Daoyuan Liu

To address the challenges of low detection accuracy and high missed detection rates in insulating oil leakage detection for power equipment—arising from small and densely distributed oil stains, structural occlusion, and complex background interference—this paper proposes a detection method based on an enhanced YOLOv11s (You Only Look Once version 11 small) architecture. First, a dedicated dataset is constructed, encompassing four representative scenarios—small object detection, complex background, multi-object detection and equipment occlusion—to evaluate detection performance. Second, in terms of network design, a proposed attention module, SimAMWS (Simple Attention Module With Slicing), is introduced. This module enhances the model’s sensitivity to subtle and irregular oil stains by utilizing slicing operations and localized energy-based weighting. For bounding box regression, a U-IoU (Unified Intersection over Union) loss is adopted, which incorporates a dynamic scaling mechanism during training to enable the model to focus more effectively on high-quality candidate boxes—leading to improved localization accuracy, particularly suited to the characteristics of oil leakage. Finally, comparative experiments are conducted against mainstream object detectors including SSD (Single Shot MultiBox Detector), Faster R-CNN (Region-based Convolutional Neural Network), YOLOv5s, YOLOv8s, and the baseline YOLOv11s. The proposed method achieves an mAP@0.5 (mean Average Precision at IoU = 0.5) of 97.7% and an mAP@0.5:0.95 of 66.9%, with an inference speed of 96.4 FPS. These results demonstrate that the proposed model delivers higher detection accuracy while maintaining high inference efficiency, making it well-suited for real-time oil leak detection in power equipment and supporting the development of intelligent operation and maintenance systems in the power industry.

Mechanical failures frequently occur in On-Load Tap Changers (OLTCs) during operation, potentially compromising the reliability and stability of power systems. The goal of this study is to develop an intelligent and accurate diagnostic approach for OLTC mechanical fault identification, particularly under the challenge of non-stationary vibration signals. To achieve this, a novel hybrid method is proposed that integrates the Gazelle Optimization Algorithm (GOA), Feature Mode Decomposition (FMD), and a Transformer-based classification model. Specifically, GOA is employed to automatically optimize key FMD parameters, including the number of filters (K), filter length (L), and number of decomposition modes (N), enabling high-resolution signal decomposition. From the resulting intrinsic mode functions (IMFs), statistical time domain features—peak factor, impulse factor, waveform factor, and clearance factor—are extracted to form feature vectors. After feature extraction, the resulting vectors are utilized by a Transformer to classify fault types. Benchmark comparisons with other decomposition and learning approaches highlight the enhanced performance of the proposed framework. The model achieves a 95.83% classification accuracy on the test set and an average of 96.7% under five-fold cross-validation, demonstrating excellent accuracy and generalization. What distinguishes this research is its incorporation of a GOA–FMD and a Transformer-based attention mechanism for pattern recognition into a unified and efficient diagnostic framework. With its high effectiveness and adaptability, the proposed framework shows great promise for real-world applications in the smart fault monitoring of power systems.