Shibin Gao

The relatively poor power quality and low capacity utilization in the existing railway traction power supply systems (TPSSs) are the pain points for the power system and railway sectors. To address this issue, this article proposes a railway virtual power plant (VPP) based on a multiphase flexibly interconnected TPSS (FITPSS) integrated with multiple photovoltaics (PVs) and energy storage systems (ESSs). First, the concept of railway VPP is introduced. Then, carbon emission reduction (CER), peak load shifting (PLS), and frequency modulation (FM) regulation modes tailored for this railway VPP are proposed, in order to effectively achieve the multifunctional demand response of an ac electrified railway to the electricity and carbon markets. To adapt to the real-time dispatching requirements of the railway VPP, a rolling scheduling approach is designed, incorporating a composite prediction error (PE) in the system and an adaptive sliding window. To achieve real-time determination of each reference power for controllable devices, the railway VPP scheduling models are developed for the optimal power flow control of the FITPSS. Finally, the effectiveness and feasibility of the proposed method are verified based on data from an actual electrified railway. A comparative assessment highlights the technical-economic advantages of the proposed strategy over literature methods.

The quality of contact between a high-speed train’s pantograph and the overhead contact system (OCS) is crucial for reliable current collection. One common disturbance, contact line irregularity, can degrade this contact quality. Whereas previous studies considered only single-pantograph operation, this work investigates the dynamic performance of a double-pantograph setup under realistic contact line irregularities. Irregularities in contact wire height were measured from a newly built high-speed line (350[ ]km/h design speed) and incorporated into a finite-element model via a shape-finding procedure. Unlike a single-pantograph scenario, the double-pantograph configuration exhibits unique dynamic interactions: the leading pantograph’s motion induces oscillations in the contact wire that affect the trailing pantograph. Quantitatively, the presence of measured contact line irregularities increases the contact force variation (standard deviation) of the leading pantograph by approximately 20% on average, about twice the increase observed for the trailing pantograph (10%). At the maximum operating speed of 350[ ]km/h, for example, the leading pantograph’s contact force standard deviation rises by 35% due to irregularities, whereas the trailing pantograph’s rises by 14%. Moreover, the trailing pantograph experiences a notable rise in contact loss occurrences under irregularities (e.g. its contact loss rate grows from 1.35% to 2.05% at 350[ ]km/h), a phenomenon absent in single-pantograph operation. These results highlight that even small, inevitable contact line irregularities can significantly impact double-pantograph systems — in ways differing from a single-pantograph case — underscoring the importance of accounting for such irregularities in the design and analysis of high-speed OCS.

High-speed railway traction substations are critical facilities in high-speed rail systems, responsible for providing stable electric power to high-speed trains through the power supply system. However, anomalies endangering the traction substation occur from time to time. Only by detecting anomalies promptly, timely and targeted countermeasures can be adopted. Therefore, timely detection of anomalies is essential for the safe operation of traction substations. Existing methods can only detect known anomalies, however, anomaly types are diverse and abnormal data is difficult to collect, resulting in high uncertainty of anomalies. To address the issue of anomaly uncertainty in visual anomaly monitoring of traction substations, this paper proposes an anomaly detection method based on out-of-distribution detection through joint entropy-energy optimization. Specifically, we construct an Anomaly Scoring Network (AS-Net) to jointly model semantic, entropy and energy feature spaces. This allows more effective characterization of data distributions and differentiation between in-distribution normal data and out-of-distribution anomalies. We further construct the Hybrid Feature Anomaly Detection Network (HFAD-Net) to fuse information from different perspectives of the AS-Net and detect anomalies. Experimental results on traction substation dataset demonstrate that the proposed method can improve anomaly detection performance, with the AP increased by 30.81% and FPR95 reduced by 27.69% on average compared to other training methods. The final anomaly detection AP reaches 61.67%, surpassing multiple comparison methods. We also validated the generalizability and effectiveness of the proposed method on three benchmark datasets. Overall, this paper provides an efficient and accurate anomaly monitoring method for traction substations.

The conventional traction power supply system (TPSS) is limited in its ability to transport energy across regions due to the presence of section posts. In contrast, flexible TPSSs enable system-wide utilization of energy. However, electric locomotives face complex working conditions and experience drastic power fluctuations, making it crucial to address the efficient utilization of energy from multiple traction substations (SSs). This article proposes a multiagent-game-based reinforcement learning (MAG-RL) energy management strategy to facilitate collaborative energy interaction among neighboring SSs. Specifically, a Markov decision process for the energy management process is first established. On this basis, a distributed RL training framework is constructed to reduce the dimensionality of the state space. The multiagent energy game model is also constructed by analyzing the operation mechanism under different operation modes. Additionally, a sequential negotiation method is presented to quickly solve the energy game model. Real-time simulation platform testing indicates that the proposed method reduces the number of convergence iterations by 28.2% compared to the traditional Q-learning approach. Compared to independent operation, the utilization efficiency of regenerative braking energy improves by 30.04%, reaching 93.28%, demonstrating a technical advantage over other strategies.

The pantograph-catenary system (PCS) is a key equipment for electric trains to obtain electrical energy from the traction power supply system. As an abnormal phenomenon in the PCS, pantograph-catenary arcing directly affects the current collection quality and operational safety of electric trains. Therefore, it is very important to achieve precise detection of pantograph-catenary arcing. At present, aiming at the difficulties in detecting small arcing and the poor performance of arcing detection in complex environments in the visual detection task of pantograph-catenary arcing, a semantic segmentation model ArcSE based on feature enhancement is proposed. This model designs a pantograph-catenary arcing segmentation model that includes semantic feature branches, detail feature branches, and feature enhancement mechanisms. To address the difficulty of detecting small arcing, a dual branch structure is designed, which utilizes the semantic information extracted from the semantic feature branch to adjust the detail feature map in the detail feature branch, filters out interfering features, and retains small arcing features. Aiming at the difficulty of identifying arcing targets in complex scenes, a feature enhancement mechanism is designed, merging the arcing features with the detailed features at different scales through a multi-scale features fusion strategy. At the same time, based on the learnable visual center module, the difference between arcing features and background features is further strengthened, effectively improving the robustness of the model. Experiments were conducted on the constructed dataset to validate the effectiveness of ArcSE, with a segmentation accuracy of 89.70% and an inference speed of 14.91 ms.

The use of distributed photovoltaic (PV) for energy sharing is a promising solution to curb energy poverty. However, due to financial barriers, spatial issues, and regulation challenges, energy sharing potentials for devising an energy supply alternative are yet to be manifested and implemented in practice. In this paper, we propose an equitable energy sharing model for effective hardware sharing in which residential users engage the on-demand PV hardware through shareholding transfers, and the energy generated by shared PV is shared via peer-to-peer (P2P) transactions and local microgrid markets. A coordinated optimization model, which considers both energy and economic benefits, is proposed for maximizing the residential user's profits. The numerical simulation results based on the IEEE 33-bus system demonstrate that the proposed mechanism can increase the overall PV utilization by 15% and lower the residential user's daily average energy cost by 40%.

This letter proposes graph convolutional neural networks (GCN) with self-supervised learning to identify the critical branches comprehensively in power systems. Firstly, we construct a two-layer graph data model comprehensively considering the propagation properties and topological properties. Secondly, we model a GCN consisting of two convolution layers and a linear layer to extract the features of the proposed data model. Furthermore, we adopt a self-supervised learning method based on different training strategies to train unlabeled data to identify the critical branches. The simulation of the IEEE 39-bus system verifies the effectiveness of proposed method.

The efficiency of current collection of high-speed trains relies on the interaction performance of the pantograph-catenary. The geometry deviation is inevitable for the catenary even built at a high standard. To evaluate the impact of geometry deviation in the interaction performance of pantograph-catenary, focusing on speeds exceeding 400 km/h, which characterise the next generation high-speed rail. Realistic geometry deviation data of catenary from a newly-built high-speed line is collected. A numerical model is used to simulate the interaction of pantograph-catenary, comparing results with ideal design and realistic geometries. Field tests and simulations at speeds above 400 km/h highlight the significant effect of geometry deviation, with a 15% error when not considered. The quantitative analysis reveals that at speeds below 400 km/h, the difference in results between ideal and realistic geometries is minimal (below 5%), but this disparity increases to over 14% at 440 km/h. These findings emphasise the need to incorporate geometry deviation in the design of 400 km/h and above railway catenary systems, indicating a necessity for updated assessment and validation standards at such speed levels