As an important component of the catenary system in electrified railways, the operating status of the catenary support components (CSC) is related to the operational safety of the railway. They are usually detected by a camera on the inspection equipment using computer vision techniques. However, current CSC detection research faces the following challenges. 1) The components in the captured images have issues of multi-scale and multiple categories; 2) Some components are covered because of the limited shooting angle; 3) The existing detectors based on CNN and Transformer are difficult to associate features between long-interval regions. To solve these problems, a novel graph-guided Mamba-DETR detector (Graph-MDETR) is proposed for detecting CSC. Firstly, a new backbone network (DAMamba-Adapter) is proposed to enhance the model’s ability to associate features between long-interval regions, and multi-scale training (MST) is used for solving multi-scale problems. Secondly, the semantic graph-guided module (SGM) and positional graph-guided module (PGM) are designed to alleviate the problems of occlusion and multi-category components by extracting prior category relationships with global semantic features and component geometric information. Finally, a CSC dataset taken by a UAV is established, and various experiments are conducted to verify the effectiveness and superiority of the proposed method.

Energy-efficient and grid-friendly railway power system is critical for the sustainable development of electrified railways. In this paper, a cascaded energy storage system (CESS) is investigated for energy efficiency and power quality improvement of the railway power system. First, the detailed operation principles of the CESS for multiple control objectives, including regenerative braking energy (RBE) utilization, reactive power compensation, and negative sequence current suppression, are analyzed. On this basis, a hierarchical power flow control strategy is developed for achieving the power flow management and control of multiple objectives for CESS. Specifically, a multi-objective power flow management strategy is designed in the system layer to allocate the power references for the CESS under sufficient and insufficient capacity scenarios. In the sufficient capacity scenario, the power references are directly obtained by the operation principles. In the insufficient capacity scenario, the power references are allocated through a three-stage optimization method considering the power shaving rate, grid power factor, and grid imbalance degree. Subsequently, the power references from the system layer are tracked in the converter layer to achieve power flow control. Finally, the effectiveness of the proposed hierarchical power flow control strategy is verified through experimental results. The results show that the proposed CESS can achieve superior performance on RBE utilization and power quality improvement under complex operating conditions in electrified railways.

The voltage imbalance and the open-circuit (OC) switch fault present significant issues for a single-phase cascaded H-bridge multilevel converter, leading to negative effects such as unbalanced dc voltages, overshoot current, and high current harmonics. To address these challenges, this article presents a novel technique for OC switch fault diagnosis and dc voltage balancing control using the space vector modulation (SVM) concept. Initially, this article provides an analysis of the SVM technique, which efficiently controls the dwelling time of the vector to generate driving signals. Leveraging the advantages of SVM, a voltage residual is employed as a detection variable. Subsequently, an active vector transition scheme is devised to identify the OC switch in a power cell-by-power cell manner. Furthermore, a prefault selection principle for redundant vectors is implemented to achieve power cell voltage balancing, considering voltage differences, and ensuring smooth voltage level transitions to minimize switching losses and voltage stress. In addition, by optimizing the vector arrangement based on the reduced vector range in the postfault condition, the proposed technique optimizes current harmonics and maximizes voltage output capability while maintaining smooth switching transitions. Finally, the effectiveness of the proposed technique is verified through the simulation and experimental results.

A cascaded H-bridge multilevel converter (CHBMC), which is composed of power cells, is an attractive ac–dc converter for medium- and high-voltage applications in terms of its modularity, improved harmonic performance, and higher rated voltage withstand capability. Nevertheless, the overall operation and stability of the CHBMC can be affected by inevitable issues, such as the open-circuit (OC) switch fault. This issue causes adverse effects. Hence, it is vital to develop an effective but straightforward method to identify the OC switches. This article proposes an OC switch fault diagnostic method, which is based on counters. These counters are defined to have a one-to-one correspondence with the CHBMC switches. First, a fault feature variable is obtained based on a current residual generated from the estimated current model and the OC switch fault output voltage conditions. Subsequently, based on a postfault logic judgment process, the output voltage levels, the driving signal combinations, and the current polarity, the counters indicate the OC switch fault and its position. The proposed method is simple and voltage level-independent, making it ideal for diagnosing multiple OC switch faults for the CHBMC in a power cell and different cells. The effectiveness of the proposed method is verified by simulation and experimental results.

Brace sleeve (BS) fasteners, i.e., nut and bolt, are small components but play essential roles in fixing BS and cantilever in railway catenary system. They are commonly inspected by onboard cameras using computer vision to ensure the safety of railway operation. However, most BS fasteners cannot be directly localized because they are too small in the inspection images. Instead, the BS is first localized for detecting the BS fastener. This leads to a new problem that the localized BS boxes may not contain the complete BS fasteners due to low localization accuracy, making it infeasible to further diagnose the fastener conditions. To tackle this problem, this article proposes a novel pipeline for BS fastener looseness diagnosis. First, the competitive deep learning model Faster RCNN ResNet101 is used to coarsely localize BSs. Second, an action-driven reinforcement learning agent is adopted to refine the coarse-localized boxes through a dynamic position searching process. Then, BS fasteners are extracted from the refined localized BS image by the deep segmentation model YOLACT++, which is fast and interpretable. Finally, a looseness diagnosis criterion based on segmented information are proposed. We evaluate the performance of submodels independently and the overall performance of the whole model on a real-life catenary image dataset collected from a high-speed line in China. The test results show that the proposed method is effective for BS looseness detection in railway catenary.

With the rapid development of deep learning technologies, researchers have begun to utilize convolutional neural network (CNN)-based object detection methods to detect multiple catenary support components (CSCs). The literature has focused on the detection of specified large-scale CSCs. Additionally, CNN architectures have faced difficulties in identifying overlapping CSCs, especially small-scale components. In this paper, a unified CNN architecture is proposed for detecting all components at various scales of CSCs. First, a detection network for CSCs with large scales is proposed by optimizing and improving Faster R-CNN. Next, a cascade network for the detection of CSCs with small scales is proposed and is integrated into the detection network for CSCs with large scales to construct the unified network architecture. The experimental results demonstrate that the detection accuracy of the proposed CNN architecture can reach 92.8%; hence, it outperforms the popular CNN architectures.