SPDNet: A Real-Time Passenger Detection Method Based on Attention Mechanism in Subway Station Scenes

In order to implement real-time detection of passengers in subway stations, this paper proposes the SPDNet based on YOLOv4. Aiming at the low detection accuracy of passengers in the subway station due to uneven light conditions, we introduce the attention mechanism CBAM to recalibrate the extracted features and improve the robustness of the network. For the crowded areas in the subway station, we use the K-means++ algorithm to generate anchors that are more consistent with the passenger aspect ratio based on the dataset KITTI, which mitigates the missing caused by the incorrect suppression of true positive boxes by the Nonmaximum Suppression algorithm. We train and test our SPDNet on the KITTI dataset and prove the superiority of our method. Then, we carry out transfer learning based on the subway surveillance video dataset collected by ourselves to make it conform to the distorted passenger targets under the angle of the surveillance camera. Finally, we apply our network in a Beijing subway station and achieve satisfactory results.

Deformation Control Analysis of Unequal-Length Bridge Piles of Urban Rail Transit Tunnel Crossing Urban Expressway Elevated Bridge

Taking the viaduct project of the main road bridge of Fushi road between Jinding Street station and Jinanqiao Station of line 11 of Beijing subway as the background, due to geological reasons, the length of the pile of the main road bridge of Fushi road is 60m to the west of Xin’an road in the north, the pile length of the bridge east of North Xin’an road is 21.5m, and line 11 of Beijing subway passes between two piles of unequal length. This paper simulates the whole construction process with the method of numerical analysis, analyzes the causes of the deformation of the bridge piles by calculation, puts forward the measures of grouting reinforcement in the whole section, sums up the deformation rules of the unequal length bridge piles, and puts forward the deformation control values, in order to ensure the safety of Rail Transit Construction, for the follow-up similar projects to provide a basis and reference.

Integrated Train Timetabling and Rolling Stock Circulation Planning for a Metro Line with Multiple Depots

With the expansion of metro networks, it is common that a metro line is equipped with multiple depots. The operation costs of rail operators and the passenger satisfaction are highly dependent on train timetables and rolling stock circulation plans, which are closely related to each other. Therefore, it is important to investigate the integrated train timetabling and rolling stock circulation planning problem for metro lines with multiple depots. A mixed integer linear programming formulation is proposed to generate train timetables and rolling stock circulation plans simultaneously for a metro line with multiple depots, in which the capacity of each depot and the rolling stock balances between depots are considered. Several numerical experiments based on real-world data of Beijing Subway Line 5 and Beijing Subway Line 6 are carried out to demonstrate the effectiveness of the presented model.

Flood Risk Assessment of Subway Systems in Metropolitan Areas under Land Subsidence Scenario: A Case Study of Beijing

Flooding is one of the most destructive natural events that severely damage the ground and inundate underground infrastructure. Subway systems in metropolitan areas are susceptible to flooding, which may be exacerbated when land subsidence occurs. However, previous studies have focused on flood risk evaluation on regional/watershed-scales and land subsidence monitoring in plains, instead of on subway flood risk evaluation and how land subsidence aggravates the flood risk in subway systems. Using the proposed risk indicators and field survey data, we present a method assessing the flood risk of metropolitan subway systems under a subsidence condition based on the fuzzy analytic hierarchy process (FAHP) combined with a geographic information system (GIS). We use the regional risk level within the 500 m buffer zone of the subway line to depict the flood risk of the subway system. The proposed method was used to evaluate the flood risk of the Beijing subway system. The results show that the flood risks of the Beijing subway show a ring-like distribution pattern—risk levels decreasing from the central urban area to the suburbs. Very high and high risks are mainly located within third and fourth ring roads, accounting for 63.58% (29.40 km2) and 63.83% (81.19 km2) of the total area. Land subsidence exacerbated the Beijing subway system’s flood risk level—the moderate to very high risk increased by 46.88 km2 (16.33%), indicating that land subsidence is an essential factor affecting the flood risk level of subway systems. In addition to enhancing flood warnings, future subway flooding could be reduced by elevating the height of the stations’ exit (entrance) and installing water stop plates and watertight doors. This study is of great significance for flood warning and prevention in the Beijing subway system; it provides a theoretical basis for flood risk evaluation in other metropolitan areas.

Life cycle carbon emission of monorail transit and its comparison in operation stage with other city transit modes

In this paper, carbon emissions of a monorail transit are calculated using the method of whole life cycle, which can be divided into four stages: material manufacturing stage, construction stage, operation stage and demolition stage. In the operation phase, the units are PKT (Passenger Kilometers Travelled) and VKT (Vehicle Kilometers Travelled), while in other three stages, the unit is 1 km. The results show that the carbon emissions from the 1km length monorail are 6271.204 tons. In addition, in the operation stage, it is found that the PKT index and VKT index of Chongqing monorail transportation are 0.07468 and 3.31933 respectively, far lower than those of subways in the same city. For PKT indicators of other rail transits, from small to large, they are light rail, tram, subway, APM and maglev. As for VKT indicators, from small to large, they are tram, light rail, subway, APM and maglev. The PKT index of Beijing subway is the lowest compared with that of other cities.