A5020725031- Wind and Air Flow Studies
- Wind Energy Research and Development
- Probabilistic and Robust Engineering Design
- Structural Health Monitoring Techniques
- Machine Fault Diagnosis Techniques
- Fluid Dynamics and Vibration Analysis
- Energy Load and Power Forecasting
- Fatigue and fracture mechanics
- Fluid Dynamics and Turbulent Flows
- Mechanical stress and fatigue analysis
- Vehicle Noise and Vibration Control
- Wind Turbine Control Systems
- Real-time simulation and control systems
- Mechanical Behavior of Composites
- Turbomachinery Performance and Optimization
- Sensor Technology and Measurement Systems
- Neural Networks and Applications
- Mechanical Failure Analysis and Simulation
- Time Series Analysis and Forecasting
- Structural Load-Bearing Analysis
- Reliability and Maintenance Optimization
- Engineering Applied Research
- Icing and De-icing Technologies
- Marine and Offshore Engineering Studies
- Meteorological Phenomena and Simulations
Technical University of Denmark
2015-2024
Lozenetz Hospital
2022
Danish Energy Association
2020-2021
University of Food Technologies
2018
Siemens (Denmark)
2012
Siemens (Germany)
2011-2012
California Maritime Academy
2009-2010
Institute of Marine and Coastal Research
2009
University of Stuttgart
1987
Abstract. We define and demonstrate a procedure for quick assessment of site-specific lifetime fatigue loads using simplified load mapping functions (surrogate models), trained by means database with high-fidelity simulations. The performance five surrogate models is assessed comparing predictions at 10 sites an aeroelastic model the DTU MW reference wind turbine. methods are polynomial chaos expansion, quadratic response surface, universal Kriging, importance sampling, nearest-neighbor...
Environmental contours are used to simplify the process of design response analysis. A wide variety contour methods exist; however, there have been a very limited number comparisons these date. This paper is output an open benchmarking exercise, in which contributors developed based on their preferred and submitted them for blind comparison study. The exercise had two components—one, focusing robustness across different offshore sites and, other, characterizing sampling uncertainty. Nine...
Abstract We analyse high‐frequency wind velocity measurements from two test stations over a period of several years and at heights ranging 60 to 200 m, with the objective validate shear predictions as used in load simulations for turbine design. A validated model is thereby proposed flat terrain that can significantly decrease uncertainty associated fatigue turbines large rotors. An essential contribution conditioning on 90% quantile turbulence, such appropriate magnitude design achieved....
Abstract. We present two methods to characterize turbulence in the turbine inflow using radial velocity measurements from nacelle-mounted lidars. The first uses a model of three-dimensional spectral tensor combined with spatial averaging lidars, and second ensemble-averaged Doppler spectrum. With former, filtered estimates can be predicted, whereas latter model-free method allows us estimate unfiltered measures. Two types forward-looking nacelle lidars are investigated: pulsed system that...
Abstract A procedure for mapping wake‐induced load predictions computed with the dynamic wake meandering model to a computationally efficient surrogate approximation is defined and demonstrated. Using function, variation can be efficiently estimated wind farm arbitrary layout. The resulting assessment provides continuous, differentiable output known analytical derivatives used applications such as turbine layout optimization, estimation of lifetime, uncertainty analysis.
Abstract This article presents a joint statistical model, which is needed in probabilistic design and structural risk assessment, that has been fitted to data of wind wave conditions for an offshore location off South Brittany. The are from numerical model contain hourly values several variables over period 32 years. distribution presented this considers the direction, mean speed, significant height, peak period. A conditional turbulence given speed introduced yield additional variable...
Abstract Minimizing the cost of energy a wind farm is difficult task, which involves reducing wake effects while satisfying several constraints. Due to its multidisciplinary nature, this problem usually solved through numerical optimisers. TOPFARM one these tools, and in paper, we have added it constraint on fatigue loads. The efficiency implementation guaranteed by an extensive use gradients load surrogate models. paper concluded showing some case studies.
Abstract Modern wind turbines have multiple sensors installed and provide constant data stream outputs; however, there are some important quantities where installing physical is either impractical or the sensor technology not sufficiently advanced. An example of such a problem is, for example, sensing shape location wake‐induced deficits caused by upwind turbines—a feature which would relevant application in farm control; it hard to detect physically due need scanning airflow front turbine...
Beyond the existing, limited IEC prescription to describe fatigue loads on wind turbines, we look towards probabilistic characterization of via analogous atmospheric flow, particularly for today's "taller" turbines with rotors well above surface layer. Based both data from multiple sites as theoretical bases boundary-layer meteorology and turbulence, offer descriptions shear turbulence intensity, elucidating connection each other stability terrain. These are used input calculation, a...
We demonstrate a method for incorporating wind velocity measurements from multiple-point scanning lidars into threedimensional turbulence time series serving as input to turbine load simulations.Simulated lidar patterns are implemented by imposing constraints on randomly generated Gaussian fields in compliance with the Mann model neutral stability.The expected efficiency of various is estimated means explained variance associated constrained field.A numerical study made using HAWC2...
Previous studies have suggested the use of reduced-order models calibrated by means high-fidelity load simulations as for computationally inexpensive wind turbine assessments; so far best performing surrogate modelling approach in terms balance between accuracy and computational cost has been polynomial chaos expansion (PCE). Regarding growing interest advanced machine learning applications, potential using Artificial Neural-Network (ANN) based improved simplified assessment is investigated...
Abstract We define and demonstrate a procedure for carrying out wind turbine load validation based on measurements from nacelle‐mounted scanning lidars. Two coherent Doppler lidar systems, pulsed continuous‐wave lidar, are mounted 2.3‐MW equipped with measurement sensors. Wind meteorological mast at 2.5 rotor diameters distance used as reference. The study shows how processed applied inputs to aeroelastic simulations, the results then compared simulations where have been determined using...
Abstract. This study proposes two methodologies for improving the accuracy of wind turbine load assessment under wake conditions by combining nacelle-mounted lidar measurements with field reconstruction techniques. The first approach consists incorporating flow field, obtained from nacelle lidars, into random, homogeneous Gaussian turbulence fields generated using Mann spectral tensor model. second imposes deficit time series, which are derived fitting a bivariate shape function to...
Leading edge erosion has become one of the most prevailing failure modes wind turbines. Its effects can evolve from an aerodynamic modification properties blade to a potential structural leading edge. The first produces reduction energy production and second produce catastrophic blade. Considering uncertainties constraints involved in design optimal operation maintenance (O&M) strategies for offshore assets influence site-specific parameters on dynamics this particular mode, task becomes...
Bread is one of the most consumed foods globally and incorporating walnut flour will enhance its nutritional value by including biologically active components. The porosity a key characteristic sensory properties overall acceptability bread crumb. objective this study was to investigate effect on crumb employing image analysis. Six breads were prepared with mixtures wheat:walnut in proportions from 100:0 85:15. number pores, pore size, total area, mean size circularity crumbs investigated....
Abstract. For measurements taken over a decade at the coastal Danish site Høvsøre, we find variance associated with wind speed events from offshore direction to exceed prescribed extreme turbulence model (ETM) of International Electrotechnical Commission (IEC) 61400-1 Edition 3 standard for turbine safety. The velocity fluctuations manifested during these is not due turbulence; rather, it primarily caused by ramp-like increases in larger-scale meteorological processes. are both linearly...
Abstract We devise a methodology to predict failures in wind turbine drive‐train components and quantify its utility. The consists of two main steps. first step is the set up predictive model for shutdown events, which able raise an alarm advance fault‐induced shutdown. trained on data events retrieved from log offshore farm. Here, it assumed that timely prediction low‐severity typically caused by abnormal component operation, allows intervention can prevent premature failures. models are...
This paper introduces a novel, transfer-learning-based approach to include physics into data-driven normal behavior monitoring models which are used for detecting turbine anomalies. For this purpose, model is pretrained on large simulation database and recalibrated the available SCADA data via transfer learning. two methods, feed-forward artificial neural network (ANN) an autoencoder, it investigated under conditions can be helpful simulations SCADA-based systems. The results show that when...
This paper describes the development of a wind turbine (WT) component lifetime fatigue load variation map within an offshore farm. A case study on farm Horns Rev I is conducted with this purpose, by quantifying wake effects using Dynamic Wake Meandering (DWM) method, which has previously been validated based CFD, Lidar and full scale measurements. Fully coupled aeroelastic simulations turbulent conditions are for all directions mean speeds between cut-in cut-out site specific turbulence...
A common challenge in the decision making process regarding operation and life extension of existing wind farms is lack accurate information about actual dynamic states turbines terms its from inception. SCADA records normally contain limited number channels, are not necessarily kept for entire operating period farm; design site data may be outdated or inaccessible. Nevertheless, as long a minimum amount available, statistical analysis augmentation with artificial intelligence based...