Wind Turbine Blade Surface Research Articles

Lightweight algorithm based on you only look once version 5 for multiple class defect detection on wind turbine blade surfaces

A lightweight wind turbine blade surface multiple class defect detection algorithm based on You Only Look Once version 5 (YOLOv5) is proposed to address the issues of low accuracy, poor efficiency, and high power consumption associated with manual and traditional automation algorithms used for detecting surface defects on wind turbine blades. Firstly, by replacing the Concentrated-Comprehensive Convolution Block module in YOLOv5 with the GhostBottleneckv2 module, the overall parameter count of the model is significantly reduced. Additionally, advocating the inclusion of the Convolutional Block Attention Module (CBAM) into the YOLOv5 backbone network to address the issue of small targets and significant scale variations in surface defects found on wind turbine blades. The YOLOv5 feature fusion module replaces the Feature Pyramid Network (FPN) structure with the Bidirectional Feature Pyramid Network (BiFPN) structure to enhance the fusion capability of multi-scale weighted features. Finally, the optimized model files underwent structured pruning through the application of pruning strategies, retraining fine-tuning, and knowledge distillation techniques. The experimental results on the self-made dataset of surface defects on wind turbine blades indicate that the optimized model has a parameter count that is only 48.79% of the small version of the YOLOv5 model. However, its mean Average Precision (mAP) and frame rate have increased by 5.05% and 8.1 frames/s, respectively, reaching 94.56% and 43.7 frames/s. The lightweight algorithm proposed in this study meets the accuracy and real-time requirements for surface defect detection on wind turbine blades.

Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features

Abstract An enhanced wind turbine blade surface defect detection algorithm, CGIW-YOLOv8, has been introduced to tackle the problems of uneven distribution of defect samples, confusion between defects and background, and variations in target scales that arise during drone maintenance of wind turbine blades. This algorithm is given based on the YOLOv8 model. Initially, a data augmentation method based on geometric changes and Poisson mixing was used to enrich the dataset and address the problem of uneven sample distribution. Subsequently, the incorporation of the Coordinate Attention (CA) mechanism into the Backbone network improved the feature extraction capability in complex backgrounds. In the Neck, the Reparameterized Generalized Feature Pyramid Network (Rep-GFPN) was introduced as a path fusion strategy and multiple cross-scale connections are fused, which effectively enhances the multi-scale expression ability of the network. Finally, the original CIOU loss function was replaced with Inner-WIoU, which was created by applying the Inner-IoU loss function to the Wise-IoU loss function. It improved detection accuracy while simultaneously speeding up the model's rate of convergence. Experimental results show that the mAP of the method for defect detection reaches 92%, which is 5.5% higher than the baseline network. The detection speed is 120.5 FPS, which meets the needs of real-time detection.

An adaptive detection approach for multi-scale defects on wind turbine blade surface

Multi-scale defects will inevitably appear in wind turbine blade defect detection within practical detection applications. An adaptive multi-scale detection approach is proposed to accurately classify and locate defects on the wind turbine blade surface. The proposed approach includes two main detection procedures: rough detection and precise detection. In the rough detection, the multi-level features extraction module with an adaptive bounding box proposal module is used to depict multi-scale defect regions and train a binary classifier to distinguish defects from non-defects. At the precise detection stage, three defect categories and four coordinates representing defect locations are obtained based on a multi-class defect classifier and regression of rough location boxes. The proposed method is evaluated on a real wind turbine blade surface defect dataset collected in a commercial wind farm and annotated manually. Results show that (1) the proposed model can detect the class of the blade multi-scale defects and outperforms other schemes with 96.89% mAP in the same model training epochs, (2) the positioning performance analysis of the model for multi-scale defects is conducted to validate the accuracy of the proposed model for multi-scale defect location.

Preparation and performance study of SiO2 aerogel thermal insulation coating with nanoporous structure for wind turbine blade surface

To prepare a base layer of thermal insulation for the surface of wind turbine blades, we utilized SiO2 aerogel material with a nanoporous structure as the thermal insulation layer of an inorganic composite photothermal de-icing coating. We further investigated the influence of raw material ratio and process on the thermal insulation layer’s performance. By adjusting the process parameters, the microstructure and characteristics of the SiO2 aerogel thermal insulation coating were controlled. Microstructure scanning and EDS analysis were employed to assess the results. The results of the experiment suggest that, as opposed to brush coating, atomized spraying is a better technique for creating SiO2 aerogel thermal insulation coating. The coating possesses excellent mechanical stability and applicability, maintaining a coating wear rate within 5% after 20 wear cycles under a load of 1734 N/m2, and the coating achieves the highest class 0 surface adhesion to the substrate surface. The coating exhibits exceptional heat insulation performance, reaching a temperature of 32.7 °C after five spraying cycles, which is 20.0 ∼ 29.4 °C higher than the surface without producing a complete heat insulation layer. The SiO2 aerogel aqueous slurry has a viscosity of 31.3 ∼ 36.9 mPa∙s, which is capable of forming a uniform membrane surface when sprayed at a 60° angle at 0.95 MPa pressure. Consequently, this paper’s design of photothermal thermal insulation offers strong anti-abrasion and excellent thermal insulation, presenting a new avenue for scientific investigation into coating photothermal de-icing.

Defect detection of the surface of wind turbine blades combining attention mechanism

The proposed work introduces a novel, lightweight feature fusion network model based on the attention mechanism to address the issues of high time consumption and poor effectiveness in wind turbine blade surface defect detection. A bidirectional feature fusion network enhances the YOLOX network. The model employs an attention mechanism module to learn channel and spatial feature information. A classification loss function with an attenuation factor is designed to tackle the unbalanced distribution of learning weights for multiple samples in the blade surface defect detection task. Additionally, to mitigate recognition loss caused by target occlusion or overlap, the Soft-NMS method is utilized to eliminate redundant detection boxes. Sample imbalance is addressed by creating new samples using a geometric transformation-based multi-sample fusion data enhancement method and unsupervised learning MFF-GAN image editing techniques. A feature fusion network comparison test, classification loss function comparison test, and attention mechanism module position ablation test were completed using the blade surface defect dataset. The results indicate that the blade surface defect detection model incorporating the attention mechanism can effectively identify five types of defects with an average detection accuracy of mAP-0.5, achieving approximately 95.03 % and a detection frame rate of about 54.56 frame·s−1. Compared with the YOLOX-s network, YOLOv7-tiny network, and YOLOv8-s network, this model can improve the recognition rate and shorten the detection time, achieving high precision and rapid identification and localization of blade surface defects.

Multiple damage segmentation and extraction of wind turbine blades surface under complex background based on SKRT approach

Machine vision detection technology has been widely used in detecting wind turbine blade surface damage, but the complex background often has a significant impact on blade damage detection. Aiming at the complex background blade images caused by various natural background features, complex blade surface features, low contrast features, non-uniform illumination, weather conditions, and other factors, the work proposes an SKRT segmentation method based on K-means clustering and adaptive threshold fusion to segment and extract multiple damage features on the surface of wind turbine blades with complex backgrounds. Firstly, an single-scale retinex(SSR) algorithm is adopted to enhance the contrast between surface damage and image background in the research. Then the image is roughly segmented by K-means clustering to remove most of the background features, and the background mask is replaced with the peak of the remaining feature image pixels to solve the problem of wrong segmentation. Finally, making use of the adaptive threshold local segmentation method to accurately segment the blade surface damage. The experimental results show that the proposed SKRT segmentation method can significantly improve the segmentation accuracy of wind turbine blade surface damage with complex background, and the accuracy, intersection over union, and F-measure value are increased by 37.20%, 35.71%, and 28.69%, respectively. The method performs better robustness to multiple damage feature segmentation.

Complex damage mechanisms and roughness evolution of wind turbine blade surface: Multiphysics and stochastic effect modelling

Leading edge erosion of wind turbine blades is the most often observed damage mechanism of wind turbines. The surface erosion of blades is influenced by many multiphysics and stochastic factors including humidity and related degradation processes, rough uneven surface and roughness development, random defects in the materials. In this paper, the effects of these factors and possibilities of their computational modelling and prediction are discussed. Competing damage mechanisms in erosion, including debonding and impact damage, as well moisture ingress and weathering are investigated. A predictive model for roughness evolution of leading edge due to the surface damage is presented.

Wind tunnel test of ice accretion on blade airfoil for wind turbine under offshore atmospheric condition

For wind turbines under offshore conditions, ice accretion occurs on the blade surface in winter because of the cold and humid environment, which leads to the performance degradation of the wind turbine. The characteristics of icing on the wind turbine blade surface under the in-cloud condition with salinity are explored. A blade segment with an airfoil of NACA0018 is selected. The tests of icing on the blade surface in different conditions of salinities, temperatures, and wind speeds are conducted in a return-flow icing wind tunnel. Two parameters are defined to evaluate icing characteristics. The distribution and amount of ice are analyzed quantitatively. Results show that salinity can restrain the amount of ice accretion. Oppositely, low temperatures and high wind speeds can increase the amount of ice. A self-developed device was manufactured to measure the adhesion strength of ice. The effects of salinity, temperature, and wind speed on adhesion strength are studied. Research indicates that adhesion strength decreases sharply first and then slowly with an increase in salinity. Even though low temperature and high wind speed both increase the adhesion strength, the growth rate decreases. The research provides a reference for anti- and de-icing technologies of offshore wind turbines.

Analyzing the aerodynamic characteristics of eroded wind turbine blades

The study investigates the influence of erosion over the leading edge of S809 series airfoil and its impact on the aerodynamic performance of wind turbine blades. The research work determines how erosion affects the performance of wind turbine blades at an angle of attack varying from 0° to 45° for the fixed Reynolds number of 2.1 x 105. To analyse the flow characteristics, the average pressure acting around the surface of the airfoil is scanned with the aid of 20 pressure taps. The time series pressure of 10,000 samples at each point has been collected. The roughness is introduced at the leading edge of the airfoil and measures 0.1% of the total chord at the top and bottom surface, respectively. Six different intensities of roughness are introduced and tested. The base airfoil, as well as roughed surface airfoil, is compared and the results are discussed. The aerodynamic parameters like lift, drag, and pressure coefficients are included. The estimated data suggests the erosion that occurs on the wind turbine blade surface alters the flow pattern and the aerodynamic properties of the wind turbine blades. Collectively, the experiment on analysing the eroded surface possibly taking place over the wind turbine blade surface affects the lift parameter with a considerable increment in drag. However, the roughness allows the airflow to remain fixed with the airfoil surface, favouring the concept of extending the stall.

Dynamic response of wind turbine blade surface material under water droplet high velocity impact

When a wind turbine blade exposure to some degree of rainfall, water droplets in the atmosphere impact the blade at high velocity will causes erosion damages of the leading edge. The damages have significant influence to the aerodynamic performance of the blade, and also the cost of power generation. In this paper, a high velocity impact dynamics model of water droplet impact on wind turbine blade is develops based on SPH-FEM coupling method, where SPH model is used to model the large deformation region of the fluid domain, and FEM method is used to model the small deformation region of the structure domain. Together with properties of gelcoat material on the blade surface, effects of varying impact velocities and angles on the impact response of the gelcoat material are analysed through numerical simulation. The results show that during a high velocity impact event of water droplet, the contact force history appears two peaks. Also contact forces and strains on the blade material decreases with decreasing droplet impact velocity and impact angle, the total contact duration of the droplet during the impact decreases with increasing droplet impact velocity and angle. The present study is expected to provide theoretical guidance for enhancing the erosive capacity of wind turbine blade gelcoat material.

Research on Wind Turbine Blade Surface Damage Identification Based on Improved Convolution Neural Network

Wind turbine blades are easily affected by the working environment and often show damage features such as cracks and surface shedding. An improved convolution neural network, ED Net, is proposed to identify their damage features. An EAC block based on the improved asymmetric convolution is introduced which strengthens the feature extraction during convolution. A DPCI_SC block, which is improved based on the attention module, is embedded to enhance the ability to obtain spatial location information of the damage. GELU is used as the activation function. The loss function is smoothed and labeled during training. Finally, three sets of experiments were conducted. Experiment 1 confirmed the efficacy of the ED Net for identifying damaged wind turbine blades. Experiment 2 confirmed the efficacy of the relevant improvements proposed in this work. Experiment 3 compares the recognition of wind turbine blade damage by commonly used lightweight networks and shows that the ED Net model proposed has a better performance with an accuracy range of 99.12% to 99.23% and a recall of 99.23%

Fault identification of fan blade based on improved ResNet-18

With the continuous development of the wind industry, wind turbine blade failure causes wind turbine shutdown, which damages the revenue of wind power. In order to more effectively detect the damage on the surface of wind turbine blade. By comparing ResNet-18 /50, VGG-19 and GoogLeNet models, ResNet-18 has the advantages of short training time and high accuracy. Therefore, ResNet-18 convolutional neural network is selected as the basic network, and a deep learning training and detection model is built based on MATLAB. In order to further optimize the model, multiple convolution kernel is selected for the main branch of residual structure of ResNet-18 to extract features, and add 3×3 pool layer, it is used to improve the learning ability of the model. In addition, Aadm, Sgdm and Rmsprop optimization algorithms are added to further improve the robustness of the model. Experiments show that the classification and recognition accuracy of the improved ResNet-18 model for fan blades is 94.17%, which is 2.5% higher than that of the original model. The improved ResNet-18 model has good recognition effect,it improves a faster and more accurate solution for fan blade surface damage identification in complex weather environment.

SOD‐YOLO: A Small Target Defect Detection Algorithm for Wind Turbine Blades Based on Improved YOLOv5

AbstractEarly and effective detection of wind turbine blade (WTB) surface defects can avoid complex and expensive repair problems and serious safety hazards. The traditional target detection methods have the problems of insufficient detection capability, long model inference time and low recognition accuracy for small targets and long strip defects in WTB datasets. This paper proposes a high‐precision model SOD‐YOLO for WTB surface defect detection based on UAVs image analysis of YOLOv5. First, the WTB images are preprocessed by foreground segmentation and Hough transform to build the WTB defect dataset. Then, a micro‐scale detection layer is added to the original YOLOv5, and the K‐means algorithm is used to re‐cluster anchors and add the CBAM attention mechanism to each feature fusion layer to reduce the loss of feature information for small target defects and other defects. In addition, to improve the detection efficiency, the channel pruning algorithm is used to reduce the model size. The experimental results show that the average accuracy (mAP) of the SOD‐YOLO algorithm on the WTB dataset reaches 95.1%, which is 7.82% better than YOLOv5, and the FPS is 28.3% better. Therefore, SOD‐YOLO is able to detect small target defects and other defects quickly and effectively.

Research on surface damage detection of wind turbinebladebased on machine vision

Abstract The research of wind turbine blade surface damage identification is of great significance to reduce the risk of wind turbine operation. Aiming at the limitation of manual inspection based damage detection, this paper takes the wind turbine as the object of identification and detection, takes the advanced machine detection method as the research content. In order to realize the task of identification and detection of turbine blade damage, an image classification system of turbine blade based on convolution neural network is proposed. In this paper, the convolution neural network (CNN) based damage identification system is compared with the support vector machine (SVM) based damage identification system. The experimental results show that the CNN based damage identification system has stronger generalization ability and higher classification accuracy in the case of randomly dividing experimental sample data.

Simulation and Experimental Study on the Ultrasonic Micro-Vibration De-Icing Method for Wind Turbine Blades

In cold and humid regions, ice accretion sometimes develops on the blades of wind turbines. Blade icing reduces the power generation of the wind turbine and affects the safe operation of the wind farm. For this paper, ultrasonic micro-vibration was researched as an effective de-icing method to remove ice from the wind turbine blade surface and improve the efficiency of wind turbine power generation. A blade segment with NACA0018 airfoil and the hollow structure at the leading edge was designed. The modal analysis of the blade was simulated by ANSYS, and the de-icing vibration mode was selected. Based on the simulation results, the blade segment sample with PZT patches was machined, and its natural frequencies were measured with an impedance analyzer. A return-flow icing wind tunnel system, and a device used to measure the adhesive strength of ice covering the airfoil blade, were designed and manufactured. The experiments on the adhesive strength of the ice were carried out under the excitation of the ultrasonic vibration. The experimental results show that the adhesive strength of the ice, which was generated under the dynamic flow field condition, was lower than the ice generated by water under the static flow field condition. Under the excitation of the ultrasonic vibration, the adhesive strength of the ice decreased. When the excitation frequency was 21.228 kHz, the adhesive strength was the lowest, which was 0.084 MPa. These research findings lay the theoretical and experimental foundations for researching in-depth the application of the ultrasonic de-icing technology to wind turbines.

Effect of perfluoropolyether and the micro nano structure of ZnO on anti icing performance of fluorinated organic superhydrophobicity coatings on wind turbine blade surface

The ice coating on the blade surface of wind turbine in winter seriously affects the operation safety and power generation efficiency of wind turbine, and anti icing and deicing is an urgent technical problem. In this paper, a super-hydrophobic anti-icing coating was prepared by fluorination modification, and the effect of the addition of perfluoropolyether (PFPE) and ZnO on the performance of the coating was studied. The research results show that the addition of PFPE reduces the surface energy of the coating from 18.12 mJ m−2 at 0% to 13.22 mJ m−2 at 15%. The surface free energy has a linear relationship with the binding force of icing. With the increase of the amount of PFPE, the contact angle of the coating increases, the sliding angle decreases, and the delayed icing performance improves. When the amount of PFPE reaches 15%, the contact angle reaches the highest value of 158°, and the sliding angle is 2°; adding PFPE coating, the freezing delay time can reach 107.1 s. The addition of PFPE can effectively reduce the icing strength, and the icing bonding force of the coating is reduced to 0.59 N at 15%. The addition of a amount of ZnO can greatly change the microstructure of the coating surface, thereby changing its superhydrophobicity. The fluorinated superhydrophobicity coating modified by the addition and removal of ZnO has obvious anti-icing performance. The 6 wt% sample has the best delayed icing energy.

Nanoengineered Graphene-Reinforced Coating for Leading Edge Protection of Wind Turbine Blades

Possibilities of the development of new anti-erosion coatings for wind turbine blade surface protection on the basis of nanoengineered polymers are explored. Coatings with graphene and hybrid nanoreinforcements are tested for their anti-erosion performance, using the single point impact fatigue testing (SPIFT) methodology. It is demonstrated that graphene and hybrid (graphene/silica) reinforced polymer coatings can provide better erosion protection with lifetimes up to 13 times longer than non-reinforced polyurethanes. Thermal effects and energy dissipation during the repeated soft impacts on the blade surface are discussed.

Unsteady characteristics of pressure on wind turbine blade surface in the field

Field experiments are carried out to investigate the fluctuation characteristics of pressure on wind turbine blade surface in real operating environments. The tested wind turbine positioned pressure tapes along the blade span to obtain the dynamic pressure of seven airfoil sections over six rotation periods, and analyzed its time-domain, frequency domain, and standard deviation. The results show that the pressure on blade surface presents highly unsteady characteristics, and rotation of the wind turbine is one of the main factors that causes pressure fluctuations on the blade surface. The standard deviation of pressure near the leading edge of each airfoil section is relatively larger, indicating that the leading edge is the most sensitive to field wind conditions. Compared with the pressure surface, the blade suction surface contributes more to wind turbine power fluctuation and is more sensitive to various unsteady sources. Moreover, the standard deviation of the pressure on the blade pressure surface is largest at the tip, and the standard deviation of the pressure on the blade suction surface is clearly divided into two regions along the chord direction with different scales at different spanwise sections.

Numerical Study of Lightning Protection of Wind Turbine Blade with De-Icing Electrical Heating System

In order to solve the problem of icing on the surface of wind turbine blade, a heating system that includes a carbon fiber net (CFN) and power cables is proposed recently. When lightning strikes at the blade with a de-icing heating system, the blade and its heating system are more easily damaged due to the overvoltage between the lightning protection system (LPS) of the blade and the heating system. In this paper, the models of a wind turbine blade with the de-icing heating system are established by Alternative Transients Program/Electromagnetic Transients Program (ATP–EMTP) and the accuracy of models is verified through an experiment. With these models, the influence of lightning current, surge protective devices (SPDs) and earthing resistance of wind turbine are analyzed by calculating the voltage between the down-conductor of the LPS and the heating system. The results show that the voltage is positively correlated with lightning current amplitude and negatively correlated with the front time of lightning current. SPDs are quite useful to reduce the voltage, and an optimal installation scheme of SPDs is obtained by simulation. It is noted that voltage decreases slightly with the increasing earthing resistance with the optimal installation scheme of SPDs.

Wind turbine blade surface inspection based on deep learning and UAV-taken images

As a key component of wind turbines (WTs), the blade conditions are related to the WT normal operation and the WT blade inspection is a significant task. Most studies of WT blade inspection focus attention on acquired sensor signal processing; however, there exist problems of stability, sensor installation, and data storage and processing. Onsite visual surface inspection is still the most common inspection method, but it is inefficient and requires a long downtime. Aimed at solving the above issues, a novel blade inspection method based on deep learning and unmanned aerial vehicles is proposed. Since common defect types are visible, the inspection problem is regarded as an image recognition problem. Three convolutional neural networks are trained by using the constructed dataset for image recognition, and the F1-score is applied to evaluate the models. The VGG-11 model is chosen for the final model due to its best performance. Then, the alternating direction method of multipliers algorithm is employed to compress the model to reduce the requirements on hardware devices. The blind area of the WT can be reduced, the efficiency of subsequent maintenance can be improved, maintenance costs can be reduced, and the economic performance can be increased. Finally, a comparison experiment of different inspection methods is carried out to demonstrate the proposed advantages.