Weili Wu

The integration of distributed generation (DG) changes the fault current path, magnitude, direction, and transient characteristics of distribution networks, which increases the difficulty of fault type identification. In particular, weak fault features and high-frequency transient components may reduce the reliability of traditional feature-based diagnosis methods. To improve the representation and classification capability of fault signals, this paper proposes a fault type identification method based on wavelet packet transform and an improved CNN-BiGRU model with a channel attention mechanism. First, three-phase voltage, three-phase current, and zero-sequence voltage signals are decomposed by wavelet packet transform, and the corresponding time–frequency matrices are constructed. Then, these matrices are integrated and converted into time-frequency images, so that multi-source fault information can be represented in a unified form. On this basis, CNN is used to extract local spatial features from the time-frequency images, while BiGRU is employed to capture bidirectional dependency information of fault features. Furthermore, a channel attention mechanism is introduced to enhance informative feature channels and suppress redundant information, thereby improving the fault classification performance. Simulation results based on a 10 kV DG-integrated distribution network show that the proposed method achieves high recognition accuracy under different DG capacities and access configurations. Compared with CNN, BiGRU, and CNN-BiGRU models, the proposed CNN-BiGRU-Attention model shows better classification accuracy and adaptability, demonstrating its effectiveness for fault type identification in active distribution networks.

Aiming at the problem that the multi-scale feature interaction ability of the traditional deep learning-based line selection algorithm is insufficient, resulting in the decline of line selection accuracy, a multi-scale feature fusion line selection method based on transfer learning is proposed, abbreviated as TLM-Net. Firstly, to address the issue of the insufficient generalization ability of the line selection network in small-sample scenarios, a simulation data pre-training framework is constructed, and a robust feature representation basis is established through a cross-domain knowledge transfer mechanism. Secondly, aiming at the problem of insufficient extraction of feature information by traditional algorithms, a multi-scale feature fusion network (MFFN) is designed to integrate global context information and local detail features, achieving cross-level semantic complementarity and spatial alignment optimization. Then, to enhance the representation ability of weak fault feature information, an EKA mechanism integrating variable kernel convolution is designed. The background interference is reduced through adaptive multi-region feature focusing, and the edge recognition accuracy of the model for irregular targets is improved. Finally, the pre-trained model is transferred to the target domain by adopting the transfer learning strategy, and the network parameters are fine-tuned in combination with the on-site data to achieve cross-domain adaptation of the feature space. The experimental results show that the TLM-Net algorithm’s mAP@0.5 reaches 98.5%, the accuracy rate and recall rate reach 98.3% and 96.5%, respectively, and the accuracy is improved by 37.5% compared with the original model.

Aiming at the problem that fault characteristics cannot be effectively expressed due to the low pixel proportion of the hot spot target and background interference when detecting hot spot faults in complex environments, a photovoltaic module hot spot fault detection method integrating U-Net and YOLOv8 is proposed. Firstly, the U-Net segmentation network is introduced to remove pseudo-high-brightness heat sources in the background and highlight the contour features of the photovoltaic panels, laying a good foundation for the subsequent photovoltaic hot spot fault detection tasks. Secondly, a detection network is built based on the YOLOv8 framework. Aiming at the problems that it is difficult to extract the hot spot features of photovoltaic panels of different sizes and to balance the reasoning speed and detection accuracy, a detection network based on deformable convolution and GhostNet is designed. Furthermore, to enhance the adaptability of the convolutional neural network to multi-scale hot spot targets, deformable convolution (DCN) is introduced into the YOLOv8 network. By adaptively adjusting the shape and size of the receptive field, the detection accuracy is further improved. Then, aiming at the issue that it is difficult to balance accuracy and speed in the detection network, the C2f_Ghost module is designed to simplify the network parameters and improve the model inference speed. To verify the effectiveness of the algorithm, a comparison is made with SSD, YOLOv5, YOLOv7, and YOLOv8. The results show that the proposed algorithm can accurately detect hot spot faults, with an accuracy of up to 88.5%.