Metocean Research Articles

Wind field reconstruction using nacelle based lidar measurements for floating wind turbines

In this work we investigate the influence of floater motions on nacelle based lidar wind speed measurements. The analysis is focused on wind field characteristics which are most relevant for performance analysis, namely rotor effective wind speed, shear and turbulence intensity. A numerical approach, coupling the publicly available in-house lidar simulation framework ViConDAR (Virtual Constrained turbulence and liDAR measurements) with an aeroelastic floating offshore wind turbine simulation is employed. Synthetic turbulent wind fields are generated with the open source turbulence generator TurbSim. The dynamics of the floating offshore wind turbine are simulated in the aeroelastic simulation code OpenFAST. Turbine dynamics and the corresponding synthetic wind field are then passed to the lidar simulation module of ViConDAR, which has been adapted for the consideration of turbine dynamics in all six degrees of freedom to simulate the lidar measurements under influence of motion. The simulation framework is demonstrated in a case study, simulating a lidar system on the nacelle of the IEA 15 MW turbine in combination with the WindCrete floater concept. Two different realistic lidar patterns are investigated under different metocean conditions. Different motion cases are examined individually and combined to evaluate the influence of rotational and translational degrees of freedom. Results show an increase of mean absolute error between lidar estimated and full wind field rotor effective wind speed of 0 to 25% depending on the environmental conditions. Observed overestimation of mean rotor effective wind speed was found to be in the region of up to 1%.

Study of the influence of met-ocean data in fatigue loads calculations of a floating offshore wind turbine

The design of a floating offshore wind turbine (FOWT) requires the computation of loads with time domain simulations that depend heavily on the metocean conditions of the site. Guidelines, such as IEC 61400-3, recommend the use of scatter tables with the combined probability of the significant wave height (Hs) and peak period (Tp) as a function of the mean wind speed. Nevertheless, at early engineering phases it may be difficult to obtain the adequate environmental wind-wave data where both parameters are correlated for the site of interest. The main objective of this work is to assess the influence of neglecting or not the correlation between wind speed and wave data in the resulting fatigue Damage Equivalent Load (DEL) for different FOWT components. The analysis is performed through the comparison of the fatigue loads obtained for correlated and non correlated wind-wave data. Results show that fatigue loads are very similar for both data sets at the rotor loads, and present slight differences, at mooring tensions and tower base. The use of a simpler metocean data, where wave height and period occurrences are given independently of the wind speed, might be a sufficient solution for initial fatigue load estimations.

A surrogate model of offshore wind farm support structures for wind farm design and financial valuation

In the preliminary phases of offshore wind farm development, very little information on project design are available for supporting financial valuation and site design. In this work, we develop a surrogate model for offshore wind support structure mass for input to techno-economic analysis that is based on a small set of input parameters. Using reference turbines and a broad set of met-ocean conditions, a large design spaced is developed from which a sampling of conditions is used to estimate the dimensions and mass of monopile support structures. The results are parameterized using statistical methods to create a functional model of costs relative to high-level site and technical inputs. To preserve the transparency of the model input-output relationships, a statistical surrogate model is used based on quadratic functions of the inputs. Overall, the rated power and rotor diameter of the turbine has the greatest influence on the mass, followed by the specific power. The water depth was the next most important environmental parameter, followed by wave period. The full surrogate model captures 98.5% of the variance of the monopile mass as a function of the above inputs. We present results related to monopile foundations, but the methodology is flexible and can be applied also in the case of other types of support structures.

Experimental Analysis of the Wake Meandering of a Floating Wind Turbine under Imposed Surge Motion

With the development of floating wind farms, the understanding of the far wake becomes of utmost importance because it cannot only be based on the experience acquired on fixed offshore turbine as the range of motion of the floater due to met-ocean conditions is in the same order of magnitude compared to the energetic turbulent scales of the atmosphere and to the characteristic scales of the wake. The objective of this wind-tunnel experiment is to analyse the behaviour of the wake center of a floating turbine subject to imposed surge motion. The experiment is carried out in the LHEEA wind tunnel where a 1/500 scale wind turbine model is immersed in a realistic offshore atmospheric-boundary layer. The model is actuated in surge by a linear motor able to reproduced idealised second order surge motion of a floating platform (mainly due to the waves) with a realistic amplitude and range of frequencies. Stereoscopic particle image velocimetry measurements are performed at two planes normal to the free-stream at distances of 4.6 and 8.1 diameters downstream of the turbine. Instantaneous velocity fields acquired at 14.1 Hz are individually analysed by convolution to find the location of the wake center. Results show that the wake center spreading is largely influenced by the downstream distance and only slightly affected by the surge motion. However, when analysing the wake position time series by a power spectral density, the signature of the surge motion becomes very clear for both downstream distances. These findings tell that the far wake (at least up to 8.1 diameters downstream) of a floating wind turbine has a “memory” of the motion in its frequency content. When extrapolating to a floating wind farm, this result suggests that a turbine inside a floating wind farm will be immersed in a flow including a significant dynamic signature in the range of its own motion.

Assessment of Current Downtime Analysis

Abstract: The approach used to calculate downtime for complicated projects is consistent with the literature: simulation. Researchers evaluate current downtime study on Shipping Corporation in this report. As a result, the existing approach for determining downtime for complex projects is untrustworthy; in fact, due to the significant simulation uncertainty, it is highly unsafe to utilise current methodology to make predictions on downtime distribution. We advocate utilising a stochastic metocean condition generator in the simulation to reduce simulation uncertainty. These uncertainties, however, may be managed by the model used, & they can also be quantified. Finally, a comparison1should be drawn between simulation uncertainty in HADDOCK & combined model, parametric, & simulation uncertainty in a stochastic model simulation. Keywords: Marine operation, downtime analysis, Assessment advantages and limitations

The sensitivity to metocean data on using data-driven methods for a Valemax vessel speed prediction

Accurate ship speed prediction is not only important for economic benefits and emission reduction, but also plays a vital role in making satisfactory routing plans. To investigate a speed prediction method of the largest bulk carriers ever constructed, we firstly compare several widely-used data-driven methods and then pick out the most appropriate model for the prediction task. Being as the input attributes of the data-driven models, the metocean data are collected from forecasting products and carefully cleaned up by data pre-processing steps. However, we have observed from literatures that the data for training are mostly collected from observational systems, such as the noon-reports, sensor measurements, on-board monitoring systems and reanalysis products, which may have certain differences with the data from forecasting systems. To figure out whether the data inconsistency problem will affect the model effectiveness on prediction, exhaustive experiments of a sensitivity study are carried out by utilizing observational data for training models while testing the prediction performance with forecasting data.

Predictive Maintenance of Floating Offshore Wind Turbine Mooring Lines using Deep Neural Networks

Abstract The recent massive deployment of onshore wind farms has caused controversy to arise mainly around the issues of land occupation, noise and visual pollution and impact on wildlife. Fixed offshore turbines, albeit beneficial in those aspects, become economically unfeasible when installed far away from coastlines. The possibility of installing floating offshore wind turbines is currently hindered by their excessive operation and maintenance costs. We have developed a comprehensive model to help companies plan their operations in advance by detecting failure in mooring lines in almost real time using supervised deep learning techniques. Given the lack of real data, we have coupled numerical methods and OpenFAST simulations to build a dataset containing the displacements and rotations of a turbine’s floating platform across all directions. These time series and their corresponding frequency spectra are used to obtain a set of key statistical parameters, including means and standard deviations, peak frequencies, and several relevant momenta. We have designed and trained a Deep Neural Network to understand and distinguish amongst a series of common failure modes for mooring lines considering a range of metocean and structural conditions. We have obtained promising results when monitoring severe changes in the line’s mass and damping using short time spans, achieving a 95.7% validation accuracy when detecting severe biofouling failure.

The Skill Assessment of Weather and Research Forecasting and WAVEWATCH-III Models During Recent Meteotsunami Event in the Persian Gulf

This study aims to use a fully realistic high-resolution mesoscale atmospheric and wave model to reproduce met-ocean conditions during a meteotsunami in the Persian Gulf. The atmospheric simulations were performed with the Weather and Research Forecasting (WRF) model by varying planetary boundary layer, microphysics, cumulus, and radiations parameterizations. The atmospheric results were compared to the meteorological observations (e.g., air pressure and wind speed) from the coastal and island synoptic and buoy stations of the nearest area to the meteotsunami event. The results show that using Mellor-Yamada-Nakanishi-Niino (MYNN) scheme for planetary boundary and surface layer had the best performance for stations over the water, whereas applying Mellor-Yamada-Janjic scheme for planetary boundary and Eta similarity surface layer had the best performance for stations over the land. For wave simulations, the WAVEWATCH-III model was employed with the well-known WAM-Cycle4 formulation and a more recent ST6 package. Six WRF experiments and ERA5 wind data were used to force the wave models. The new error parameter was introduced to identify the optimum wind data for wave simulation. EXP4 configuration which uses the MYNN scheme for planetary boundary and surface layer was led to minimum error, while ERA5 severely underestimated Hs and Tp parameters. For the first time, the Gaussian Quadrature Method (GQM) was implemented in the WAVEWATCH-III model and combined with a depth scale to be used in the Persian Gulf. This method is more accurate for non-linear wave-wave interaction than the default Discrete Interaction Approximation (DIA) method. Lower coefficients for dissipation term were required for GQM and the resulted bulk wave parameters were improved compared to the DIA method. The calibrated ST6 formulation with GQM resulted in a more realistic prediction of wave spectrum than the default settings of the WAVEWATCH-III.

Experimental study of an ocean surface cleaning system

This paper presents a novel design of an ocean surface cleaning system (OSCS), which can be operated at various sea sites. The system weathervanes around the mooring system, so that the marine debris can be effectively concentrated by the OSCS. In order to verify the technical feasibility and efficiency of this novel OSCS, a series of experiments in a wave tank were conducted. The prototypes and metocean data were designed to clean marine debris at Ningbo Xiangshan Port. From the statistics of the testing results, it is found that the novel OSCS can effectively concentrate and clean marine debris of different types. We also measured the hydrodynamic responses of the OSCS under various sea conditions. The results show that the system's motion responses in the vertical plane are small, indicating the marine debris could not escape from the top and bottom of the system. Meanwhile, the motions in the horizontal plane are sufficiently high to cover a large ocean surface area. The present study indicates that the novel OSCS can be used to clean a large ocean surface at high efficiency.

Characterization of tropical cyclones to identify the response based design metocean conditions for an FPSO mooring system

Design metocean conditions (DMC) for a single-point moored weathervaning vessel exposed to the cyclonic environment are obtained by characterising the tropical cyclones on the basis of extreme responses induced in its mooring system. Determination of design metocean conditions, as combinations of wave, wind and current associated with the extreme response, is a multivariate problem that does not have a unique solution. Its probabilistic nature dictates that the design metocean condition should be a short-term sea state which is most probable to occur while at the same time inducing the extreme N-year response of the system. This study aims at the development of a methodology for obtaining the DMC for a mooring system of a weathervaning vessel by characterizing the tropical cyclonic environment based on the mooring responses. First, a method for identifying the DMCs within the metocean history is described, along with its limitations. Then, the storm characterisation approach is introduced, based upon the maximum responses of the system rather than the extreme sea states in a particular event. The metocean conditions at a specific location during the passage of a cyclone are described by a set of dimensionless shape functions, with their main parameters linked to the most probable mooring tension experienced during each cyclonic event. It is further determined that for a weathervaning floating system, cyclones should be separated into two distinct types according to the relative wave-wind direction highly affecting the mooring response. The cyclone shape functions are generalized to describe the characteristic cyclones at the N-year return period response, which are determined by considering a limited group of storms inducing the mooring responses close to the N-year response. By averaging the dimensionless shape functions for each of the two cyclone types, the characteristic cyclones can be synthesized to describe two types of events, where the N-year mooring response is most likely to occur. The robustness of the method is checked through a case study, in which the characteristic 5-year, 10-year and 100-year DMCs are produced for a mooring system of a floating production, storage and offloading vessel (FPSO) permanently moored offshore Australia. By using a 44-year hindcast time history of cyclonic metocean conditions, the DMCs were developed from the response-based synthesized N-year cyclones and compared with the specific short-term conditions retrieved from the same time history. An alternative averaging method for the cyclone shape functions is also considered and results are discussed.

Multi-parameter full probabilistic modeling of long-term joint wind-wave actions using multi-source data and applications to fatigue analysis of floating offshore wind turbines

In the fatigue assessment of offshore wind turbines, joint probabilistic models of long-term wind and wave parameters are usually required. In practice, annual met-ocean data typically exhibit non-stationarity due to the seasonal variations and extreme weather effects, and therefore cannot be considered as being from the same probability space. Thus, data separation and data segmentation should be performed. In this paper, the full probabilistic modeling of wind and wave parameters for a site in the South China Sea usually hit by typhoons is studied. For this purpose, the typhoon data is firstly separated from the normal wind data using multi-source data and physically based approach. Then, a modified Fisher's optimum partition method is proposed for the seasonal effects segmentation of the normal wind data. On this basis, the full probabilistic model of the environmental variables is developed using the C-vine copula method. The application of the full probabilistic model to the fatigue analysis of a floating offshore wind turbine (FOWT) is illustrated through an example. Numerical results indicate that the separation and segmentation of the long-term met-ocean data is quite significant to the full probabilistic modeling of the environmental variables, and the proposed methods can deal with this problem effectively.

The influence of the Arctic Oscillation on North Atlantic wind and wave climate by the end of the 21st century

This work analyses the change in the Arctic Oscillation (AO) index by the end of the 21st century and its relationship with present and future met ocean conditions on the North Atlantic Ocean. The seasonal distribution of the AO index changes is analysed and an annual regression line is obtained. Pearson correlation is computed between AO index and wind and wave parameters. It is observed that for future climate, negative AO events become weaker but more frequent and positive index events remain stable in intensity, but their frequency is reduced. At the end of the 21st century, the correlation between significant wave height, peak period and zonal wind remains the same as in the present climate: positive (negative) correlation north (south) of 50°N. Results show that future climate correlation on Wave Watch III forced by EC-Earth (EC-Earth/WW3) projections using the high-emissions scenario (RCP 8.5) and AO index will remain similar to the one observed in the present climate.

Application and improvement of iterative methodology for response based analysis of an FPSO mooring system in tropical cyclones

Response based analysis (RBA) is an advanced alternative to traditional design approach for mooring systems of floating structures, which reduces uncertainty associated with extreme responses. In the traditional approach, mooring analysis is performed for specified sets of extreme metocean conditions, representing typically a return period of 100 years. In RBA, the dynamic analysis is carried out over a long-term database of metocean conditions and statistics of the responses are used to identify responses associated with a return period of 100 years. Whilst the RBA has a more rigorous probabilistic basis, it is also computationally demanding. An “Iterative methodology” has been developed recently, comprising a combination of time domain (TD), frequency domain (FD) and probabilistic analyses of the structural system into a single iterative process, to minimise the number of time-consuming simulations through distinguishing between “mild” and “severe” response events. In the study, the “iterative methodology” is applied for RBA of the mooring system of a turret moored FPSO in the tropical cyclone environment of the North West Shelf of Australia. In the first stage, the time-efficient FD analysis is utilised to identify the contribution of all the storms to the 100-year response. In the second stage, an iteration process is set up, in which only the most contributing response events (tropical cyclones) and sea states are analysed by TD simulations. The iterations proceed by selecting the next most contributing storm until the convergence of the 100-year response is achieved. Application of the iterative methodology leads to significant reduction in computational time and makes RBA feasible as a part of the usual design process by using fully coupled TD analysis. Further optimisation can be achieved through a new method of applying thresholds to metocean parameters, which can reduce the analysis effort further. In the case study considered, it is found that 61% of metocean time history could be excluded from any dynamic analysis by applying a variable storm-dependent threshold. Among the remaining intervals, only 5.2% need to be re-analysed by TD simulations. Several methods for assessing errors in the convergence process are compared, and a sensitivity analysis on the number of storms that should be selected at each iteration step is performed.

Comparison of supervised machine learning methods to predict ship propulsion power at sea

As the shipping moves towards digitization, a large amount of ship energy performance-related information collected during a ship’s sailing provides opportunities to derive data-driven performance models using different machine learning algorithms. This paper compares several typical supervised machine learning algorithms, i.e., eXtreme Gradient Boosting (XGBoost), artificial neural network, support vector machine, and statistical regression methods, for the ship speed–power modeling. First, a general data pre-processing framework is presented. The different machine learning based models are trained by both ship operational parameters and encountered metocean conditions. Based on the full-scale measurement data collected at two types of worldwide sailing ships, the pros and cons of different machine learning algorithms for the ship’s speed–power performance modeling are compared. Finally, the best performed XGboost model is chosen to analyze the sensitivity due to the amount of available ship data, assumed time period for each stationary waypoint (data sample) used for the model training, and their impact on online performance prediction.

Weather Downtime Assesment for Complex Offshore Projects

Analysis of the expected downtime in complex offshore operations is performed with metocean analysis of the area of interest which is then compared to the operative limits defined by installation analyses. The metocean conditions are commonly represented by seastate hindcast time series. On the other end, the operative limits can be defined as maximum allowable sea state, as well as the maximum allowable vessel motion. This paper presents the methodology to evaluate the operative weather downtime based on classical operative sea state limit.

Neural Network Meta-Models for FPSO Motion Prediction From Environmental Data With Different Platform Loads

The current design process of mooring systems for Floating Production, Storage, and Offloading units (FPSOs) depends on the availability of the platform’s mathematical model and the accuracy of dynamic simulations. These simulations then provide the FPSO’s time series motion which is evaluated according to design constraints. This process can be time-consuming and present inaccurate results due to the mathematical model’s limitations and the overall complexity of the vessel’s dynamics. We propose a Neural Simulator, called NeuroSim, a set of data-based surrogate models with environmental data as input, each model specialized in predicting different motion statistics relevant to mooring system design: Maximum Roll, Platform Offset, and Fairlead Displacements. The surrogate models are trained by current, wind, and wave data given in 3 hours periods at a Brazilian Offshore Basin from 2003 to 2010, and the associated dynamic response of a spread-moored FPSO is obtained through time-domain simulations using the Dynasim software. Hyperparameter Optimization techniques are performed to obtain optimal Artificial Neural Network (ANN) models specialized in different platform drafts. Finally, the proposed models are shown to correctly capture platform dynamics, providing good results when compared to motion statistics obtained from Dynasim. We conclude that an ANN surrogate model can be trained directly on actual measured metocean conditions and corresponding FPSO motion statistics to provide increased accuracy and reduced computational time over traditional methods based on dynamic simulation. Moreover, the proposed architecture can be integrated into an automated learning framework: The data-based surrogate models can be continuously fine-tuned and updated with newly measured data, improving accuracy over time.

Bias Correction Techniques for Uncertainty Reduction of Long-term Metocean Data for Ocean Renewable Energy Systems

Bias Correction Techniques for Uncertainty Reduction of Long-term Metocean Data for Ocean Renewable Energy Systems

Modelling sea surface wind measurements on Australia’s North-West Shelf

The environment experienced by offshore industries is inherently volatile, providing a strong impetus for accurate modelling and prediction of meteorological and oceanographic conditions. Numerical forecasts, obtained from physics-based deterministic models, are informative yet imperfect in their prediction, motivating the quantification of their associated uncertainty. To achieve this, we present a Bayesian hierarchical model that updates numerical forecasts of metocean conditions and examines their associated error structure. This involves a linear regression to remove systematic forecast biases and the use of time series techniques, namely autoregressive and generalised autoregressive conditional heteroscedasticity models, to remove residual time-evolving error structure. We find that the hierarchical statistical model outperforms a numerical weather prediction model in terms of mean squared error and continuous ranked probability score, a metric which is able to account for uncertainty information contained in probabilistic forecasts. While the method is generalisable to a range of metocean conditions with an available set of past measurements and numerical forecasts, we present our findings using a case study of sea surface wind data. The data is collected from a location on Australia’s North-West Shelf, a region home to an extent of offshore operations and often characterised by extreme unexpected metocean conditions.

Storm Anatol over Europe in December 1999: impacts on societal and energy infrastructure

Abstract. Storm Anatol impacted the North Sea and northern Europe on 3–4 December 1999. It brought hurricane force winds to Denmark and northern Germany, and high winds also in Sweden and countries around the Baltic Sea. For many meteorological stations in Denmark, the wind speeds were the highest on record and the storm was ranked as a century event. The storm impacts included extensive forest damage, fatalities, hundreds of injuries, power outages, transportation interruptions, as well as storm surge flooding on the west coast of Denmark. Strongly committed to wind energy, Denmark lost 13 onshore wind turbines destroyed during the storm. An important industry insurer noted that this was a remarkably low number, considering the storm intensity and the large number of turbines (>3500) installed in the country. In 1999, offshore wind energy was just getting started in Europe, and the storm provided an important test of environmental extreme conditions impacting offshore infrastructure. This contribution takes a closer look at the regional met-ocean conditions during the storm. A brief overview is made of the wind field and available wave measurements from the North Sea. An examination is made of water level measurements from around the North Sea to characterize the storm surge and identify possible meteo-tsunamis and infragravity waves. Offshore accidents are briefly discussed to assess if there had been unusual wave strikes on shipping or platforms. At the time of the storm in 1999, there was a growing awareness in the scientific community of possible changes in ambient sea state conditions and the increasing threat of rogue waves. The offshore wind energy community had become aware from the impact of rogue waves from damage at the research platform FINO1 in the southern North Sea during severe storms in 2006, 2007, 2009, and 2013. Storm Anatol may have been another rogue wave storm at an earlier stage of offshore wind energy development.

Generation of Offshore Environments in the Numerical Wave Tank to Model Metocean Conditions Interaction with Offshore Structure Near the Free Surface

Offshore structures play a vital role in the economy of offshore oil-producing countries, where mostly fixed jacket type structures are used to produce oil and gas installed in shallow water. In an offshore environment where structures are installed, there exist met ocean forces such as wind, waves, and currents. These met ocean conditions when interacting with offshore structures near the free surface, generate loads. The estimation of such loads is very much important for the proper design of these structures. The primary aim of this study is to investigate the interaction of waves with a jacket platform by generating offshore environments in the numerical wave tank (NWT). To achieve this goal, ANSYS Fluent is used for the flow analysis by using continuity and Navier Stokes equation. Results are verified and validated with the analytical work. Wave crests under operating condition generate a force of 1.3 MN which is the lowest in magnitude as compared to wave crest which produces 4.5 MN force under extreme conditions. Unlike operating wave crest, the operating wave trough generates a higher force of 1 MN than extreme conditions which account for 1.5 MN forces. Forces produced by the extreme offshore environment are 30% higher than those generated under operating conditions. It is concluded from the results that a positive force is exerted onto the structure during the water entry phase while a negative force is observed when the water leaves the structure.