Research on Railway Passenger Flow Prediction Method Based on GA Improved BP Neural Network

This paper chooses passenger flow data of some stations in China from January 2015 to March 2016, and the time series prediction model of BP neural network for railway passenger flow is established. But because of its slow convergence speed and easily falling into local optimal solution of the problem, we propose to improve the time series model of BP neural network by genetic algorithm to predict railway passenger flow. Experimental results show that the improved method has higher prediction accuracy and better nonlinear fitting ability.

Numerical simulation on flow field,wind erosion and sand sedimentation patterns over railway subgrades

The railway subgrades in the sandy areas act as an obstacle interfering wind-blown sand,causing sand erosion and sedimentation,which can disrupt the safe and stable operation of the railway system.Most previous studies mainly focus on the flow field around railway subgrades,however,the real erosion and sedimentation patterns are rarely studied.This study aims to analyze the erosion and sand sedimentation patterns of wind-blown sand over the subgrades with different heights and steel rails using the ratio of the wall shear stress to the critical value of erosion shear stress.Results show that wind erosion near the top of the upwind slope of the embankment and the shoulder on the upwind side are more severe,and the severity increases with an increase in the height of the embankment.With the increase of wind velocity,sand sedimentation both on the windward and leeside of the subgrade decreases and wind erosion by reverse flow occur.This study indicates that railways in sandy areas should be constructed with a moderate subgrade height(4 m).

Prevention of debris flow disasters on Chengdu-Kunming Railway

Chengdu-Kunming Railway is an important transport line on southwestern China. However, this railway's safety is often threatened by debris flows. How to effectively forecast and alarm the debris flow disasters and reduce the losses is the aim to study the prevention system in this paper. The factors to cause or influence debris flow are divided into four parts-the basin environmental factors, the basin meteoric factors, the prevention work's elements and the flood-relief work's elements, and the prevention system is made up of three models-a judgment model to assess the debris flow gully's seriousness, a forecast model to predict the debris flow's occurrence and an alarm model to evaluate the debris flow's disaster. Afterwards, a concise structure chart is worked out and verified by the field data from Chengdu-Kunming Railway. This prevention system will provide beneficial reference for the debris flow's monitoring network to be executed on Chengdu-Kunming Railway.

Railway Passenger Flow Forecasting by Integrating Passenger Flow Relationship and Spatiotemporal Similarity

Railway passenger flow forecasting can help to develop sensible railway schedules,make full use of railway resources,and meet the travel demand of passengers.The structure of passenger flow in railway networks and the spatiotemporal relationship of passenger flow among stations are two distinctive features of railway passenger flow.Most of the previous studies used only a single feature for prediction and lacked correlations,resulting in suboptimal performance.To address the above-mentioned problem,we proposed the railway passenger flow prediction model called Flow-Similarity Attention Graph Convolutional Network(F-SAGCN).First,we constructed the passenger flow relations graph(RG)based on the Origin-Destination(OD).Second,the Passenger Flow Fluctuation Similarity(PFFS)algorithm is used to measure the similarity of passenger flow between stations,which helps construct the spatiotemporal similarity graph(SG).Then,we determine the weights of the mutual influence of different stations at different times through an attention mechanism and extract spatiotemporal features through graph convolution on the RG and SG.Finally,we fused the spatiotemporal features and the original temporal features of stations for prediction.The comparison experiments on a railway bureau’s accurate railway passenger flow data show that the proposed F-SAGCN method improved the prediction accuracy and reduced the mean absolute percentage error(MAPE)of 46 stations to 7.93%.

Study on the Organization Model of Wagon Flows in Railway Terminal

Railway terminal is an important part of railway network. Transport organization of railway terminal is the key of the railway transport organization. Moreover, the organization of transport work is based on the organization of wagon flows in the railway terminal. Because of the great amounts of equipment and a large number of train operations, the study on railway terminal transport organization is mostly focused on a marshalling station in railway terminal or a part of it. Systematic study taking railway terminal as a whole is very few. In this paper, the organization of wagon flows in a railway terminal is analyzed and a wagon flow model in a railway terminal is established. The main principles of organization of local trains are also presented.

A Study on the Analysis of Load Flow for Railway DC System

Urban railway systems differ greatly from general power systems in that they use direct current(DC)power supply and that the location and power requirements of the loads change.The position and power consumption of the load shall be interpreted continuously every second,or in a fixed unit of time,for a specific period of time during which the operating conditions are repeated.The additional analysis of energy-saving systems being considered as energy efficiency improvement methods requires more complex load flow analysis algorithms.Simulations are performed load flow every time step.The power of an electric railway power feeding system is the power consumed or produced by a train.Because the amount and position of the load change rapidly over time,load flow analysis continues over time.Therefore,based on the method of obtaining solutions by constructing node equations for load flow analysis in this study,load flow analysis was performed through algorithms with energy-saving systems applied.Both thetrain performance simulation(TPS)and power simulation results show that the actual measurement data are estimated almost equally.

Scheme of long distance power supply for electrified railway traction network based on traction cable

Purpose–The traction cable is paralleled with the existing traction network of electrified railway through transverse connecting line to form the scheme of long distance power supply for the traction network.This paper aims to study the scheme composition and power supply distance(PSD)of the scheme.Design/methodology/approach–Based on the structure of parallel traction network(referred to as“cable traction network(CTN)”),the power supply modes(PSMs)are divided into cableþdirect PSM and cableþautotransformer(AT)PSM(including Japanese mode,French mode and new mode).Taking cableþJapanese AT PSM as an example,the scheme of long distance power supply for CTN under the PSMs of co-phase and out-of-phase power supply are designed.On the basis of establishing the equivalent circuit model and the chain circuit model of CTN,taking the train working voltage as the constraint condition,and based on the power flow calculation of multiple train loads,the calculation formula and process for determining the PSD of CTN are given.The impedance and PSD of CTN under the cableþAT PSM are simulated and analyzed,and a certain line is taken as an example to compare the scheme design.Findings–Results show that the equivalent impedance of CTN under the cableþAT PSM is smaller,and the PSD is about 2.5 times of that under the AT PSM,which can effectively increase the PSD and the flexibility of external power supply location.Originality/value–The research content can effectively improve the PSD of traction power supply system and has important reference value for the engineering application of the scheme.

Thermal features of Qinghai-Tibet Railway Embankment with Crushed-stone Slope Protection in Permafrost Regions

The Embankment with Crushed-Stone Slope Protection(ECSSP) in permafrost regions is an effective measure to cool subgrade and protect permafrost.It can mitigate the engineering hazards of the Qinghai-Tibet railway in the permafrost regions. Considering the influence of the noctumal cold air during summer months in Qinghai-Tibet Plateau。

Assessment of Ballast Flying in the National Railway Network of Ethiopia

One of the major problems in ballasted railroads is ballast flying, which is the projection of ballast particles from the at-rest position as the train passes over the track of a railway structure, mainly due to high speed. In this research, the possibility of railway ballast flying for the double track Addis-Adama section of the new Addis-Djibouti railway line is assessed by determining the major causes of ballast flying and applying Discrete Element Modeling (DEM) with the aid of Particle Flow Code (PFC3D) software. The analysis comprised of an impact load and ballast material behavior which were used to determine the vibrational speed of individual ballast particles. The governing result from the series of discrete element analyses performed by considering fouled ballast gradation with grain-size diameter of 22.4 mm gives rise to a ballast maximum vibrational speed of 0.014 m/s. Since the ballast vibrational speed for Addis Ababa-Adama line is less than 0.02 m/s that is recommended by the literature, no ballast flight is expected under the present traffic and ballast conditions.

Medium-term forecast of daily passenger volume of high speed railway based on DLP-WNNMedium-term forecast of dailypassenger volume of high speedrailway based on DLP-WNN

Purpose – This paper aims to propose a medium-term forecast model for the daily passenger volume of HighSpeed Railway (HSR) systems to predict the daily the Origin-Destination (OD) daily volume formultiple consecutivedays (e.g. 120 days).Design/methodology/approach – By analyzing the characteristics of the historical data on daily passengervolume of HSR systems, the date and holiday labels were designed with determined value ranges.In accordance to the autoregressive characteristics of the daily passenger volume of HSR, the Double LayerParallel Wavelet Neural Network (DLP-WNN) model suitable for the medium-term (about 120 d) forecast of thedaily passenger volume of HSR was established. The DLP-WNN model obtains the daily forecast result byweighed summation of the daily output values of the two subnets. Subnet 1 reflects the overall trend of dailypassenger volumes in the recent period, and subnet 2 the daily fluctuation of the daily passenger volume toensure the accuracy of medium-term forecast.Findings – According to the example application, in which the DLP-WNN modelwas used for the medium-termforecast of the daily passenger volumes for 120 days for typical O-D pairs at 4 different distances, the averageabsolute percentage error is 7%-12%, obviously lower than the results measured by the Back Propagation (BP)neural network, the ELM (extreme learning machine), the ELMAN neural network, the GRNN (generalizedregression neural network) and the VMD-GA-BP. The DLP-WNN model was verified to be suitable for themedium-term forecast of the daily passenger volume of HSR.Originality/value – This study proposed a Double Layer Parallel structure forecast model for medium-termdaily passenger volume (about 120 days) of HSR systems by using the date and holiday labels and WaveletNeural Network. The predict results are important input data for supporting the line planning, scheduling andother decisions in operation and management in HSR systems.

Installation position determination of wind speed sensors on steel pole along a high-speed railway

In order to consider the influence of steel pole on the measurement of wind speed sensors and determinate the installation position of wind speed sensors, the flow field around wind speed sensors was investigated. Based on the three-dimensional steady Reynolds-averaged Navier-Stokes equations and k-ε double equations turbulent model, the field flow around the wind speed sensor and the steel pole along a high-speed railway was simulated on an unstructured grid. The grid-independent validation was conducted and the accuracy of the present numerical simulation method was validated by experiments and simulations carried out by previous researchers. Results show that the steel pole has a significant influence on the measurement results of wind speed sensors. As the distance between two wind speed sensors is varied from 0.3 to 1.0 m, the impact angles are less than ±20°, it is proposed that the distance between two wind speed sensors is 0.8 m at least, and the interval between wind speed sensors and the steel pole is more than 1.0 m with the sensors located on the upstream side.

Research on Optimization of Selection of Train Passing Routes and Division of Shunting Work in Railway Hub

in this paper, a general integer prosramming model is set up, and its solution ispresented to optimize the organization of wagon flows and the determination ofthe train running route as well as division of shuntins work among stations in ahub. As application example is put forword, with some conclusions reached.

Experiment and Simulation of Metal Flow in Multi-stage Forming Process of Railway Wheel

To investigate the metal flow during the railway wheel forming process, experiments and finite element method （FEM） simulation were carried out. An axisymmetric modeling for the wheel rolling process was proposed to predict the metal flow in radial direction, by which the whole multi-stage forming process could be simulated in axisymmetric and integral way. The result shows that the axisymmetric simulation method was an effective method to explore the metal flow in radial direction and to analyze the relationships of tools motion during the wheel rolling. The detail information about metal flow in railway wheel forming process was obtained. The metal in the wheel web was from the area near the half radius of the original billet; the chill zone of the billet became an envelope of the rim and part of the web with a maximum thickness of about 6 mm below the tread. At the wheel rolling stage, the metal in the rim flowed towards the web; the metal near the surfaces of the conjunction region between the web and rim suffered severe shear deformation.

Wind tunnel test on the effect of metal net fences on sand flux in a Gobi Desert, China

The Lanzhou-Xinjiang High-speed Railway runs through an expansive windy area in a Gobi Desert, and sand-blocking fences were built to protect the railway from destruction by wind-blown sand. However, the shielding effect of the sand-blocking fence is below the expectation. In this study, effects of metal net fences with porosities of 0.5 and 0.7 were tested in a wind tunnel to determine the effectiveness of the employed two kinds of fences in reducing wind velocity and restraining wind-blown sand. Specifically, the horizontal wind velocities and sediment flux densities above the gravel surface were measured under different free-stream wind velocities for the following conditions： no fence at all, single fence with a porosity of 0.5, single fence with a porosity of 0.7, double fences with a porosity of 0.5, and double fences with a porosity of 0.7. Experimental results showed that the horizontal wind velocity was more significantly decreased by the fence with a porosity of 0.5, especially for the double fences. The horizontal wind velocity decreased approximately 65% at a distance of 3.25 m（i.e., 13 H, where H denotes the fence height） downwind the double fences, and no reverse flow or vortex was observed on the leeward side. The sediment flux density decreased exponentially with height above the gravel surface downwind in all tested fences. The reduction percentage of total sediment flux density was higher for the fence with a porosity of 0.5 than for the fence with a porosity of 0.7, especially for the double fences. Furthermore, the decreasing percentage of total sediment flux density decreased with increasing free-stream wind velocity. The results suggest that compared with metal net fence with a porosity of 0.7, the metal net fence with a porosity of 0.5 is more effective for controlling wind-blown sand in the expansive windy area where the Lanzhou-Xinjiang High-speed Railway runs through.

Evaluation of embankment foundation creep in conditions of deep bedding of roof in permafrost soils

Railroad operating experience in permafrost conditions has shown that deformations of embankments on thawing foun- dations last for a long time. After an initial period of heat settlement due to permafrost degradation, the determining factor is the plastic flow of seasonal thawed soils of the foundation upper layer under the embankment. This paper provides a method to evaluate these deformations, and calculation examples using data from line sections of the Chum-Labytnangi Railway in northwestern Russia. It also discusses several methods of embankment stabilization, including the use of ver- tical thermosiphons.

Experimental definition and its significance on the minimum safe distance of blown sand between the proposed Qinghai-Tibet Expressway and the existing Qinghai-Tibet Railway

The Qinghai-Tibet Expressway is a major strategic project planned by China that will be built along the Qinghai-Tibet Engineering Corridor. At present,important traffic line projects,such as the Qinghai-Tibet Railway,have been built within this narrow corridor,particularly at the blown sand sections. How to ensure that the wind speed and its flow field between the new expressway and existing railway subgrades are not affected by each other is a priority to prevent breaking the dynamic balance of the blown sand movement of the existing subgrade,thereby avoiding aggravating or inducing new blown sand hazards and ensure the safe operation of the existing Qinghai-Tibet Railway. Therefore,defining the minimum distance of the wind speed and its flow field,which are not affected by each other,between the subgrades become a scientific problem that should be solved immediately to implement the construction of the Qinghai-Tibet Expressway. For this purpose,the minimum safe distance between the subgrades of the Qinghai-Tibet Expressway and Qinghai-Tibet Railway was investigated from the perspective of blown sand by making subgrade models for conducting wind tunnel experiments and combining the observation data of the local field. Results indicated that the minimum safe distance between the two subgrades is 45–50 times the subgrade height when the Qinghai-Tibet Expressway is located at the downwind direction of the Qinghai-Tibet Railway,and 50 times the subgrade height when the former is located at the upwind direction of the latter. These results have guiding significance for the route selection,survey,and design of the Qinghai-Tibet Expressway at the blown sand sections and for the traffic line projects in other similar sandy regions.

Macro-microscopic dynamic characteristics of the ballast bed induced by the variation of moisture content

Due to the excellent drainage performance of the ballast,existing studies mainly focus on the dynamic response of ballast under field capacity or saturation.Attention has rarely been paid to dynamic changes in moisture content and potential influences.In this article,we firstly conduct a model test to determine the variation of ballast moisture content under artificial rainfall.After that,a full-scale model test with cyclic loading is carried out to study the effect of moisture content variation on the macro-microscopic response of the ballast bed,where several wireless particle sensors are installed to obtain ballast motion characteristics at strategic locations.The results show that the moisture content increases gradually and stabilizes at a flat peak under rainfall,despite the excellent drainage performance of ballast bed.After halting rainfall,the moisture content drops back to field capacity,which indicates dynamic flowing surface water on ballast particles under rainfall.Such flowing surface water brings changes to the original dynamic equilibrium of ballast bed:macroscopically,the deformation rate of stabilized ballast bed increases significantly,reaching a local peak under field capacity;microscopically,the x-and z-angular accelerations of the ballast show positive correlation with rainfall intensity.The multiscale responses indicate that field capacity is a critical moisture content.

A Study of the Weak Shock Wave Propagating over a Porous Wall/Cavity System

The present computational study addresses the attenuation of the shock wave propagating in a duct, using a porous wall/cavity system. In the present study, a weak shock wave propagating over the porous wall/cavity system is investigated with computational fluid dynamics. A total variation diminishing scheme is employed to solve the unsteady, two-dimensional, compressible, Navier-Stokes equations. The Mach number of an initial shock wave is changed in the range from 1.02 to 1.12. Several different types of porous wall/cavity systems are tested to investigate the passive control effects. The results show that wall pressure strongly fluctuates due to diffraction and reflection processes of the shock waves behind the incident shock wave. From the results, it is understood that for effective alleviation of tunnel impulse waves, the length of the perforated region should be sufficiently long.