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

Abstract

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.

A Comprehensive Review on Signal-Based and Model-Based Condition Monitoring of Wind Turbines: Fault Diagnosis and Lifetime Prognosis

“…Table 16 summarizes the advantages and disadvantages of machine vision-based monitoring and provides a list of selected references categorized according to the names of the components being monitored in wind turbines. With the fast development of computer science and optics devices in recent years, machine vision technology shows a growing potential for structural health monitoring in the coming years [46], [266], [270], [271]. Also, through the power of AI and the latest autonomous system technologies, such as those using unmanned aerial vehicles (UAVs), new horizons for the autonomous machine vision-based monitoring of wind turbines are opening [4], [268], [271]- [273].…”

“…With the fast development of computer science and optics devices in recent years, machine vision technology shows a growing potential for structural health monitoring in the coming years [46], [266], [270], [271]. Also, through the power of AI and the latest autonomous system technologies, such as those using unmanned aerial vehicles (UAVs), new horizons for the autonomous machine vision-based monitoring of wind turbines are opening [4], [268], [271]- [273]. However, to improve detection accuracy, speed, and online computational efficiency, further studies need to focus on image processing algorithms, simultaneous localization and mapping (SLAM), and machine learning (pattern recognition) for correct damage recognition.…”

A Comprehensive Review of Artificial Intelligence and Wind Energy

“…Artificial Neural Networks (ANN) consist of a set of nodes (neurons) organized and connected in layers, in which each connection is a signal processed and transmitted to the following layer until an output response is provided [8,9]. They are classified depending on the number of layers, the neurons per layer, and the connections between them, being common types the feedforward neural networks (FNN), recurrent neural networks (RNN), and convolutional neural networks (CNN) [10,11]. ANN are largely used for applications as false alarm detection [12][13][14], data analysis for fatigue estimation [8] and state classification [15], but their most common application is to detect faults in different components, being the most common the gearbox, the blades and the bearings [16][17][18].…”

Deep learning in automated ultrasonic NDE -- developments, axioms and opportunities

“…Unmanned aerial vehicles (UAVs) are used as a source of automated data acquisition, whose data are used for the detection of cracks with CNN-based DL models, while the characterisation of cracks are addressed using linear regression. Additionally, Xu et al [117] have addressed the inspection of wind turbine blades through images taken by an UAV. In this case, a CNN-based model is used for both detection and characterisation simultaneously by creating different defect classes ranging from no defect to different types of defect (e.g.…”