Ensuring financial sustainability through financial digital transformation constitutes an urgent requirement for urban rail transit enterprises, with business-finance integration serving as an effective enabler for such transformation. Addressing existing research gaps—where digital transformation studies in the rail transit industry predominantly focus on technological applications in specific operational scenarios but lack systematic financial digital transformation frameworks, while business-finance integration research lacks empirical validation in rail transit contexts and emphasizes unidirectional approaches—this study employs a case analysis methodology examining Guangzhou Metro's practical implementation to investigate pathways and outcomes of bidirectional business-finance integrated financial digital transformation. That findings demonstrate that synergistic advancement of conceptual/systemic reforms, methodological transformation, and ecological restructuring effectively promotes organizational integration, information integration, and value convergence between business and finance functions, yielding significant outcomes: enhanced accounting information quality, development of cross-functional finance talent, strengthened business support, and deepened strategic alignment; financial processing efficiency improvements are manifested in a 5-day reduction of month-end closing cycles, 44% increase in per capita transaction document processing volume, and 52% rise in per capita voucher creation volume; cost-effectiveness is demonstrated through an 8.6% labor productivity gain and 9.2% reduction in cost-to-revenue ratio; the pathway “top-down design leadership→bidirectional business-finance integration→ enhanced data governance→phased ecosystem synergy” proves effective in advancing financial digital transformation. This research establishes a theoretical framework and practical paradigm for urban rail transit enterprises to advance financial digital transformation.

To address technical challenges in capacity configuration, safety design, energy management, and energy-saving evaluation for urban rail transit regenerative braking energy utilization systems, this paper proposes the new-generation stored regenerative braking energy utilization system, eRAM. We develop: (1) capacity optimization model and algorithm based on parameter decomposition; (2) a multi-level safety architecture considering rail transit application scenarios; (3) an interval energy dynamic allocation algorithm based on key parameter identification; (4) distributed cooperative control strategy based on deep reinforcement learning; and (5) data-driven condition monitoring and life prediction of energy storage components based on data driven. We present the basic principles and key technologies of each major component and evaluate their demonstration performance on the Beijing Metro Batong Line, providing a useful reference for advancing regenerative braking energy recovery and utilization in urban rail transit.

In view of the numerous controversies surrounding urban rail transit Public-Private Partnership (PPP) projects, this paper analyzes the main problems faced by PPP projects during the operation period of such projects. To ensure the stability of urban rail transit operations during the operation period and safeguard the interests of all investors, three feasible solutions are proposed from the government's perspective, with an analysis of their implementation processes as well as their respective advantages and disadvantages. The research concludes that early termination of franchise rights through equity transfer is the relatively superior approach, as it offers strong operability, does not affect normal operations, helps maintain workforce stability, imposes relatively low fiscal pressure on the government, and aligns with policy requirements. Finally, taking China's first urban rail transit PPP project that was terminated during the operation phase—the Fuzhou Metro Line 2 electromechanical equipment PPP project—as a typical case, the study provides a detailed account of the entire process of successfully repurchasing the equity held by private investors and terminating the franchise agreement in advance, with the aim of offering reference and guidance for equity repurchase and exit practices in urban rail transit PPP projects.

In the operation of ultra-large-scale urban rail transit networks, train operation delays occur frequently. Accurately and efficiently predicting the number and scope of affected passengers is crucial for network command and dispatch. Based on the normal passenger flow situation, this paper combines the advantages of data-driven and model-driven methods and proposes a short-term prediction method for the scope and number of affected passengers under urban rail train delays. First, based on the detailed OD path data of passenger travel chains under normal conditions, by analyzing the passenger flow direction in the delayed section, the scope of affected passengers is defined and a passenger flow quantity prediction model is constructed. Second, for the affected passenger group, according to the travel choice behavior of passengers under delay conditions, the passenger impedance sensitivity coefficient is calibrated to predict the structural changes of passenger components (waiting, detouring, and loss). Finally, multiple actual train delay scenarios in the Beijing urban rail network are selected for verification. The results show that this method can be effectively applied to multiple scenarios in the Beijing urban rail network. The average relative error of the prediction during the morning peak on weekdays is 4.8%. The time-consuming for predicting various passenger flow indicators within the next 30 minutes with a 5 minute granularity does not exceed 140 seconds, which shows good universality, accuracy, and timeliness. Currently, this solution has been successfully applied in the Beijing Rail Transit Command Center, providing passenger flow risk warnings, passenger guidance, and emergency coordination support for network dispatchers.

As urban rail transit networks continue to expand, junction areas in cross-line operations have increasingly become critical bottlenecks limiting capacity improvement. To alleviate passenger congestion during peak hours, this paper proposes a virtually coupled train coordination and scheduling method that incorporates passenger service level considerations. A train dynamics model is first established based on traction-braking characteristics and state transition rules. Then, a passenger- oriented Nash equilibrium game framework is constructed by introducing passenger priority levels to achieve coordinated optimization of train operation strategies. Furthermore, a bi-objective optimization function is formulated to balance two key indicators: passenger travel time and platform crowding. Finally, taking the junction area of Chongqing Rail Transit Lines 4, Loop Line, and Line 5 as the research object, eight typical operation scenarios are constructed to conduct simulation verification of the virtual coupling strategy, moving block strategy, and First Come First Serve (FCFS) strategy, and the impact of passenger flow changes on optimization effectiveness is analyzed. Simulation results demonstrate that, under peak operating conditions, the proposed method reduces passenger travel time cost by approximately 3% and platform congestion by around 30% compared to the traditional First Come First Serve (FCFS) strategy and the moving block-based game approach, significantly enhancing transportation efficiency and passenger satisfaction in cross-line operation zones.

In view of the growing demand for vehicle-ground integrated communication services and the limitation of frequency resources, this study investigates 5G network slicing technology for signal system business transmission under the vehicle-ground wireless communication architecture. First, the study analyzes the requirements of the current vehicle-ground integrated communication services and the status of the transmission system, and proposes the feasibility of using 5G network slicing technology to address these issues. Second, the system is designed from two aspects: network scheme and slicing scheme. Finally, taking the vehicle-ground business transmission of the signal system as an example, an experimental verification of the signal system business transmission at the 160 km/h speed was conducted on Guangzhou Metro Line 18. The results show that compared with the PRB reserved slice, the probability of the PRB reserved slice with a delay less than 150ms is 7.94% higher than that of the ordinary slice, reaching 99.87%, and the packet loss rate is 1.47% lower than that of the ordinary slice, reaching 0.08%; The percentage of PRB reservation+QoS slice delay less than 150ms has increased to 99.99%, while the packet loss rate has decreased to 0.5G network slice configuration PRB reservation and increasing QoS can meet the service transmission requirements of the signal system. The research results can provide reference for the transmission of signal system services in 5G networks at 160 km/h.

In order to effectively identify and prevent and control the safety hazards of ground settlement along the subway lines and surrounding areas, it is crucial to carry out systematic settlement monitoring. In this study, the spatial and temporal evolution characteristics of ground settlement in Urumqi metropolitan area, including cumulative settlement and settlement time series data, are obtained based on the small baseline set interferometry (SBAS-InSAR) technique. At the same time, the autoregressive integral sliding average (ARIMA) prediction model is combined to predict the deformation trend of the key subsidence areas, and an in-depth analysis is carried out for the typical subsidence sections along the subway line. The results of the study show that: Shayibak district, Shuimogou district, Urumqi county and the new urban area have obvious subsidence and subsidence funnel areas, and the annual subsidence in some areas reaches -31.4～-7.3 mm. the subway line 1 and 2 have obvious subsidence in some sections, and the area of subsidence area is large, which needs to be paid attention to the road section. The ARIMA model is used to predict the future 12-period data of the two settlement areas of D4 and D5 with high prediction accuracy, which can be used as a reference for later monitoring.

Based on the shield tunneling subway crossing through the existing anchor cable group between Wenfu Road Station and Fuming Street Station of Shenyang Subway Line 3 as the background. The paper analyzed various treatment schemes for shield tunneling through anchor cable areas. Through comprehensive comparison of technology, economy, and construction period, combined with the actual project conditions, a scheme was proposed involving the mining primary lining method to excavate tunnels, remove intruding anchor cables, backfill with mortar, and subsequently conduct shield tunneling. Numerical model analyses were performed for various construction stages, and comparative analyses were conducted by integrating on-site monitoring data. The results show that the mining primary lining method can effectively remove anchor cables, and mortar backfilling eliminates obstacles and hidden dangers for subsequent shield construction. The shield advancement proceeded normally without requiring chamber opening. The surface deformation data from construction monitoring is slightly larger than that from numerical analysis, but the deformation patterns are basically consistent, both showing a trend of larger values in the middle and smaller values towards the edges. Engineering practice has proven that the method is safe and reliable, and it can provide a reference for similar projects.

To reduce the operational cost of air-conditioning systems in subway stations, a cold storage air-conditioning system using phase change materials (PCMs) is proposed. The system can operate in four modes, namely, space cooling and cold charging by chillers, space cooling by chillers, space cooling by cold discharging, space cooling by chillers and cold discharging modes. The PCM used is a commercially available inorganic material, and its phase change temperature range is 7.5℃～8.5℃ and heat of fusion is 140 kJ/kg. The PCM is encapsulated in panel containers. A series of PCM slabs are parallelly arranged in the cold storage tank. The gap between the two adjacent PCM slabs serves as the flow channel for water, which exchanges heat with PCMs to store and release cold energy. The characteristics of chillers and cooling towers are modeled with the curve fitting method using the measured performance data. A dynamic computational model of the cold storage system which consists of a heat transfer numerical model of PCM slabs and the performance curves of chillers and cooling towers is built. The simulations are performed under the measured data in a subway station in Guangzhou. A control strategy is developed, then, the performance of the cold storage system is studied and compared to the conventional air-conditioning system. The variations of the melted fractions of PCMs, the inlet temperature, outlet temperature and water flow rate of cold storage tank, chillers and cooling towers, and the input power of system versus time on May 1st in 2023 are analyzed. In addition, a economic analysis of the proposed cold storage system is performed. The results show that, under the demand tariff, 8.27% annual operational cost is saved by the cold storage system over the conventional system, with a static payback period of 6.3 years. While 13.34% can be saved by the cold storage system against the conventional system under the discounted electricity tariff, with a static payback period of 3.7 years.

To address the difficulty of on-site verification for urban rail transit rail potential and stray current prevention measures, the rail potential hardware dynamic emulator system of rail potential has attracted widespread attention. Existing passive hardware emulator schemes usually achieve dynamic adjustment of the equivalent output resistance by modulating the average insertion time of a fixed power resistor. However, due to the heating and power limitations of fixed power resistors, it is difficult to increase the power rating. Therefore, this paper proposes an active rail potential hardware dynamic emulator (ARPHDE) system. Based on the analysis of the operating principle of the ARPHDE, a coordinated control strategy for the virtual resistance converter unit (VRCU), large resistance converter unit (LRCU), and section switching switches is developed, enabling precise regulation of the equivalent virtual resistance value. Simulation results demonstrate that the ARPHDE can accurately reproduce the spatiotemporal dynamic distribution patterns of rail potential. Compared to traditional passive hardware emulator schemes, the ARPHDE system replaces fixed power resistor regulation with virtual resistance regulation, avoiding issues of the fixed power resistor heating and showing promising application prospects.