Fig 1 - uploaded by Jinkyoo Park
Content may be subject to copyright.
Source publication
Site-specific and time-specific wind field characteristics have a significant impact on the structural response and the lifespan of wind turbines. This paper presents a machine-learning approach towards analyzing and predicting the response of a wind turbine structure to diurnal and nocturnal wind fields. Machine-learning algorithms are applied (1)...
Contexts in source publication
Context 1
... study is based on the monitoring data collected on a 500-kW wind turbine located in Dortmund, Germany (Fig. 1). The wind turbine has a hub height of 65 m and upwind rotor diameter of 40.3 m. A life-cycle management (LCM) framework has been de- veloped for the monitoring and operational management of the wind turbine. The LCM framework provides valuable information that can be used not only to monitor the integrity of the wind turbine structure, ...
Context 2
... the tower to gauge tem- perature gradients. For measuring wind speed, wind direction, and air temperature, two anemometers are deployed. The first one, a cup anemometer that is directly connected to the integrated supervisory control and data acquisition (SCADA) system of the wind turbine, is installed on top of the nacelle at a height of 67 m (Fig. 1). The second one, a three-dimensional ultrasonic anemometer, is mounted on a telescopic mast adjacent to the wind turbine at a height of 13 m; it continuously monitors the horizontal and vertical wind speed (0-60 m=s), the wind direction (0-360°) and the air temper- ature (−4-60°C). In addition to the structural and environmental sensor ...
Context 3
... study the wind turbine responses corresponding to the diur- nal and the nocturnal wind fields, the class probabilities Pðy C jt P Þ Fig. 10. Time-specific PDFs: (a) diurnal; (b) nocturnal cases; the response classification function g C ðxÞ for evaluating the class probabilities and the expected response classes Interquartile range (mg) 0.0-3.8 3.8-7.6 7.6-11.4 11.4-15.2 ...
Context 4
... ( t P = 1) Nighttime( t P = 2) Fig. 11. Comparison of the estimated class probabilities for the day and the night times and the expected response classes E½y C jt P for the wind turbine tower acceleration during the daytime t P ¼ 1 and during the night- time t P ¼ 2 are constructed. Fig. 10 shows pðxjt P ¼ 1Þ and pðxjt P ¼ 2Þ overlapped with the load classification ...
Context 5
... ( t P = 1) Nighttime( t P = 2) Fig. 11. Comparison of the estimated class probabilities for the day and the night times and the expected response classes E½y C jt P for the wind turbine tower acceleration during the daytime t P ¼ 1 and during the night- time t P ¼ 2 are constructed. Fig. 10 shows pðxjt P ¼ 1Þ and pðxjt P ¼ 2Þ overlapped with the load classification function g C ðxÞ, which divides the input feature domain into five different classes. For display purposes, the PDFs are represented by the con- tour surfaces with pðxjt P Þ ¼ 0.02. As shown in Fig. 10, the PDF for the daytime wind disperses more widely in the ...
Context 6
... daytime t P ¼ 1 and during the night- time t P ¼ 2 are constructed. Fig. 10 shows pðxjt P ¼ 1Þ and pðxjt P ¼ 2Þ overlapped with the load classification function g C ðxÞ, which divides the input feature domain into five different classes. For display purposes, the PDFs are represented by the con- tour surfaces with pðxjt P Þ ¼ 0.02. As shown in Fig. 10, the PDF for the daytime wind disperses more widely in the region associated with the high load classes (y C ¼ 3, 4 and 5) than the PDF for the nighttime wind. As a result, as shown in Fig. 11, the diurnal wind fields have higher probabilities for classes 3, 4, and 5 but lower class probabilities for classes 1 and 2 than the nocturnal ...
Context 7
... feature domain into five different classes. For display purposes, the PDFs are represented by the con- tour surfaces with pðxjt P Þ ¼ 0.02. As shown in Fig. 10, the PDF for the daytime wind disperses more widely in the region associated with the high load classes (y C ¼ 3, 4 and 5) than the PDF for the nighttime wind. As a result, as shown in Fig. 11, the diurnal wind fields have higher probabilities for classes 3, 4, and 5 but lower class probabilities for classes 1 and 2 than the nocturnal wind ...
Similar publications
The optical turbulence above Dome C in winter is mainly concentrated in the first tens of metres above the ground. Properties
of this so-called surface layer (SL) were investigated during the period 2007–2012 by a set of sonic anemometers placed on
a 45 m high tower. We present the results of this long-term monitoring of the refractive index struct...
Citations
... In addition, robust multivariate statistical methods are introduced to account for the influence of operational and environmental variation on damage-sensitive features; the algorithms described are the Minimum Covariance Determinant Estimator and the Minimum Volume Enclosing Ellipsoid. Park et al. [16], also focusing on wind energy research, couple Gaussian Discriminative Analysis and Gaussian Mixture Models to analyze and to predict wind turbine loads in various atmospheric conditions. Nick et al. [17], reporting significant trade-offs between accuracy and runtime of the machine learning techniques proposed, have used unsupervised learning for identifying the existence and location of damage (k-means and self-organizing maps) and supervised learning for identifying the type and severity of damage (support vector machines, naive Bayes classifiers, and feed-forward neural networks). ...
Data-driven approaches are particularly useful for computer-supported assessment of civil engineering structures (i) if large quantities of sensor data are available, (ii) if the physical characteristics of the structure are complex to model (or even unknown), or (iii) if the computational efforts are to be reduced. This paper, upon a classificational review of machine learning techniques in structural health monitoring, reports on an embedded machine learning approach for decentralized, autonomous sensor fault detection in wireless sensor networks, facilitating reliable and accurate structural health monitoring. Based on decentralized artificial neural networks, the embedded machine learning approach is applied to perform autonomous detection of sensor faults injected in the acceleration response data collected by a prototype structural health monitoring system. As demonstrated through laboratory tests, the results highlight the ability of the embedded machine learning approach to autonomously detect sensor faults in a decentralized manner, thus enhancing the reliability and accuracy of structural health monitoring systems.
