for the amount of plastic deformation caused by an impact load based on simple strain measurements. This is verified by a lab-scale experiment with a uni-axial stress state. Second, measurement data collected during pile driving of a large-diameter pile show that the proposed one-dimensional model, while able to predict the peak stresses, fails to accurately predict the full time history of the measured stress state. In contrast, an advanced model based on shell membrane theory is able to do that, showing that a bi-axial stress state is needed for these type of structures. This requires an extension of the theory for plasticity quantification presented in this paper. |
Kalverla, Peter C; Steeneveld, Gert-Jan; Ronda, Reinder J; Holtslag, Albert A M Evaluation of three mainstream numerical weather prediction models with observations from meteorological mast IJmuiden at the North Sea Journal Article Wind Energy, 0 (0), 2018. @article{doi:10.1002/we.2267, title = {Evaluation of three mainstream numerical weather prediction models with observations from meteorological mast IJmuiden at the North Sea}, author = {Peter C. Kalverla and Gert-Jan Steeneveld and Reinder J. Ronda and Albert A. M. Holtslag}, doi = {10.1002/we.2267}, year = {2018}, date = {2018-00-00}, journal = {Wind Energy}, volume = {0}, number = {0}, abstract = {Abstract Numerical weather prediction models play an important role in the field of wind energy, for example, in power forecasting, resource assessment, wind farm (wake) simulations, and load assessment. Continuous evaluation of their performance is crucial for successful operations and further understanding of meteorology for wind energy purposes. However, extensive offshore observations are rarely available. In this paper, we use unique met mast and Lidar observations up to 315 m from met mast “IJmuiden,” located in the North Sea 85 km off the Dutch coast, to evaluate the representation of wind and other relevant variables in three mainstream meteorological models: ECMWF-IFS, HARMONIE-AROME, and WRF-ARW, for a wide range of weather conditions. Overall performance for hub-height wind speed is found to be comparable between the models, with a systematic wind speed bias <0.5 m/s and random wind speed errors (centered RMSE) <2 m/s. However, the model performance differs considerably between cases, with better performance for strong wind regimes and well-mixed wind and potential temperature profiles. Conditions characterized by moderate wind speeds combined with stable stratification, which typically produce substantial wind shear and power fluctuations, lead to the largest misrepresentations in all models.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Abstract Numerical weather prediction models play an important role in the field of wind energy, for example, in power forecasting, resource assessment, wind farm (wake) simulations, and load assessment. Continuous evaluation of their performance is crucial for successful operations and further understanding of meteorology for wind energy purposes. However, extensive offshore observations are rarely available. In this paper, we use unique met mast and Lidar observations up to 315 m from met mast “IJmuiden,” located in the North Sea 85 km off the Dutch coast, to evaluate the representation of wind and other relevant variables in three mainstream meteorological models: ECMWF-IFS, HARMONIE-AROME, and WRF-ARW, for a wide range of weather conditions. Overall performance for hub-height wind speed is found to be comparable between the models, with a systematic wind speed bias <0.5 m/s and random wind speed errors (centered RMSE) <2 m/s. However, the model performance differs considerably between cases, with better performance for strong wind regimes and well-mixed wind and potential temperature profiles. Conditions characterized by moderate wind speeds combined with stable stratification, which typically produce substantial wind shear and power fluctuations, lead to the largest misrepresentations in all models. |
Leontaris, Georgios; Morales-Nápoles, Oswaldo; Wolfert, A R M Probabilistic decision support for offshore wind operations: a Bayesian Network approach to include the dependence of the installation activities Inproceedings Probabilistic Safety Assessment and Management (PSAM 14, September 2018, Los Angeles, CA), 2018. @inproceedings{Leontaris2018a, title = {Probabilistic decision support for offshore wind operations: a Bayesian Network approach to include the dependence of the installation activities}, author = {Georgios Leontaris and Oswaldo Morales-Nápoles and A. R. M. Wolfert}, year = {2018}, date = {2018-00-00}, booktitle = {Probabilistic Safety Assessment and Management (PSAM 14, September 2018, Los Angeles, CA)}, abstract = {Offshore wind operations are logistical challenges and require improved management of the installation and maintenance processes. For this reason, numerous models have been developed concerning different aspects of these operations. Most of these models assume constant durations for the installation or maintenance activities or employ probability distributions to describe the associated uncertainty. However, these two approaches do not take into account the dependence between the activities. This paper proposes a method to describe the dependence between the main installation activities of offshore wind turbines (WTGs) by the use of a non-parametric Bayesian Network (NPBN). To achieve this, different tests were performed and the NPBN was quantified based on real data from a realized project. To illustrate the impact of neglecting the dependence between the activities, a hypothetical case regarding the installation of 150 WTGs was simulated for all three aforementioned approaches (non-dependent: deterministic and probabilistic vs. dependent). It was found that the proposed approach allows for a proper representation of the dependence between the installation activities. Moreover, it can lead to more accurate and reliable estimated installation duration. Hence, this NPBN model can effectively support decision makers in optimizing the work planning of offshore wind processes.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Offshore wind operations are logistical challenges and require improved management of the installation and maintenance processes. For this reason, numerous models have been developed concerning different aspects of these operations. Most of these models assume constant durations for the installation or maintenance activities or employ probability distributions to describe the associated uncertainty. However, these two approaches do not take into account the dependence between the activities. This paper proposes a method to describe the dependence between the main installation activities of offshore wind turbines (WTGs) by the use of a non-parametric Bayesian Network (NPBN). To achieve this, different tests were performed and the NPBN was quantified based on real data from a realized project. To illustrate the impact of neglecting the dependence between the activities, a hypothetical case regarding the installation of 150 WTGs was simulated for all three aforementioned approaches (non-dependent: deterministic and probabilistic vs. dependent). It was found that the proposed approach allows for a proper representation of the dependence between the installation activities. Moreover, it can lead to more accurate and reliable estimated installation duration. Hence, this NPBN model can effectively support decision makers in optimizing the work planning of offshore wind processes. |
Leontaris, Georgios; Morales-Nápoles, Oswaldo ANDURIL — A MATLAB toolbox for ANalysis and Decisions with UnceRtaInty: Learning from expert judgments Journal Article SoftwareX, 7 , pp. 313–317, 2018, ISSN: 2352-7110. @article{LEONTARIS2018b, title = {ANDURIL — A MATLAB toolbox for ANalysis and Decisions with UnceRtaInty: Learning from expert judgments}, author = {Georgios Leontaris and Oswaldo Morales-Nápoles}, url = {http://resolver.tudelft.nl/uuid:d8210946-8d27-4e85-b395-a479f7addd9d}, doi = {10.1016/j.softx.2018.07.001}, issn = {2352-7110}, year = {2018}, date = {2018-00-00}, journal = {SoftwareX}, volume = {7}, pages = {313–317}, abstract = {The Classical model (or Cooke’s model) for elicitation and combination of expert judgments has been used in science and engineering since at least the early 1990’s. The most widely used program for applications of this model is EXCALIBUR. However, its code is not available for practitioners, which limits the accessibility and potential of the method. In this paper, we discuss a MATLAB toolbox (ANDURIL11In order to avoid confusion of the minority of people, who are not familiar with the universe of Lord of the Rings by J.R.R. Tolkien, the authors would like to clarify the inspiration for the name of the developed Matlab toolbox. Andúril was the name of the sword of Aragorn, the son of Arathorn, which was reforged from the shards of Narsil (the sword that was used by Isildur to cut the One Ring from Sauron’s hand). Excalibur is also the name of the legendary sword of King Arthur. Similarly to the sword, the source code of EXCALIBUR software remained accessible only to a few worthy ones. Therefore, the researchers and practitioners could only admire and use the software without being able to further investigate and explore developments of the method. To change this, the existing software had to be “broken to pieces” and then “reforged”. Naturally, the name of the resulting new open-source Matlab toolbox is ANDURIL. Hopefully, this will help in bringing peace to troubled researchers and practitioners of Cooke’s classical model. ) intended to fill in this gap. The software has been tested in a recent real-life application reproducing the results of EXCALIBUR. We discuss different advantages for the users from having the developed source code available for practice.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The Classical model (or Cooke’s model) for elicitation and combination of expert judgments has been used in science and engineering since at least the early 1990’s. The most widely used program for applications of this model is EXCALIBUR. However, its code is not available for practitioners, which limits the accessibility and potential of the method. In this paper, we discuss a MATLAB toolbox (ANDURIL11In order to avoid confusion of the minority of people, who are not familiar with the universe of Lord of the Rings by J.R.R. Tolkien, the authors would like to clarify the inspiration for the name of the developed Matlab toolbox. Andúril was the name of the sword of Aragorn, the son of Arathorn, which was reforged from the shards of Narsil (the sword that was used by Isildur to cut the One Ring from Sauron’s hand). Excalibur is also the name of the legendary sword of King Arthur. Similarly to the sword, the source code of EXCALIBUR software remained accessible only to a few worthy ones. Therefore, the researchers and practitioners could only admire and use the software without being able to further investigate and explore developments of the method. To change this, the existing software had to be “broken to pieces” and then “reforged”. Naturally, the name of the resulting new open-source Matlab toolbox is ANDURIL. Hopefully, this will help in bringing peace to troubled researchers and practitioners of Cooke’s classical model. ) intended to fill in this gap. The software has been tested in a recent real-life application reproducing the results of EXCALIBUR. We discuss different advantages for the users from having the developed source code available for practice. |
2017 |
Kalverla, Peter C; Steeneveld, Gert-Jan; Ronda, Reinder J; Holtslag, Albert A M An observational climatology of anomalous wind events at offshore meteomast IJmuiden (North Sea) Journal Article Journal of Wind Engineering and Industrial Aerodynamics, 165 , pp. 86–99, 2017. @article{Kalverla2017, title = {An observational climatology of anomalous wind events at offshore meteomast IJmuiden (North Sea)}, author = {Peter C. Kalverla and Gert-Jan Steeneveld and Reinder J. Ronda and Albert A. M. Holtslag}, doi = {10.1016/j.jweia.2017.03.008}, year = {2017}, date = {2017-06-00}, journal = {Journal of Wind Engineering and Industrial Aerodynamics}, volume = {165}, pages = {86–99}, abstract = {Uncertainty reduction in offshore wind systems heavily relies on meteorological advances. A detailed characterization of the wind climate at a given site is indispensable for site assessment, and its accurate representation in load assessment models can reduce costs of turbine design and the risk of failure. While regular wind conditions are reasonably described by established methods, some atypical wind conditions are poorly understood and represented, although they contribute substantially to load on turbines. In this study, 4 years of high-quality observations gathered up to 300 m are analyzed to characterize the wind climate at the IJmuiden tower, focusing on these ill-defined conditions. Following a systematic approach, six ‘anomalous wind events’ are identified and described: low-level jets, extreme wind speeds, shear, veer, turbulence and wind ramps. In addition, we identify typical weather conditions that favour their formation. Stable stratification in spring and summer leads to low-level jets (up to 12% of the time) for moderate wind conditions, and to extreme wind shear for stronger wind regimes. Typical wind ramps lead to a change in wind speed of 2 m/s in one hour. The applicability of turbulence intensity as a measure of turbulence and gusts is found to be questionable.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Uncertainty reduction in offshore wind systems heavily relies on meteorological advances. A detailed characterization of the wind climate at a given site is indispensable for site assessment, and its accurate representation in load assessment models can reduce costs of turbine design and the risk of failure. While regular wind conditions are reasonably described by established methods, some atypical wind conditions are poorly understood and represented, although they contribute substantially to load on turbines. In this study, 4 years of high-quality observations gathered up to 300 m are analyzed to characterize the wind climate at the IJmuiden tower, focusing on these ill-defined conditions. Following a systematic approach, six ‘anomalous wind events’ are identified and described: low-level jets, extreme wind speeds, shear, veer, turbulence and wind ramps. In addition, we identify typical weather conditions that favour their formation. Stable stratification in spring and summer leads to low-level jets (up to 12% of the time) for moderate wind conditions, and to extreme wind shear for stronger wind regimes. Typical wind ramps lead to a change in wind speed of 2 m/s in one hour. The applicability of turbulence intensity as a measure of turbulence and gusts is found to be questionable. |
van den Bos, Laurent M M; Koren, Barry; Dwight, Richard P Non-intrusive uncertainty quantification using reduced cubature rules Journal Article Journal of Computational Physics, 332 , pp. 418–445, 2017. @article{vandenBos2017, title = {Non-intrusive uncertainty quantification using reduced cubature rules}, author = {Laurent M. M. van den Bos and Barry Koren and Richard P. Dwight}, url = {https://ir.cwi.nl/pub/25311}, doi = {10.1016/j.jcp.2016.12.011}, year = {2017}, date = {2017-03-00}, journal = {Journal of Computational Physics}, volume = {332}, pages = {418--445}, abstract = {For the purpose of uncertainty quantification with collocation, a method is proposed for generating families of one-dimensional nested quadrature rules with positive weights and symmetric nodes. This is achieved through a reduction procedure: we start with a high-degree quadrature rule with positive weights and remove nodes while preserving symmetry and positivity. This is shown to be always possible, by a lemma depending primarily on Carathéodory's theorem. The resulting one-dimensional rules can be used within a Smolyak procedure to produce sparse multi-dimensional rules, but weight positivity is lost then. As a remedy, the reduction procedure is directly applied to multi-dimensional tensor-product cubature rules. This allows to produce a family of sparse cubature rules with positive weights, competitive with Smolyak rules. Finally the positivity constraint is relaxed to allow more flexibility in the removal of nodes. This gives a second family of sparse cubature rules, in which iteratively as many nodes as possible are removed. The new quadrature and cubature rules are applied to test problems from mathematics and fluid dynamics. Their performance is compared with that of the tensor-product and standard Clenshaw–Curtis Smolyak cubature rule.}, keywords = {}, pubstate = {published}, tppubtype = {article} } For the purpose of uncertainty quantification with collocation, a method is proposed for generating families of one-dimensional nested quadrature rules with positive weights and symmetric nodes. This is achieved through a reduction procedure: we start with a high-degree quadrature rule with positive weights and remove nodes while preserving symmetry and positivity. This is shown to be always possible, by a lemma depending primarily on Carathéodory's theorem. The resulting one-dimensional rules can be used within a Smolyak procedure to produce sparse multi-dimensional rules, but weight positivity is lost then. As a remedy, the reduction procedure is directly applied to multi-dimensional tensor-product cubature rules. This allows to produce a family of sparse cubature rules with positive weights, competitive with Smolyak rules. Finally the positivity constraint is relaxed to allow more flexibility in the removal of nodes. This gives a second family of sparse cubature rules, in which iteratively as many nodes as possible are removed. The new quadrature and cubature rules are applied to test problems from mathematics and fluid dynamics. Their performance is compared with that of the tensor-product and standard Clenshaw–Curtis Smolyak cubature rule. |
Beltman, René; Anthonissen, Martijn J H; Koren, Barry Mimetic Staggered Discretization of Incompressible Navier–Stokes for Barycentric Dual Mesh Inproceedings Cancès, Clément; Omnes, Pascal (Ed.): Finite Volumes for Complex Applications VIII - Hyperbolic, Elliptic and Parabolic Problems, pp. 467–475, Springer, 2017. @inproceedings{Beltman2017, title = {Mimetic Staggered Discretization of Incompressible Navier–Stokes for Barycentric Dual Mesh}, author = {René Beltman and Martijn J. H. Anthonissen and Barry Koren}, editor = {Clément Cancès and Pascal Omnes}, url = {http://repository.tue.nl/5149ebcb-983a-4b1b-8b31-92360e865705}, doi = {10.1007/978-3-319-57394-6_49}, year = {2017}, date = {2017-01-01}, booktitle = {Finite Volumes for Complex Applications VIII - Hyperbolic, Elliptic and Parabolic Problems}, pages = {467–475}, publisher = {Springer}, abstract = {A staggered discretization of the incompressible Navier–Stokes equations is presented for polyhedral non orthogonal nonsmooth meshes admitting a barycentric dual mesh. The discretization is constructed by using concepts of discrete exterior calculus. The method strictly conserves mass, momentum and energy in the absence of viscosity.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } A staggered discretization of the incompressible Navier–Stokes equations is presented for polyhedral non orthogonal nonsmooth meshes admitting a barycentric dual mesh. The discretization is constructed by using concepts of discrete exterior calculus. The method strictly conserves mass, momentum and energy in the absence of viscosity. |
Leontaris, Georgios; Morales-Nápoles, Oswaldo; Wolfert, A R M Planning cable installation activities for offshore wind farms including risk of supply delays Inproceedings Walls, Lesley; Revie, Matthew; Bedford, Tim (Ed.): Risk, Reliability and Safety: Innovating Theory and Practice: Proceedings of ESREL 2016 (Glasgow, Scotland, 25-29 September 2016), pp. 660–666, CRC Press, 2017, ISBN: 9781138029972. @inproceedings{Leontaris2016b, title = {Planning cable installation activities for offshore wind farms including risk of supply delays}, author = {Georgios Leontaris and Oswaldo Morales-Nápoles and A. R. M. Wolfert}, editor = {Lesley Walls and Matthew Revie and Tim Bedford}, isbn = {9781138029972}, year = {2017}, date = {2017-01-01}, booktitle = {Risk, Reliability and Safety: Innovating Theory and Practice: Proceedings of ESREL 2016 (Glasgow, Scotland, 25-29 September 2016)}, pages = {660–666}, publisher = {CRC Press}, abstract = {Installation of Offshore Wind Farms (OWF) is subject to various uncertainties which concern environmental conditions, failures and availability of components. These should be taken into account when estimating the duration in order to efficiently plan the construction. This paper presents a methodology to include possible disturbances on the supply of cable, in order to extend a developed decision support tool for the cable installation. A realistic test case was simulated for two scenarios, when uncertainty regarding the availability of the cable is neglected and when it is modelled using expert assessments. It was found that the proposed methodology can help professionals and/or researchers in investigating cost-effective alternatives concerning the assets that are used in the installation. Concluding, it is suggested to apply this methodology to the supply chain of the entire OWF installation as well as use structured expert judgement in order to improve the quality of resulting estimates.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Installation of Offshore Wind Farms (OWF) is subject to various uncertainties which concern environmental conditions, failures and availability of components. These should be taken into account when estimating the duration in order to efficiently plan the construction. This paper presents a methodology to include possible disturbances on the supply of cable, in order to extend a developed decision support tool for the cable installation. A realistic test case was simulated for two scenarios, when uncertainty regarding the availability of the cable is neglected and when it is modelled using expert assessments. It was found that the proposed methodology can help professionals and/or researchers in investigating cost-effective alternatives concerning the assets that are used in the installation. Concluding, it is suggested to apply this methodology to the supply chain of the entire OWF installation as well as use structured expert judgement in order to improve the quality of resulting estimates. |
Quaeghebeur, Erik; Sanchez Perez-Moreno, Sebastian ; Zaaijer, Michiel OWFgraph: A graph database for the offshore wind farm domain Inproceedings Sørensen, J N (Ed.): WESC 2017: Wind Energy Science Conference (Book of Abstracts), pp. 116, Technical University of Denmark Lyngby, Denmark, 2017. @inproceedings{Quaeghebeur2017, title = {OWFgraph: A graph database for the offshore wind farm domain}, author = {Erik Quaeghebeur and Sebastian {Sanchez Perez-Moreno} and Michiel Zaaijer}, editor = {J. N. Sørensen}, url = {http://resolver.tudelft.nl/uuid:e3d75a9d-3e76-4721-b59f-fd782aaa29aa http://wesc2017.org/book%20of%20abstracts.html }, year = {2017}, date = {2017-00-00}, booktitle = {WESC 2017: Wind Energy Science Conference (Book of Abstracts)}, pages = {116}, address = {Lyngby, Denmark}, organization = {Technical University of Denmark}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Meijers, Peter C; Tsouvalas, Apostolos; Metrikine, Andrei V The effect of stress wave dispersion on the drivability analysis of large-diameter monopiles Inproceedings Vestroni, F; Gattulli, V; Romeo, F (Ed.): EURODYN 2017: X International Conference on Structural Dynamics, pp. 2390–2395, Elsevier, 2017. @inproceedings{Meijers2017b, title = {The effect of stress wave dispersion on the drivability analysis of large-diameter monopiles}, author = {Peter C. Meijers and Apostolos Tsouvalas and Andrei V. Metrikine}, editor = {F. Vestroni and V. Gattulli and F. Romeo}, url = {http://resolver.tudelft.nl/uuid:e27fb9ab-2a9f-4245-a6e8-ead0bc55f030}, doi = {10.1016/j.proeng.2017.09.272}, year = {2017}, date = {2017-00-00}, booktitle = {EURODYN 2017: X International Conference on Structural Dynamics}, journal = {Procedia Engineering}, volume = {199}, pages = {2390--2395}, publisher = {Elsevier}, series = {Procedia Engineering}, abstract = {Due to the increasing need for energy from renewable resources, a large number of offshore wind farms are planned to be constructed in the near future. Despite the plethora of available foundation concepts for offshore wind turbines, the monopile foundation is the most widely adopted concept in practice. To predict the installation process for a monopile a so-called drivability study is performed. Such a study allows one to decide on a number of key parameters for the installation process, such as, the appropriate size of the hydraulic hammer, the number of hammer blows and energy input needed to reach the final penetration depth, and the induced stresses in the system. The latter is important for the prediction of the fatigue life of the pile. Currently, drivability studies are based on one-dimensional wave equation models as first proposed by Smith in the 1950s. These models are valid as long as the diameter of the pile is small compared to the excited wavelengths in the structure due to the hammer impact. For large-diameter monopiles that are currently being used in the offshore wind industry, the latter condition is not met and the effect of stress wave dispersion can no longer be neglected. In this paper the classical wave equation model is amended by an extra term which accounts for the lateral inertia of the cross-section, resulting in the so-called Rayleigh-Love rod theory. With this new model, a parametric study is performed in which the effect of stress wave dispersion on the induced stresses and the number of hammer blows needed to reach the final penetration depth are assessed. A comparison with the results obtained from the classical model is also included in order to define the applicability range of the models. It is shown that the effect of stress wave dispersion can not be neglected for a drivability study of large-diameter monopiles.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Due to the increasing need for energy from renewable resources, a large number of offshore wind farms are planned to be constructed in the near future. Despite the plethora of available foundation concepts for offshore wind turbines, the monopile foundation is the most widely adopted concept in practice. To predict the installation process for a monopile a so-called drivability study is performed. Such a study allows one to decide on a number of key parameters for the installation process, such as, the appropriate size of the hydraulic hammer, the number of hammer blows and energy input needed to reach the final penetration depth, and the induced stresses in the system. The latter is important for the prediction of the fatigue life of the pile. Currently, drivability studies are based on one-dimensional wave equation models as first proposed by Smith in the 1950s. These models are valid as long as the diameter of the pile is small compared to the excited wavelengths in the structure due to the hammer impact. For large-diameter monopiles that are currently being used in the offshore wind industry, the latter condition is not met and the effect of stress wave dispersion can no longer be neglected. In this paper the classical wave equation model is amended by an extra term which accounts for the lateral inertia of the cross-section, resulting in the so-called Rayleigh-Love rod theory. With this new model, a parametric study is performed in which the effect of stress wave dispersion on the induced stresses and the number of hammer blows needed to reach the final penetration depth are assessed. A comparison with the results obtained from the classical model is also included in order to define the applicability range of the models. It is shown that the effect of stress wave dispersion can not be neglected for a drivability study of large-diameter monopiles. |
van den Bos, Laurent M M; Sanderse, Benjamin Uncertainty quantification for wind energy applications Technical Report Centrum Wiskunde & Informatica (SC-1701), 2017. @techreport{vandenBos2017b, title = {Uncertainty quantification for wind energy applications}, author = {Laurent M. M. van den Bos and Benjamin Sanderse}, url = {https://ir.cwi.nl/pub/26650/}, year = {2017}, date = {2017-00-00}, number = {SC-1701}, institution = {Centrum Wiskunde & Informatica}, abstract = {Uncertainties are omni-present in wind energy applications, both in external conditions (such as wind and waves) as well as in the models that are used to predict key quantities such as costs, energy yield, and fatigue loads. This report summarizes and reviews the application of uncertainty quantification techniques to wind energy problems. In the wind industry, including uncertainties in predictions has classically been done by using Monte Carlo methods. Recently, more advanced methods have been considered (e.g. polynomial chaos expansion, stochastic collocation, and Gaussian process regression), which are based on smartly sampling the model (e.g. a complex aerodynamic blade model). These methods generally have a greater efficiency compared to Monte Carlo (depending on model properties) and additionally yield computationally cheap surrogate models. Furthermore, surrogate models purely based on data (e.g. via proper orthogonal decomposition) have received significant interest, especially for the representation of turbulent wind turbine wakes. Both model-driven and data-driven surrogate models play a crucial role in making control and optimization studies feasible. In the near future, we expect that recent trends in uncertainty quantification, namely Bayesian model calibration and optimization under uncertainty, will become increasingly popular in wind energy applications.}, keywords = {}, pubstate = {published}, tppubtype = {techreport} } Uncertainties are omni-present in wind energy applications, both in external conditions (such as wind and waves) as well as in the models that are used to predict key quantities such as costs, energy yield, and fatigue loads. This report summarizes and reviews the application of uncertainty quantification techniques to wind energy problems. In the wind industry, including uncertainties in predictions has classically been done by using Monte Carlo methods. Recently, more advanced methods have been considered (e.g. polynomial chaos expansion, stochastic collocation, and Gaussian process regression), which are based on smartly sampling the model (e.g. a complex aerodynamic blade model). These methods generally have a greater efficiency compared to Monte Carlo (depending on model properties) and additionally yield computationally cheap surrogate models. Furthermore, surrogate models purely based on data (e.g. via proper orthogonal decomposition) have received significant interest, especially for the representation of turbulent wind turbine wakes. Both model-driven and data-driven surrogate models play a crucial role in making control and optimization studies feasible. In the near future, we expect that recent trends in uncertainty quantification, namely Bayesian model calibration and optimization under uncertainty, will become increasingly popular in wind energy applications. |
2016 |
Leontaris, Georgios; Morales-Nápoles, Oswaldo; Wolfert, A R M Ocean Engineering, 125 (Supplement C), pp. 328–341, 2016, ISSN: 0029-8018. @article{LEONTARIS2016a, title = {Probabilistic scheduling of offshore operations using copula based environmental time series – An application for cable installation management for offshore wind farms}, author = {Georgios Leontaris and Oswaldo Morales-Nápoles and A. R. M. Wolfert}, url = {http://resolver.tudelft.nl/uuid:f361f293-66a4-426f-9bf6-7610361fc8d1}, doi = {10.1016/j.oceaneng.2016.08.029}, issn = {0029-8018}, year = {2016}, date = {2016-00-00}, journal = {Ocean Engineering}, volume = {125}, number = {Supplement C}, pages = {328–341}, abstract = {Abstract There are numerous uncertainties that impact offshore operations. However, environmental uncertainties concerning variables such as wave height and wind speed are crucial because these may affect installation and maintenance operations with potential delays and financial consequences. In order to include these uncertainties into the duration estimation, adequate tools should be developed to simulate an installation scenario for a large number of historical environmental data. Data regarding environmental time series are usually scarce and limited, therefore they should be modelled. Since the environmental variables are in reality dependent, we propose a probabilistic method for their construction using copulas. To demonstrate the effectiveness of this method compared to the cases where observed or independently constructed environmental time series are used, a realistic cable installation scenario for an offshore wind farm was simulated. It was found that the proposed method should be followed to acquire more reliable and accurate estimation of the installation's duration.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Abstract There are numerous uncertainties that impact offshore operations. However, environmental uncertainties concerning variables such as wave height and wind speed are crucial because these may affect installation and maintenance operations with potential delays and financial consequences. In order to include these uncertainties into the duration estimation, adequate tools should be developed to simulate an installation scenario for a large number of historical environmental data. Data regarding environmental time series are usually scarce and limited, therefore they should be modelled. Since the environmental variables are in reality dependent, we propose a probabilistic method for their construction using copulas. To demonstrate the effectiveness of this method compared to the cases where observed or independently constructed environmental time series are used, a realistic cable installation scenario for an offshore wind farm was simulated. It was found that the proposed method should be followed to acquire more reliable and accurate estimation of the installation's duration. |