Rail Transit System Research Articles

The Noise Exposure of Urban Rail Transit Drivers: Hazard Classification, Assessment, and Mitigation Strategies

Prolonged exposure to high-intensity noise environments in urban rail transit systems can negatively impact the health and work efficiency of drivers. However, there is a lack of comprehensive understanding of the noise pattern and, therefore, effective mitigation strategies. To control the noise in urban rail transit systems, this study proposes a comprehensive noise assessment framework, including metrics such as average sound pressure level, peak sound pressure level, percentile sound pressure levels, dynamic range, main frequency component, and cumulative time energy to evaluate the noise characteristics. We also employ a density-based spatial clustering of applications with noise (DBSCAN) method to identify the noise patterns with the evaluation of their hazard to urban rail transit drivers. The results have revealed that: (1) The equivalent continuous sound pressure level (Leq) in the cab of Lanzhou Urban Rail Transit Line 1 averages 87.12 dB, with a standard deviation of 8.52 dB, which reveals a high noise intensity with substantial fluctuations. (2) Ten noise patterns were identified, with frequencies varying from 14.47 Hz to 69.70 Hz and Leq varying from 60 dB to 115 dB. (3) The major noise sources from these patterns are inferred to be the train’s mechanical systems, wheel–rail interaction, aerodynamic effects, and braking systems. Combined with the noise patterns and urban rail transit’s operation environment, this study proposes tailored mitigation strategies for applications aimed at protecting drivers’ hearing health, enhancing work efficiency, and ensuring driving safety.

State-dependent multi-agent discrete event simulation for urban rail transit passenger flow

Urban rail transit passenger flow modeling is the foundation of urban rail transit planning, design, and operation. The motivation of this paper is to accurately and efficiently simulate passenger flow dynamics in urban rail transit systems. To this end, we propose a State-dependent Multi-agent Discrete Event Simulation (SdMaDES). The state-dependence (or congestion-dependence) means that the service abilities (or travel times) of bottleneck facilities (e.g. platform screen doors and transfer corridors) are not constant and they interact with the congestion dynamics. Specifically, we first establish a modular Multi-agent Discrete Event Simulation (MaDES), which includes four types of modules (passenger, train, station, and network modules) and three types of agents (passenger, train, and station agents). The logical connections and event-triggering modes between the modules are analyzed and defined, and thirty types of events are designed. The state-dependence is then captured by an Improved Social Force Model (ISFM), which adds an autonomous obstacle avoidance mechanism. The ISFM reproduces passenger movement behavior within bottleneck facilities of urban rail transit systems at the microscopic level. These state-dependent functions or general rules are explicitly formulated by fitting the results of ISFM and are subsequently applied to the proposed MaDES model, resulting in the SdMaDES. This integration aims to enhance the accuracy of the simulation. We conducted a real case from the Chengdu Metro network. Some interesting results are found. (a) The maximum number of boarding passengers in a train carriage is a complex nonlinear function that is dependent on the state (density inside the train carriage). This challenges the linear function commonly utilized in most studies. (b) Compared to the actual data, the proposed SdMaDES model shows a cumulative error of 9.85% after data smoothing, while the conventional MaDES model exhibits a much higher cumulative error of 21.9% after data smoothing. (c) As the overall traffic demand level increases, the gap between the two simulation models’ results is getting wider and wider due to the amplified nonlinear impact of congestion.

Research on Evaluation of City–Industry Integration in Industrial Parks

The original meaning of city–industry integration should be understood as the coordination, balance, reasonable layout, and mutual support between urban production functional areas and service functional areas, which both have urban populations as their core element. The evaluation of city–industry integration in industrial parks can be carried out from two aspects: land–industry integration and residence–industry integration. The secondary indexes of the former mainly include industrial land efficiency and service sector land efficiency, while the secondary indicators of the latter mainly include supporting rail transit and the matching degree between residence and environment. The output value, land use structure, enterprise profile, employment rates, investments, air quality, rail transit system and other data points regarding sample industrial parks were collected by means of geofencing as well as through the creation of an enterprise credit information database and development area yearbook. The Analytic Hierarchy Process (AHP) combined with expert scoring was used to determine the index weights and implement the evaluation of city–industry integration. This study found that city–industry integration in Beijing and the Chengdu Economic and Technological Development Zones is at the forefront of sample industrial parks, and the entropy weight evaluation method verified this evaluation result. The analysis of the benchmark development zone of city–industry integration shows that the Chengdu model and the Beijing model are worthy of reference for growing and mature industrial parks when promoting city–industry integration.

Intelligent measurement method of transition resistance of insulation fasteners in urban rail transit system based on SSA-optimized algorithm

Rail-to-earth transition resistance characterizes the amplitude of stray current leakage in urban rail transit system using electric traction, which is critical for evaluating failure risk of metal infrastructures caused by electrochemical corrosion. However, detection method commonly used in the field exhibit measurement errors resulting from the limitation of measuring principle and measuring distance, which will seriously affect the accuracy of electrical insulation performance evaluation of the whole system. In view of this, this paper proposes a data-driven method for accurately measuring rail-to-earth transition resistance in urban rail transit system based on CDEGS-based simulation and SSA-optimized algorithm. CDEGS model was built to construct database for machine learning of SSA-RF network structure. SSA algorithm was used in the measurement model to optimize the topology of the random forest (RF) structure. Proposed ensemble in this paper was proved to conduct transition resistance regression with a mean relative accuracy rate of 98.85 %. Results of sensitivity analysis indicate that proposed measurement method shows acceptable robustness, which allows the algorithm parameters to fluctuate within a certain range when applied in the field. Besides, in the case of uniform and non-uniform insulation degradation, the proposed method shows good performance in assessing insulation performance. This provides a feasible foundation for future applications of the proposed method in failure risk evaluation of buried infrastructures.

LiDAR-Based Urban Three-Dimensional Rail Area Extraction for Improved Train Collision Warnings.

The intrusion of objects into track areas is a significant issue affecting the safety of urban rail transit systems. In recent years, obstacle detection technology based on LiDAR has been developed to identify potential issues, in which accurately extracting the track area is critical for segmentation and collision avoidance. However, because of the sparsity limitations inherent in LiDAR data, existing methods can only segment track regions over short distances, which are often insufficient given the speed and braking distance of urban rail trains. As such, a new approach is developed in this study to indirectly extract track areas by detecting references parallel to the rails (e.g., tunnel walls, protective walls, and sound barriers). Reference point selection and curve fitting are then applied to generate a reference curve on either side of the track. A centerline is then extrapolated from the two curves and expanded to produce a 2D track area with the given size specifications. Finally, the 3D track area is acquired by detecting the ground and removing points that are either too high or too low. The proposed technique was evaluated using a variety of scenes, including tunnels, elevated sections, and level urban rail transit lines. The results showed this method could successfully extract track regions from LiDAR data over significantly longer distances than conventional algorithms.

Impact of Metro Rail Transit Systems in Metropolitan Cities: A Bibliometric Analysis

The research investigates the evolving context of Metro Rail Transit (MRT) systems, using a thorough bibliometric analysis to identify patterns, trends, and significant contributors impacting scholarly debate. The study focuses on the effects of MRT systems in metropolitan cities, looking at both existing studies and new areas for future research. The primary goals are to examine the evolving publication scene, identify prominent individuals and entities, analyze keyword networks, and understand current research subjects. The study aims to provide a complete understanding of the effects of MRT systems to guide future research and inform urban planning decisions. Using bibliometric approaches, the study examines a varied dataset of scientific publications about MRT systems. The examination includes advancements in publication trends, identification of significant contributors, and analysis of keyword networks, titles, and abstracts to help untangle the complexity of MRT systems’ impact. The study uncovers dynamic shifts in scholarly contributions, emphasizing research themes and identifying significant contributors, organizations, and countries fostering creative discussions. Transit-oriented development, environmental sustainability, and socioeconomic ramifications are key topics that match global goals. The study meets its aims, offering significant insights into the various effects of MRT systems in metropolitan areas. By analyzing growing trends, identifying major contributors, and explaining key topics, the research lays the groundwork for future investigations, directing educated urban planning decision-making. The study’s future implications stem from its ability to drive further research, emphasizing the continued importance of investigating real-world consequences, ethical considerations, and fair access to emerging technologies. As urban mobility advances, this study offers the framework for informed decision-making and continuous innovation in metropolitan transit networks.

Passenger Route and Departure Time Guidance Under Disruptions in Oversaturated Urban Rail Transit Networks

The urban rail transit (URT) system attracts many commuters with its punctuality and convenience. However, it is vulnerable to disruptions caused by factors such as extreme weather and temporary equipment failures, which greatly affect passengers’ journeys and diminish the system’s service quality. In this study, we propose targeted travel guidance for passengers at different space–time locations by devising passenger rescheduling strategies during disruptions. This guidance not only offers insights into route changes but also provides practical recommendations for delaying departure times when required. We present a novel three-feature four-group passenger classification principle, integrating temporal, spatial, and spatiotemporal features to classify passengers in disrupted URT networks. This approach results in the creation of four distinct solution spaces based on passenger groups. A mixed integer programming model is built based on individual level considering the first-in-first-out rule in oversaturated networks. In addition, we present a two-stage solution approach for handling the complex issues in large-scale networks. Experimental results from both small-scale artificial networks and the real-world Beijing URT network validate the efficacy of our proposed passenger rescheduling strategies in mitigating disruptions. Specifically, when compared to scenarios with no travel guidance during disruptions, our strategies achieve a substantial reduction in total passenger travel times by 29.7% and 50.9%, respectively, underscoring the effectiveness in managing unexpected disruptions.

Metro Stations as Catalysts for Land Use Patterns: Evidence from Wuhan Line 11

Urban rail transit systems significantly influence land use patterns in newly developed areas, yet their impact on spatial organization and functional characteristics remains understudied. This research examines Wuhan Metro Line 11, analyzing land use within an 800 m radius of stations using Point of Interest data, ArcGIS spatial analysis, and locational entropy methods. The study reveals three station types, i.e., single-function, mixed-function, and underdeveloped, each exhibiting distinct spatial differentiation patterns. On this basis, a novel “core-diffusion” model emerges, with the highest density of functional elements observed at approximately 600 m from station centers, challenging conventional proximity assumptions. Three spatial organization modes are identified: single-core independent in two-level axis, single-core continuous in single-level axis, and double-core continuous in two-level axis. These findings contribute to the Transit-Oriented Development literature, offering sustainable insights into optimizing land use around metro stations in rapidly urbanizing contexts. This study also provides a methodological framework applicable to similar urban environments, enhancing the understanding of the complex relationships between metro development and surrounding land use patterns. These results have significant implications for urban planning and policy-making, particularly in emerging economies seeking to balance transit efficiency with sustainable urban growth.

Review of linear flux-switching machines for variable speed applications

Purpose This paper aims to present an overview of permanent magnet linear flux-switching machines (PMLFSM), field excited LFSM and hybrid excited LFSM (HELFSM) topologies as presented in literature for transportation systems such as high-speed trains and maglev systems. Design/methodology/approach The structural designs of different configurations are thoroughly investigated, and their respective advantages and disadvantages are examined. Based on the geometry and excitation sources, a detailed survey is carried out. Specific design and space issues, such as solid and modular structures, structure strength, excitation sources placement, utilization of PM materials, and flux leakage are investigated. Findings PMLFSM provide higher power density and efficiency than induction and DC machines because of the superior excitation capability of PMs. The cost of rare-earth PMs has risen sharply in the past few decades because of their frequent use, so the manufacturing cost of PMLFSM is increasing. Owing to the influence of high-energy PMs and magnetic flux concentration, the efficiency and power density are higher in such machines. PM is the only excitation source in PMLFSM and has constant remanence, limiting its applications in a wide speed operation range. Therefore, the field winding is added in the PMLFSM to flexibly regulate the magnetic field, making it a hybrid excited one. The HELFSM possess better flux linkage, high thrust force density and better flux controlling ability, leading to a wide speed range. However, the HELFSM have problems with the crowded mover, as PM, field excited and armature excitation are housed on a short mover. So, for better performance, the area of each excitation component has to compete with each other. Originality/value Transportation of goods and people by vehicles is becoming increasingly prevalent. As railways play a significant role in the transportation system and are an integral part of intercity transportation. So, this paper presents an overview of various linear machines that are presented in literature for rail transit systems to promote sustainable urban planning practices.

Simulation and Risk Assessment of Flood Disaster at the Entrance to a Rail Transit Station under Extreme Weather Conditions—A Case Study of Wanqingsha Station of Guangzhou Line 18

With the rapid development of urbanization and underground transportation, as well as the frequent occurrence of extreme weather conditions such as extreme rainfall, flooding disasters for rail transit are becoming severe, and need to be urgently clarified in terms of the mechanism causing them. In this study, a comprehensive model for water damage at the entrance to a rail transit station is proposed, emphasizing the entire process of extreme weather–surface ponding–underground intrusion. The model is validated by the inundation process of Line 5 of the Zhengzhou Metro during the “7.20” event and further applied to Wanqingsha Station of Guangzhou Metro Line 18 in China to determine the surrounding water depth, distribution, total water inflow volume, and water damage time under different rainfall intensities, rain patterns and protection scenarios. It was found that when rainfall reaches the level of a 1-in-2000-years event, the surface water begins to invade the internal rail transit system through the rail transit entrances. When facing extreme rainfall akin to the “7.20” event in Zhengzhou, the rail transit system in Wanqingsha Station meets a heightened risk of water damage, resulting in significantly deeper water levels compared to 1-in-5000-year rainfall event in Guangzhou and exceeds the height of the subway entrances. Analysis of the water intrusion process reveals that, as rainfall intensity escalates, the total inflow water volume into the rail transit system increases while escape time diminishes. Moreover, under identical rainfall intensity, pre-type rainfall yields the highest total water inflow, whereas mid-type rainfall exhibits the shortest escape time. Enhancing the protection conditions can markedly attenuate surface water intrusion into the subterranean rail transit system, thereby enhancing the evacuation time for individuals within the system.

Flexible self-powered triboelectric nanogenerator sensor for wind speed measurement driven by moving trains

Flexible self-powered sensors have been extensively applied to the Internet of Things, structural health monitoring (SHM), and intelligent transportation. It would be more demanding for the power supply to these sensors during the long-term maintenance of the rail transit system. The wind pressure/velocity generated by high-speed trains poses a substantial threat to safety of human, and new sensors without an external power supply should be developed to monitor wind pressure/velocity in the trackside. Flexible self-powered wind triboelectric nanogenerator (W-TENG) sensor with a single-electrode mode based on conductive hydrogel is designed to wind pressure/velocity monitoring without power supply by harvesting wind energy. It is devoted the relationship between the output voltage of the sensors and the wind pressure/velocity driven by high-speed trains. Material selection and structural design methods are adopted to enhance the energy harvesting efficiency and sensing accuracy of self-powered W-TENG sensors. Open-circuit current of 2.8 μA and open-circuit voltage of 12 V are achieved, and the output voltage signal has the linear relationship with trackside wind pressure/velocity. Field tests are implemented to evaluate the performance of self-powered W-TENG sensors in wind pressure/velocity measurement caused by moving trains, providing an idea to SHM application in intelligent transmit systems.

Application and Innovation of Information Technology in Rail Transit Courses in Higher Vocational Colleges

The innovation of information technology in the field of rail transit mainly includes big data analysis and intelligent decision making and the application of artificial intelligence technology. Through big data analysis and intelligent decision-making technology, the operational efficiency and passenger experience of rail transit can be improved. Artificial intelligence technology can realize intelligent functions such as train operation prediction, passenger demand analysis, and safety monitoring, so as to improve the safety and reliability of the rail transit system.

Novel Phononic-Like Crystal Wave Barrier Structure for Vibration Isolation

As an efficient, environmentally friendly, and economical urban public transportation tool, the subway has significant advantages compared to other urban public transportation tools and plays an irreplaceable role in the urban rail transit system. While the subway brings many conveniences to people, its operation has also brought a series of problems to urban development and residents’ lives, especially the vibration and noise problems generated by the subway have become unavoidable and urgent problems to be solved. The traditional wave barriers have a certain effect on attenuating and suppressing the vibration generated by the subway, but it has shortcomings in terms of the width and quantity of elastic wave bandgap (BG), so its application in low-frequency vibration reduction and isolation of subway systems is greatly limited. In order to broaden the elastic BG width and quantity of the wave barrier, this paper designs a novel phononic-like crystal subway wave barrier (PLCSWB) for subway vibration isolation based on locally resonant theory. Firstly, the dispersion curves of the novel PLCSWB are calculated using the improved plane wave expansion method and finite element method (FEM), and compared with the dispersion curves of traditional phononic crystal (PC) wave barriers. Secondly, the frequency response function of the novel PLCSWB is calculated using FEM to evaluate its attenuation effect on vibration, and the energy distribution characteristics at its BG boundary are analyzed. Then, the main factors affecting the BG of the novel PLCSWB are analyzed, and a spring–mass system equivalent model of the novel PLCSWB is established to theoretically estimate its BG range. Finally, the vibration data of the tunnel wall monitored on site are used to analyze and verify the vibration isolation effect of the novel PLCSWB. The results show that the novel PLCSWB opens five low-frequency BGs in the 200[Formula: see text]Hz frequency band, which increased the total width of the opened BGs by 49% compared to traditional PC wave barriers. In the frequency range of the BG, the attenuation value of the novel PLCSWB to vibration is mostly above 30[Formula: see text]dB, and the energy distribution inside the structure is mainly concentrated in the primitive cell that controls the BG, indicating good attenuation and control effects on vibration. The main factors affecting the BG of the novel PLCSWB are the density of the scatterer material, the elastic modulus, and the thickness of the wrapping layer material. Moreover, during subway operation, the novel PLCSWB has a good attenuation effect on the vibration transmitted from the tunnel wall to the soil. Among them, the maximum vibration acceleration amplitude of the novel PLCSWB with nine rows is reduced by 47%, and the vibration isolation performance is remarkable. The relevant research results of this paper provide a novel approach and method for the study of subway vibration reduction and isolation, and can also provide specific references for the design and development of wave barriers.

Research on Inverter Control Strategy Based on FPGA

Abstract This study adopts a composite repetitive controller strategy in a three-phase four-wire circuit, combining proportional resonance and repetitive control, which effectively suppresses harmonics and improves system performance. At higher switching frequencies, FPGA is used to achieve efficient control to ensure system stability and performance. This research provides new ideas for improving the performance of on-board inverters, especially in terms of high-speed control, nonlinear load and harmonic processing. SiC devices and composite control strategies provide more sustainable solutions for future rail transit systems.

The Feasibility of TOD Mode in Transportation Planning of Nantong City

The study examines the feasibility of the Transit-Oriented Development (TOD) model in the transportation planning of Nantong City. In the context of the "14th Five Year Plan for the Comprehensive Transportation System Development of Nantong City", it investigates the viability of the TOD mode within Nantong's rail transit system. This article will examine the successful TOD development cases in Shanghai, a first-tier city, and Suzhou, a new-tier city, both located in the Yangtze River Delta region. It will also compare the rail transit systems of Shanghai, Suzhou, and Nantong. Based on the current development status of rail transit in Nantong City, SWOT model is used to position the development of rail transit in Nantong City. hrough this analysis, this paper identified the advantages and disadvantages that Nantong City may face in advancing the TOD mode, as well as the potential opportunities and threats.Measures such as creating TOD channels, formulating ride discounts, expanding regional connectivity, flexible commuting time, and optimizing station facilities are proposed. hese strategies aim to promote the sustainable development of Nantong rail transit system, improve the convenience and efficiency of urban transportation, and then provide strong support for the economic and social development of Nantong city and the improvement of residents' quality of life.

Analysis of Stray Current Leakage in Subway Traction Power Supply System Based on Field-Circuit Coupling

In a rail transit system, there is a constant leakage of current from the subway rails to the earth, and these stray currents have complex propagation paths and a wide range of influence. Since no stray current collection devices are installed at subway depots, some of the stray current leaking from the mainline will converge at the depot, seriously corroding the structural reinforcement and buried metal of the station, thereby jeopardizing the normal operation of subway trains and passenger safety. In this paper, a field-circuit coupling method is proposed to analyze the current leakage and distribution law of the subway mainline and depot. It is found that the failure of the gauge block at the mainline will trigger the maximum leakage of rail current. Additionally, it is observed that the stray current distribution at the depot is mainly influenced by the operating status of the one-way conduction device (OWCD) and the change of rail potential. These results validate the applicability and effectiveness of the field-circuit coupling method proposed in this paper and provide new technical support for the study of stray current leakage distribution in subways.

Graph transformer embedded deep learning for short-term passenger flow prediction in urban rail transit systems: A multi-gate mixture-of-experts model

Urban rail transit (URT) plays a crucial role in mitigating urban traffic congestion by offering faster and higher-quality travel services. Short-term passenger flow predictions have practical significance for metro management and operation. However, the complex spatiotemporal characteristics and the relationship between entry and exit passenger flows make it challenging to detect the dynamic evolution patterns. This study proposes a Spatio-Temporal Graph Transformer (STGT) under the multi-task learning framework, utilizing Graph Transformer network and gated residual units to select and aggregate features. To account for the correlation between entry and exit passenger flow prediction tasks, the STGT model integrates a Multi-gate Mixture-of-Experts (MMoE) approach, which combines different expert networks for diverse input and explicitly learns to model passenger flow relationships in various scenarios. Metro-related characteristics such as weather conditions, train operation characteristics, and accessibility of nearby bus stops are incorporated to enhance prediction accuracy. Experimental evaluations are conducted using real-world historical passenger travel records from the Beijing subway. The results demonstrate the superior robustness and advantages of the STGT-MMoE model over basic and advanced benchmarks for passenger flow prediction tasks. These findings provide compelling evidence to address the challenges of short-term inflow and outflow prediction in urban rail transit systems.

Real-time train timetabling with virtual coupling operations on a Y-type metro line

The spatiotemporal imbalance of passenger flows is a prominent characteristic of urban rail transit systems. To match the provided transportation capacity with passenger distribution, this study considers an integer linear programming model to optimize train operation on a Y-type line, including the train timetable, rolling stock circulation plan and virtual coupling/uncoupling strategy that enables trains to switch between different compositions or configurations during operations. To enhance solution efficiency, certain integer decision variables in the model are relaxed to be continuous, and it is proved that this does not affect the optimal value of the model. To account for the dynamic nature of passenger flows, a “prediction + optimization” method with the rolling optimization framework, which utilizes real-time predicted passenger flow data to enable train operations to be performed and adjusted in response, is proposed. Three variants of the proposed model are embedded to meet the real-time requirements of operations. Numerical experiments verify the effectiveness and applicability of our proposed approach, with real-world data from Shanghai Metro Line 5. The computational results demonstrate that our method performs well under operation scenarios with both normal and abnormal passenger flows. Compared to fixed train composition, virtual coupling can perform much better in both peak and off-peak periods.

A multi-node self-powered fault detection system by triboelectric-electromagnetic nanosensors for smart transportation

The harnessing of vibrational energy is becoming increasingly pivotal in the development of intelligent rail transit systems. The integration of emerging technologies such as triboelectric nanogenerators (TENGs), electromagnetic generators (EMGs), or hybrid generators has become crucial for fault detection and energy harvesting in rail transit. This paper introduces a self-powered fault detection system (SPFDS). SPFDS combines multiple compact rotating Triboelectric-Electromagnetic Nanosensor (TENS) nodes with a deep learning-based diagnostic module to transform vibrational energy generated during train operations into electrical power and accurately identifies five distinct train bogie fault conditions. Simulations and experiments have shown that the TENS nodes, with a root mean square power of 0.21 W and a power density of 1595.7 W/m³, can efficiently detect various bogie faults. Additionally, their power output is adequate to support commercial sensors and Bluetooth modules. Through hyperparameter optimization, the diagnostic module utilizing multi-TENS nodes achieves an average diagnostic accuracy of 99.38 % for the five fault modes of freight train bogies. Implementing multiple TENS nodes in SPFDS enhances fault detection accuracy by an average of 32 % compared to a single TENS node, with a peak increase of 128 %. The multi-node TENS configuration and SPFDS's self-powered detection capabilities represent an innovative approach to complex fault detection, significantly contributing to the advancement of vibration energy harvesting and the development of distributed self-powered sensor network technologies for smart transportation.

Are Graphs and GCNs necessary for short-term metro ridership forecasting?

Short-term metro ridership prediction is of great significance to efficient and economic operation of Urban Rail Transit (URT) systems. With the popularity of Graph Convolution Networks (GCN) and Transformers, the recent notable metro ridership forecasting methods are GCN-based and Transformer-based models. However, existing methods face the following drawbacks. First, GCN-based models fail to effectively capture global spatial correlations which are significant for accurate prediction. Second, Transformer-based models are prone to loss temporal information due to the permutation-invariant and anti-order properties of the self-attention which they used for capturing temporal correlations. To overcome the drawbacks, we propose a novel sequence-to-sequence metro ridership prediction model, named SDT-GRU, with Stacked DT-GRU layers as both encoder and decoder. The core component of our model is DT-GRU, which integrates Dual-branch Transformer decoder into the GRU to effectively capture global spatial correlations and temporal correlations with Transformer decoder and GRU, separately. In particular, the DT-GRU module uses one branch Transformer encoder layer to capture spatial correlations within the same timestamp, and adopts another Transformer encoder layer to implicitly capture spatio-temporal correlations among previous timestamps. Then, outputs of the two Transformer encoder layers are fed into a GRU layer to capturing spatio-temporal patterns. To evaluate the effectiveness of the proposed SDT-GRU, we conduct comprehensive experiments on three real-world metro ridership datasets from Beijing, Shanghai and Hangzhou. Experimental results demonstrate that our SDT-GRU achieves better prediction performance than the state-of-the-art baselines.