An method for power lines insulator defect detection with attention feedback and double spatial pyramid

Chen, Jiangli; Fu, Zhangjie; Cheng, Xu; Wang, Fan

doi:10.1016/j.epsr.2023.109175

Cited by 20 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method should be possible to implement on UAVs and, as a result, should allow for improving the insulator’s diagnostics process. According to [ 56 ], in recent years, with the wide application of UAVs in electric power line inspection, inspection efficiency has been promoted rapidly. Likewise in [ 57 , 58 , 59 , 60 , 61 ] were presented arguments for the various possibilities of using UAVs in the power lines diagnostics.…”

Section: Discussionmentioning

confidence: 99%

Detection of Power Line Insulators in Digital Images Based on the Transformed Colour Intensity Profiles

Tomaszewski

Gasz

Osuchowski

2023

Sensors

View full text Add to dashboard Cite

Proper maintenance of the electricity infrastructure requires periodic condition inspections of power line insulators, which can be subjected to various damages such as burns or fractures. The article includes an introduction to the problem of insulator detection and a description of various currently used methods. Afterwards, the authors proposed a new method for the detection of the power line insulators in digital images by applying selected signal analysis and machine learning algorithms. The insulators detected in the images can be further assessed in depth. The data set used in the study consists of images acquired by an Unmanned Aerial Vehicle (UAV) during its overflight along a high-voltage line located on the outskirts of the city of Opole, Opolskie Voivodeship, Poland. In the digital images, the insulators were placed against different backgrounds, for example, sky, clouds, tree branches, elements of power infrastructure (wires, trusses), farmland, bushes, etc. The proposed method is based on colour intensity profile classification on digital images. Firstly, the set of points located on digital images of power line insulators is determined. Subsequently, those points are connected using lines that depict colour intensity profiles. These profiles were transformed using the Periodogram method or Welch method and then classified with Decision Tree, Random Forest or XGBoost algorithms. In the article, the authors described the computational experiments, the obtained results and possible directions for further research. In the best case, the proposed solution achieved satisfactory efficiency (F1 score = 0.99). Promising classification results indicate the possibility of the practical application of the presented method.

show abstract

Section: Discussionmentioning

confidence: 99%

Detection of Power Line Insulators in Digital Images Based on the Transformed Colour Intensity Profiles

Tomaszewski

Gasz

Osuchowski

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Insulators, those indispensable and pivotal components of a transmission line, 1 are deemed vital for the optimal functioning and sustainability of the power system. Nevertheless, their exposure to the protracted and deleterious influences of the harsh and demanding environment, characterized by high temperature, icing pressure, and wind erosion, among other factors, can result in detrimental outcomes, namely the occurrence of faults and defects, culminating in power system failures of catastrophic proportions 2 . Given the complexity of the power system transmission line environment, 3 traditional detection methods are slow and have low accuracy.…”

Section: Introductionmentioning

confidence: 99%

“…Nevertheless, their exposure to the protracted and deleterious influences of the harsh and demanding environment, characterized by high temperature, icing pressure, and wind erosion, among other factors, can result in detrimental outcomes, namely the occurrence of faults and defects, culminating in power system failures of catastrophic proportions. 2 Given the complexity of the power system transmission line environment, 3 traditional detection methods are slow and have low accuracy. Consequently, the process of detecting and addressing insulator defects has become fraught with challenges and obstacles.…”

Section: Introductionmentioning

confidence: 99%

“…This algorithm enables the selection of the center point closest to the global optimal solution and thus elevates the accuracy and efficiency of object detection. (2) The CBAM attention mechanism is incorporated into the backbone and head structure of YOLOv7 to bolster the feature extraction capability of the network even further. (3) The wise-IoU loss 29 is utilized as a loss function calculation, which leads to an improvement in the positioning performance of the model.…”

Section: Introductionmentioning

confidence: 99%

“…2. The whole network model structure is divided into input, backbone network, neck network, and head network four parts.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Improving insulator fault detection with effective-YOLOv7 network

Ye,

Liu,

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

Ensuring reliable and steady operation of power equipment is paramount to safeguard the livelihoods and labor of the populace. However, traditional detection techniques face obstacles when adapting to the intricate background of transmission lines, leading to innumerable incorrect and missed detections. To resolve these issues, an improved YOLOv7 transmission line insulator defect recognition algorithm has been introduced. An anchor frame, matching the size of the transmission line fault, has been constructed using K-means++, followed by a convolutional block attention module to enhance the ability to extract features. Next, the wise-IoU loss function has been incorporated, providing a gradient gain allocation strategy, thereby enhancing the positioning performance and speed of the model. Finally, the SiLU activation function has been replaced with meta-ACON to enhance the adaptive feature capability of the YOLOv7 network. Experimental results have shown that the proposed method has an average accuracy (mAP) of 91.8%, and the detection accuracy for transmission lines can be improved to 98.8%. This model resolves the persisting issues of erroneous and missing detections by addressing the traditional technical difficulties of detecting complex backgrounds and defects with insufficient accuracy.

show abstract