Structures and Integrity
De-risking critical aspects for concept selection and enable cost-effective design and fabrication of support structures through reduced uncertainty and application of novel methods.
Work Package 1 is led by Vigdis Olden from SINTEF Industry and Ana Page from NGI. Its focus areas are environmental loads, limetime and performance, design and fabrication.
Anticipated results:
- Improved methods for prediction of interactions between environmental loads, soil and structure.
- Decision support platform for lifetime prediction
- Cost-effective serial production of substructures
Hypothesis
Fundamental understanding and accurate models of load, response, and degradation will be a key enabler for design, digitalised operation and lifetime prediction of support structures.
Methodology and research tasks
Environmental loads and response modelling
Hydrodynamic load models, site investigation methodologies and soil structure interaction and integrity.
Integrated design assessment and optimization
Multi-scale aero-hydro-servo-elastic modelling and reliability-based structural design optimisation.
Lifetime, performance, and integrity
Precision in degradation models. This is linked to Work Package 4.
Fabrication
Cost-effective mass production by automation and optimised and light-weight design and materials for secondary structures. Linked to Work Package 2 and 3.
Ambition for innovations
Improved models
Improved models for reliable predictions and better
understanding of interactions between environmental loads, soil, and structure. Accurate knowledge of site conditions and extreme environmental loads is insufficiently applied in early-phase design.
Decision support platform
Decision support platform for life prediction based on advanced model and condition monitoring. Limited understanding and modelling capabilities for material damage during operation.
Cost-effective production
Cost-effective serial production of offshore sub-structures, including a high degree of automation. Manual welding and coating are bottlenecks and cost drivers in fabrication.
Innovations
Main results from 2023
Establishing a sand database
For geotechnical engineering, databases of soil test results have played an important role in the development of offshore energy infrastructure. The availability of these databases provides valuable tools for designers, instilling confidence in the process of determining geotechnical design profiles.
A database, comprising all the physical geotechnical tests conducted at NGI on offshore sands, has been established. This database includes index properties and advanced laboratory test results for sand, which will be made accessible through an Application Programming Interface (API). However, with the growing emphasis on achieving net-zero greenhouse gas emissions, there has been increased pressure on offshore wind supply chains, leading to concerns about the capacity of soil investigation laboratories.
To address this challenge, the Level Set Discrete Element Method (LS-DEM) has emerged as a promising technology for virtual simulations of soil element scale tests. The virtual laboratory employs LS-DEM, a particle-based model that enables the simulation of sands with varying grain size distributions and grain shapes. This tool can be seamlessly integrated with data from the database and used to predict soil behaviour, aiding in decision-making during the early stages of a project before site investigations are conducted. Dogger Bank sand, typically found in the North Sea region, alongside Hokksund sand, Karlsruhe sand, Øysand and Ottawa sand were selected for this study. Three different approaches were used to define grain morphology and create 3D level set avatars: clones based on material descriptions by laboratory technicians, clones based on 2D images with corresponding image processing, and full 3D CT scans with subsequent image processing.
The objective is to subject virtual samples created by these three methods to various stress paths and compare the results to physical laboratory test results available for Dogger Bank sand. A paper detailing the work conducted in 2023 has been submitted to the ECSMGE 2024 conference, and work on comparing virtual sampling test results to physical laboratory results is currently underway. These findings will eventually be compiled into a journal publication.
Nonlinear hydrodynamic loads
This year, the research focused on several activities. First, a poster presentation and paper on the numerical study examining the nonlinear and viscous loads on the INO WINDMOOR floater in regular waves was prepared and finalised for EERA DeepWind 2023. This study applied computational fluid dynamic (CFD) techniques to investigate the effect of the free surface on the local drag forces. The results from the CFD study provided a good basis for further analysis where drag coefficients for the Morison load model are extracted from the local nonlinear forces. These coefficients can be applied in more efficient global simulation models. This work will continue in 2024 and is expected to lead to another publication. In addition to the above study, a separate work on testing and validating full quadratic transfer functions (QTFs) for wave drift forces for conditions with steep waves and current has been carried out. A summary of the results is currently being prepared in is expected to be published in 2024.
Uncertainty analysis of hydrodynamic
experiments of floating wind turbines
The first stage of the uncertainty analysis of a hydrodynamic experiment of the 12MW INO WINDMOOR semisubmersible floating wind turbine was presented at the OMAE2023 conference. This work provided valuable insights into the uncertainties associated with the mooring system modelling and its impact on hydrodynamic testing of floating wind turbines. The next stage of the study will focus on a larger set of uncertainty sources – on how it impacts the model test result. A Monte Carlo method is applied, and a convergence study was carried out in order to determine how many sample simulations are needed to have confidence in the analysis. The work is still ongoing and will continue in 2024 before a full paper on the topic is published.
Materials integrity in the drive train
The research paper ‘Material Failures on Offshore Wind Turbines’, which includes a literature review and future research outlook, was presented at the Annual Innovation Forum in December, highlighting key findings and insights. Central aspects are that “the lifetime of gears in wind turbines is affected by the operation conditions, the preceding materials selection and manufacturing processes of the gear”. New technologies for increasing steel quality and enhance fatigue properties are receiving much attention. More knowledge and technologies are also needed for surface treatment processes to optimise the compressive residual stress field at the contact surface, while considering the contact stresses during operation and reduce shear stresses in the material.
ANEO have proposed to make a failed bearing available for material analyses within FME NorthWind, and we are looking forward to investigating this component in 2024.
How to ensure robust laser-arc welding process
and high-quality welds?
Laser-arc hybrid welding (LAHW) is an efficient and promising welding technology for manufacturing offshore wind substructures. In 2022, we performed laboratory testing and demonstrated that LAHW may offer 5-24 times higher productivity, resulting in a significant cost reduction for wind turbines. In 2023, we addressed another important issue related to LAHW: how to ensure good weld quality. Despite LAHW showing high efficiency, it is also important to achieve defect-free welds and fulfil strict material and structural integrity requirement due to offshore wind’s harsh service conditions.
We tested the acoustic emission (AE) technique for LAHW process monitoring in our laboratory. After welding, an X-ray Computed Tomography (μCT) scan of the weld was performed, and a full picture of defects was generated. Post-processing of AE sound signals was performed, and a first attempt for correlating defects and signals was made. Unfortunately, the result was not as good as we expected. However, this test has shown the great potential of applying cheap AE sensors for fast defect detection during welding. This test also triggered an exciting discussion between WP1 and WP4 for possible collaboration related to developing a digital twin for anomaly diagnosis in manufacturing of offshore wind substructures. The development will be continued in 2024.
Publications and PhDs
A total of four peer-reviewed publications were produced within this work package in 2023. See the Results page for the list of all publications. Work package 1 is involved in the supervision and guidance of two PhD candidates (Veronica Liverud Krathe and Afolarinwa David Oyegbile) and one associated PhD (Øyvind Torgersrud). See the Education page for details.
Previous results
Numerical simulation of cone penetration tests in silty sands
The offshore wind industry relies heavily on the Cone Penetration Test (CPTu) to measure the in-situ soil conditions and characterise large areas. However, the accuracy of CPT data interpretation depends on generic correlations. While these correlations work well for sands with limited fine particles, they are less reliable for soils with higher fines content, like silty sands, partially due to the lack of understanding on how the presence of fine particles affects the measured CPTu response. To address this issue and analyse the influence of fines and drainage on the tip resistance and developed pore pressure in CPTu, researchers in WP1 have combined two known numerical techniques: large-deformation finite element analyses with the “zipper technique” and coupled consolidation analyses.
The results from applying this technique have been compared against CPTu measurements from two locations at an offshore wind farm site, providing good agreement, and indicating that this technique may help improving the understanding of the effect of fines content on CPTu response.
Establishing a sand database
The limited capacity of site investigation vessels and laboratories to process and test soil samples remains a significant bottleneck in the offshore wind industry, causing delays in the deployment of new offshore wind sites. To address this challenge, researchers in WP1 have taken the initiative to compile a database of typical sand properties. The primary objective of this database is to provide access to data on typical soil behaviour for offshore sites.
This will facilitate the development of offshore wind projects, especially in cases where there are no site-specific laboratory test results available. Our approach involved a combination of existing laboratory tests and newly generated virtual
sand tests to enhance the coverage of the database. To conduct virtual tests, we used the recently developed Level-Set Discrete Element Method (LS-DEM). This method models the interactions between individual sand grains in an assembly of grains, generating a continuum stress-strain relationship similar to that derived from traditional laboratory tests (see illustration).
Comparison of the stress-strain curves from a virtual LS-DEM test and a triaxial laboratory test for Øysand sand.
Uncertainty analysis of hydrodynamic experiments of floating wind turbines
A numerical study was conducted to evaluate uncertainties in the mooring system of the 12MW INO WINDMOOR semisubmersible floating wind turbine based on the KPN WINDMOOR hydrodynamic experiment. The study investigated uncertainties in the mooring system and their effect on outputs such as mooring line tension and platform responses. Two different approaches were used for the uncertainty analysis: the Taylor Series Method and the Monte Carlo Method. The results will be presented at OMAE2023 in June 2023 and provide valuable insights into the uncertainties associated with the mooring system modelling and its impact on hydrodynamic testing of floating wind turbines.
Materials integrity in the drive train
A state-of-the-art report on design, operation, failures and lifetime of main bearings and gears has been written. Considerable effort to get hold of failed drive train components and relevant material for study of failures and laboratory rolling fatigue experiments have taken place without success. We hope that this can be solved in collaboration with the industry partners in 2023.
How efficient is laser-arc hybrid welding?
Laser-arc hybrid welding (LAHW) is an efficient and promising welding technology for manufacturing offshore wind substructures. Inspired by the quest for finding efficient fabrication methods, we performed an initial comparison study in a laboratory to show how efficient LAHW can be compared to conventional welding methods. We found that LAHW may offer 5-24 times higher productivity than conventional arc welding (see the figure above) with half the energy consumption.
Despite the economic and productivity advantages of LAHW, there are challenges with the quality of welds, e.g., cracking at weld centreline and poor microstructure. Improvement of the stability of LAHW process is the key to achieve defect-free welds. Therefore, we have started to investigate the potential of using acoustic emission (AE) technique for LAHW process monitoring. A state-of-the-art review has been performed. The first testing will start in 2023.
Establishing a sand database
- The foundation concept selection for wind turbines often relies on limited information on the soil conditions and some generic correlations. Having access to data and better correlations can facilitate the development of offshore wind projects and de-risk foundation concepts that are sensitive to soil conditions.
Laser welding of thick-walled steel plates
- For the accelerated deployment of large-scale offshore wind, we need cost-effective, sustainable, and mass-produced substructures. The ability to efficiently perform welding of thick steel plates resulting in a good quality weld with respect to full penetration and material properties is central. A full-penetration, defect-free weld has been achieved.
Identification and description of user cases
In close cooperation with the industry partners, three industrial user cases have been identified.
