Displacement, Strain and Failure Estimation for Multi-Material Structure Using the Displacement-Strain Transformation Matrix

In this study, we propose a method to estimate structural deformation and failure by using displacement-strain transformation matrices, i.e., strain-to-displacement transformation (SDT) and displacement-to-strain transformation (DST). The proposed SDT method can be used to estimate the complete structural deformation where it is not possible to apply deformation measurement sensors, and the DST method can be used for to estimate structural failures where strain and stress sensors cannot be applied. We applied the SDT matrix to a 1D beam, a 2D plate, rotating structures and real wind turbine blades, and successfully estimated the deformation in the structures. However, certain difficulties were encountered while estimating the displacement of brittle material such as an alumina beam. The study aims at estimating the displacement and stress to predict the failure of the structure. We also explored applying the method to multi-material structures such as a two-beam bonded structure. In the study, we used alumina–aluminum bonded structures because alumina is bonded to the substrate to protect the structure from heat in many cases. Finally, we present the results of the displacement and failure estimation for the alumina–aluminum structure.

Download Full-text

Numerical Analysis of Aeroelastic Responses of Wind Turbine Under Uniform Inflow

Volume 9: Ocean Renewable Energy ◽

10.1115/omae2020-18084 ◽

2020 ◽

Author(s):

Yang Huang ◽

Decheng Wan

Keyword(s):

Wind Turbine ◽

Structural Model ◽

Turbine Blades ◽

Structural Deformation ◽

Wind Turbine Blades ◽

Wake Field ◽

Wind Speeds ◽

Inflow Condition ◽

Blade Tip ◽

Aeroelastic Simulations

Abstract With wind turbine blades becoming longer and slender, the influence of structural deformation on the aerodynamic performance of wind turbine cannot be ignored. In the present work, the actuator line technique that simplifies the wind turbine blades into virtual actual lines is utilized to simulate the aerodynamic responses of wind turbine and capture downstream wake characteristics. Moreover, the structural model based on a two-node, four degree-of-freedom (DOF) beam element is adopted for the deformation calculation of the wind turbine blades. By combing the actuator line technique and linear finite element theory, the aeroelastic simulations for the wind turbine blades can be achieved. The aeroelastic responses of NREL-5MW wind turbine under uniform wind inflow condition with different wind speeds are investigated. The aerodynamic loads, turbine wake field, blade tip deformations and blade root bending moments are analyzed to explore the influence of blade structural responses on the performance of the wind turbine. It is found that the power output of the wind turbine decreases when the blade deformation is taken into account. Significant asymmetrical phenomenon of the wake velocity is captured due to the deformation of the wind turbine blades.

Download Full-text

Wind Turbine Maintenance Cost Reduction by Deep Learning Aided Drone Inspection Analysis

10.20944/preprints201901.0281.v1 ◽

2019 ◽

Author(s):

ASM Shihavuddin ◽

Xiao Chen ◽

Vladimir Fedorov ◽

Nicolai Andre Brogaard Riis ◽

Anders Nymark Christensen ◽

...

Keyword(s):

Deep Learning ◽

Wind Turbine ◽

Data Augmentation ◽

Turbine Blades ◽

Automated Analysis ◽

Machine Learning Algorithms ◽

Maintenance Cost ◽

Wind Turbine Blades ◽

Surface Damages ◽

Structural Failures

Timely detection of surface damages on wind turbine blades is imperative for minimising downtime and avoiding possible catastrophic structural failures. With recent advances in drone technology, a large number of high-resolution images of wind turbines are routinely acquired and subsequently analysed by experts to identify imminent damages. Automated analysis of these inspection images with the help of machine learning algorithms can reduce the inspection cost, thereby reducing the overall maintenance cost arising from the manual labour involved. In this work, we develop a deep learning based automated damage suggestion system for subsequent analysis of drone inspection images. Experimental results demonstrate that the proposed approach could achieve almost human level precision in terms of suggested damage location and types on wind turbine blades. We further demonstrate that for relatively small training sets advanced data augmentation during deep learning training can better generalise the trained model providing a significant gain in precision.

Download Full-text

Wind Turbine Surface Damage Detection by Deep Learning Aided Drone Inspection Analysis

Energies ◽

10.3390/en12040676 ◽

2019 ◽

Vol 12 (4) ◽

pp. 676 ◽

Cited By ~ 22

Author(s):

ASM Shihavuddin ◽

Xiao Chen ◽

Vladimir Fedorov ◽

Anders Nymark Christensen ◽

Nicolai Andre Brogaard Riis ◽

...

Keyword(s):

Deep Learning ◽

Wind Turbine ◽

Data Augmentation ◽

Turbine Blades ◽

Surface Damage ◽

Automated Analysis ◽

Machine Learning Algorithms ◽

Wind Turbine Blades ◽

Surface Damages ◽

Structural Failures

Timely detection of surface damages on wind turbine blades is imperative for minimizing downtime and avoiding possible catastrophic structural failures. With recent advances in drone technology, a large number of high-resolution images of wind turbines are routinely acquired and subsequently analyzed by experts to identify imminent damages. Automated analysis of these inspection images with the help of machine learning algorithms can reduce the inspection cost. In this work, we develop a deep learning-based automated damage suggestion system for subsequent analysis of drone inspection images. Experimental results demonstrate that the proposed approach can achieve almost human-level precision in terms of suggested damage location and types on wind turbine blades. We further demonstrate that for relatively small training sets, advanced data augmentation during deep learning training can better generalize the trained model, providing a significant gain in precision.

Download Full-text