PD-YOLOv11: A power distribution enabled YOLOv11 algorithm for power transmission tower component detection in UAV inspection

At this stage, the inspection of transmission lines is dominated by UAV inspection. Insulators, as essential equipment for transmission line equipment, are susceptible to various factors during UAV detection, and their detection results often lead to leakages and false detection. Combining deep learning detection algorithms with the UAV transmission line inspection system can effectively solve the current sensing problem. To improve the recognition accuracy of insulator detection, the MS-COCO pre-training strategy that combines the FPN module with a cascading R-CNN algorithm based on the ResNeXt-101 network is proposed. The purpose of this paper is to systematically and comprehensively analyze mainstream isolator detection algorithms at the current stage and to verify the effectiveness of the improved Cascade R-CNN X101 model by combining the mAP (mean Average Precision) value and other related evaluation indices. Compared with Faster R-CNN, Retina Net, and other detection algorithms, the model is highly accurate and can effectively deal with the false detection, leakage, and non-recognition of the environment in online special detection. The research in this paper provides a new idea for intelligent fault detection of transmission line insulators and has some reference value for engineering applications.

To accurately complete the inspection task of transmission lines, a UAV inspection trajectory planning method for high-precision tower models is discussed. By using GIM three-dimensional modeling technology, high-precision models of power towers and transmission lines are obtained, and then the coordinates of target points during the UAV inspection process are obtained. A target model based on the inspection point coordinates is established to plan the trajectory path. A trajectory planning model based on particle swarm algorithm and beetle whisker algorithm is established to improve the algorithm model’s ability to escape from difficulties. In response to sudden obstacles during UAV inspections, a wall-walking strategy is introduced to prevent UAVs from falling into static traps. Through experimental verification, the improved algorithm can effectively reduce the flight distance by 2.81[Formula: see text]km and improve the efficiency of UAV operation. The verification of sudden static and dynamic threats has been conducted, and the experimental results show that this method can effectively ensure the effectiveness of the UAV power inspection process.

In order to overcome the low efficiency and high risk of the traditional manual inspection method of optical cable lines, this paper proposes a method of identification and positioning of engineering vehicle in UAV inspection. The method is based on the Pytroch deep learning framework and uses YOLOv4 target detection algorithm to realize intelligent detection of hidden faults. At the same time, the paper designs the positioning algorithm of the UAV inspection target based on the image coordinates of the recognition target and the multiple coordinate conversion method. Through test experiments, the results show that the AP value of engineering vehicles inspected by UAV reaches 83.28%, and the target positioning accuracy is about 20 meters, which can meet the inspection requirements. The test results show that based on the method proposed in the paper, the hidden faults of the optical cable can be accurately identified and located, which can effectively improve the inspection efficiency and reduce the workload of the inspectors.

Insulators are widely used in transmission lines due to their good insulation performance and pollution resistance. The damage to insulators will affect their service life and insulation performance, which brings potential safety hazards to transmission lines. With the replacement of manual inspection with UAV inspection, image detection technology based on deep learning is gradually combined with UAV inspection. Usually, the insulator defect target accounts for a small proportion of the acquired image. Therefore, the manuscript introduces an enhanced algorithm for detecting flaws on the surface of insulators, designated as YOLOv8-ISDD. In the backbone network, the YOLOv8-ISDD algorithm introduces reparameterized convolution (Repconv) to enhance the network’s capability in extracting features of defects; to fuse the features of different resolutions, PANet in the neck network is replaced by Bi-FPN for integrating features. Within the head network, the multi-scale detection head output is used to increase the diminutive object identification stratum and boost the algorithm’s capability to discern lesser targets. Experiments show that the mAP@0.5 of the YOLOv8-ISDD algorithm is 88.9%, which is 6% higher than that of the original YOLOv8 algorithm, indicating that YOLOv8-ISDD can detect insulator defects more accurately.

With the continuous expansion of the grid scale and the development of artificial intelligence, UAV inspection is gradually replacing manual inspections and has become a mainstream of transmission line inspections by its high efficiency, low cost and high safety, which has greatly improved the inspection efficiency and inspection accuracy of transmission lines. However, the existing UAV inspections mainly rely on manual methods, requiring staff to operate and perform a lot of interventions in the field. The operating specifications are not uniform, which brings a lot of difficulties to the subsequent intelligent identification. Therefore, this paper proposes an autonomous transmission line UAV inspection system based on the improved yolov4. First, a priori box is set through K-means clustering to enhance the size adaptability, and the improved yolov4 could identify the key structure of the transmission tower. Second, the system could move the PTZ to place the relevant structure in the center of the image to complete the collection of image information. The test results show that the system can collect relevant information of transmission line towers in a standardized manner, improve the accuracy of image collection, and expand the scope of application of autonomous drone inspections, and provide a new direction for subsequent intelligent inspections.

As the scale of the power grid continues to expand, the human-based inspection method struggles to meet the needs of efficient grid operation and maintenance. Currently, the existing UAV inspection system in the market generally has short endurance power time, high flight operation requirements, low degree of autonomous flight, low accuracy of intelligent identification, slow generation of inspection reports, and other problems. In view of these shortcomings, this paper designs an intelligent inspection system based on self-developed UAVs, including autonomous planning of inspection paths, sliding film control algorithms, mobile inspection schemes and intelligent fault diagnosis. In the first stage, basic data such as latitude, longitude, altitude, and the length of the cross-arms are obtained from the cloud database of the power grid, while the lateral displacement and vertical displacement during the inspection drone operation are calculated, and the inspection flight path is generated independently according to the inspection type. In the second stage, in order to make the UAV’s flight more stable, the reference-model-based sliding mode control algorithm is introduced to improve the control performance. Meanwhile, during flight, the intelligent UAV uploads the captured photos to the cloud in real time. In the third stage, a mobile inspection program is designed in order to improve the inspection efficiency. The transfer of equipment is realized in the process of UAV inspection. Finally, to improve the detection accuracy, a high-precision object detector is designed based on the YOLOX network model, and the improved model increased the mAP0.5:0.95 metric by 2.22 percentage points compared to the original YOLOX_m for bird’s nest detection. After a large number of flight verifications, the inspection system designed in this paper greatly improves the efficiency of power inspection, shortens the inspection cycle, reduces the investment cost of inspection manpower and material resources, and successfully fuses the object detection algorithm in the field of high-voltage power transmission lines inspection.

The inspection of power transmission and distribution systems is performed visually by foot patrolling and helicopter methods. These methods have various disadvantages such as time-consuming, high operating cost, safety issues, and improper results. To overcome this issue, power utility companies are searching for the alternatives like Unmanned Aerial Systems (UAS) or drones. UAS are safe, cost effective, and requires less time for power transmission line inspection compared to the regular methods. In this manuscript, a decisive flight path planning for UAS to visually inspect a power transmission line and towers is explained. The objective of this paper is to maximize the performance of three functions such as coverage of transmission tower, quality of captured image, and flight time. Second, proposing an automated inspection strategy for UAS to follow the overhead power transmission lines. These objectives are achieved by formulating a cost function, to convert the path planning into a safe operation for UAS. The results of Particle Swarm Optimization (PSO) and Simulated Annealing (SA) are compared to find the best path for UAS. The experimental finding shows that PSO algorithm has higher efficiency and effectiveness compared to the SA algorithm and benefits the UAS with a safe flight path for the inspection of power transmission towers.

With the continuous improvement of intelligent operation and inspection requirements of transmission lines in power system, the popularization of online monitoring of transmission lines and uav inspection, a large number of transmission line defect image data are generated. However, traditional identification technology has low efficiency and inaccurate detection in the identification of transmission line defects. Based on transmission line defect source data, automatic transmission line defect image recognition technology based on artificial intelligence is studied. The transmission line defect identification technology is optimized and improved through automatic video image defect identification and defect sample annotation technology of transmission line components. The field scene verification proves that, compared with the traditional identification technology, the proposed adaptive algorithm for the selected area of transmission line defects can effectively improve the identification efficiency of transmission line defects, and the identification accuracy is close to 100%, which provides a guarantee for the stable operation of the power transmission system in the power industry.

Fallen detection of power distribution poles in UAV inspection using improved YOLOX with particle swarm optimization

Electric power construction is related to the national economy and people’s livelihood. However, some natural disasters, such as typhoons and earthquakes, make the power poles and transmission towers of the distribution network prone to damage. In this context, we use advanced deep learning algorithm to realize automatic detection of power transmission towers. Specifically, we show fast and accurate detection of power transmission towers from video frames taken by unmanned aerial vehicles (UAVs). First of all, we film the power transmission towers using UAVs, and make the datasets through the manual annotation method of video frame taking. Then, an efficient intersection over union (EIoU) is introduced to calculate the loss of predicted box and an activation function Mish is used to replace original activation function ReLU base on YOLO-V4 algorithm. Finally, the center line of the power transmission tower can be obtained by using ResNet-50 to locate its endpoints. Combined with the center line and detected box, the tilt angle of the tower can be calculated. Via testing and comparison, our algorithm can give consideration to both speed and accuracy, which is shown to be more suitable to be applied in power grid disaster survey as compared to other approaches. We believe that this method will play a positive role in the future detection of damages in power transmission towers.

Power line detection is an important content of power system inspection, which of important significance for UAV inspection, obstacle avoidance, and visual measurement. Traditional power line detection algorithms are mostly based on straight line detection, and the power lines are further extracted from the detected straight lines. Therefore, the disturbance of background information is large, and the detection effect for curved lines is not satisfied. Classical deep learning object detection algorithms are not suitable for the detection of elongated objects such as overhead power lines. Based on the Holistically-Nested edge detection (HED) algorithm, this paper uses the Tensorflow deep learning framework to build a HED network model and train it. Through pixel-level classification of the image, the pixel-level segmentation of the wires in the image is realized. Through experimental verification, the power line detection achieves a high accuracy rate.

This paper proposes a design of wireless monitoring system of tower body running state such as tilt angle, temperature, humidity, wind speed, etc. This design adopts the structural health monitoring (SHM) techniques to monitor the state of tower, and can be applied to both the power transmission tower and the communication tower. Although the SHM has been widely applied to civil engineering and building structures subjected to various loadings, there are few applications in the running state monitoring for the power transmission and communication towers. In this study, micro-electro-mechanical system (MEMS)-based acceleration sensor is used, in which a method is employed for calculating the tilt based on the difference between the acceleration due to combination of gravity and other stresses and the acceleration due to gravity alone. The wireless system uses wireless sensor nodes to transmit the tower running state data to the monitoring server. The wireless sensor node system consists of a short-distance wireless transmission network (ZigBee 2.4GHz) and a remote telecommunication network (Global System for Mobile Communication - GSM). By so doing, the important problem about the communication distance limitation is resolved. The performance of the monitoring system is evaluated through several experiments. The experimental results indicate the wireless monitoring system can accurately monitor the tower body running state in real time.

Because of a large number of tower elements in transmission tower, and direction of cross section is complex, it is difficult to quickly set up a space model at the same direction with the actual transmission line tower structure. In this paper, using the method of vector orientation. Define tower element cross section in the local coordinate system, and by targeting vector method locating element of tower in the global coordinate system. The orientation of vector method of tower material has simple steps, easy to implement, and modelling fast. 1 Foreword Because of a large number of the tower element in the power transmission tower, section of the tower element consists of the angle steel and the steel tube etc several types, and direction of the cross section of the tower element is very complex, it is difficult to quickly set up a space model which is consistent with direction of the tower element in the actual power transmission tower structure (1) , orientation of the tower element in the power transmission tower becomes the main technical bottle neck for modelling of the spatial model of the power transmission tower. The beam unit modelling of the power transmission tower is realized if orientation of the tower element in the power transmission tower structure is solved. Finite element model of the power transmission tower generally applies the rod unit model for modelling. The rod unit model is characterized with bearing drawing and compressing and not bearing bending moment. This characteristic isn't same as the real model tower, the real model tower is also affected by the bending moment. Therefore orientation direction of the section shall be considered during modelling of the beam unit. Only when orientation of the tower element is basically consistent with actual conditions, accurate simulation effect can be obtained (2) .

The small scale of transmission towers and the environmental diversity around their situations make their detection in Synthetic Aperture Radar (SAR) images a challenging task. This paper presents a new hierarchical detection algorithm for transmission towers. First, Signal-to-Clutter Ratios (SCRs) of pixels are calculated. Afterwards, a SCR threshold is set. Since transmission towers possess strong scattering characteristics, pixels with SCRs above the threshold are considered as potential transmission tower pixels. Second, spatial densities of potential transmission tower pixels are calculated. According to the aggregation characteristics of transmission tower pixels, some potential transmission tower pixels with small spatial densities are removed. The remained potential transmission tower pixels are considered as candidate transmission tower pixels. The candidate transmission tower pixels are grouped by the nearest neighbour scheme such that in each group the distance between pixels is under a given threshold. Thus, each of the groups is viewed as a quasi-transmission tower. Convex-hulls of quasi-transmission towers are built, and then Minimum Bounding Rectangle (MBR) for each convex-hull is generated. According to the rectangle aspect ratios of MBRs, the real transmission towers are extracted. C-band HH-polarization GaoFen-3 (GF-3) amplitude images are used for experiments and four of the most popular transmission tower detection algorithms are selected as comparing algorithms to validate the proposed algorithms. The detection performance of transmission towers is evaluated with detection rate and quality factor. Experimental results verify that the proposed algorithm can efficiently and accurately detect transmission towers while maintaining the transmission tower geometry to a certain extent, which indicates that the proposed algorithm is efficient and promising.

The bird’s nest on the power transmission tower may cause the occurrence of bird flash accidents and threaten the safe and reliable operation of the power grid. The realization of the independent identification and positioning of the bird’s nest has always been a research hotspot. With the gradual deepening of the application of UAV inspection, higher requirements are put forward for the accuracy and speed of the bird’s nest recognition algorithm. This paper proposes a bird’s nest defect recognition method based on YOLOv5, which is composed of backbone network, FPN and YOLO head. After multiple rounds of training on the construction of the bird’s nest defect database and model of the transmission line, the independent identification and positioning of the bird’s nest has been realized. The results show that the recognition rate of the YOLOv5 model for the bird’s nest can reach 83.4%, and the FPS can reach 85.32. The recognition algorithm based on YOLOv5 proposed in this paper can meet the real-time and accuracy requirements of UAV inspection for target detection.