Search

• A robustness index considering cascading failures caused by failed units is proposed. • Robustness improvement by structure adjustment is studied for fixed characteristics. • TS with directed and undirected edge swap operators is designed to solve the problem. • A case study of structure adjustment for improving robustness in the HST is performed. • Robustness of the optimized HST is greatly enhanced and hardly affected by failures. To enhance the ability of a high-speed train (HST) to provide services under various adverse conditions, this paper studies robustness improvement of the HST by adjusting its structure. We model the HST as an interdependent machine-electricity-communication network (IMECN) composed of a machine network (MN), an electricity network (EN) and a communication network (CN), and propose a robustness metric of the IMECN subject to node failures considering failure propagation. Then, a robustness optimization model is constructed for structure adjustment. A Tabu search algorithm with directed and undirected edge exchange operators is designed to solve this problem. A case study on a practical HST is used to verify the feasibility and effectiveness of the proposed method. The results show that robustness of the IMECN is significantly improved by structure adjustment. Furthermore, most nodes have little effect on robustness, and impact of failures on any node is minimized after optimization. In addition, in terms of topology, nodes with low and high degrees in the MN and CN reconnect with those with a similar degree, their clustering coefficients become larger, and closeness of all nodes in the subnetworks increases. Finally, adjusting the structures leads to a slight difference of loads in the EN and CN. [ABSTRACT FROM AUTHOR]

• The general speed-density diagram of bidirectional pedestrian flow is proposed. • The speed-density diagram of bidirectional pedestrian flow is calibrated by the experimental data. • The pedestrian counter flow is simulated by using the collection of ordinary differential equations. • A pedestrian counter flow control model is proposed to avoid blocking in bidirectional corridors. In this paper, we develop a control model for pedestrian counter flow to avoid blocking in bidirectional corridors. Considering the conflict between two streams, we formulate a general speed-density diagram of bidirectional pedestrian flow based on the previous models in the literature. To accommodate a possible inhomogeneous pedestrian density along the corridor and inflows from various directions, the corridor is divided into several sections. A pedestrian counter flow model is developed to simulate the pedestrian flow between sections by using a collection of ordinary differential equations. The control objective is proposed based on user equilibrium and a capacity-oriented principle, and a linear feedback control model is developed to achieve a controlled inflow that reduces the density of each section to an acceptable level. The level of service of the corridor is improved, and blocking is eliminated by the proposed control model. Finally, sensitivity analysis suggests that the model performance can be improved if the conflict level is reduced. [ABSTRACT FROM AUTHOR]

• The pedestrians cell transmission model is extended to simulate the pedestrian following behavior. • The leader movement is simulated by an improved social force model. • A multi-scale model is proposed to simulate the guided crowd dynamics. • A strategic guidance model including leader location plan and exit selection is proposed. • The proposed model can suggest the optimal number of leaders and the optimal exit selection. Guidance is an efficient crowd management measure used to save lives in emergencies because it can reduce the disorientation of pedestrians. This study investigates an optimal guidance strategy for large-scale crowd evacuations. To increase computational efficiency, a pedestrian cell transmission model is extended to create a rapid simulation of guided crowd dynamics. To solve the conflicts between the limited guidance capacity and the desire to improve evacuation efficiency, a strategic guidance model is proposed, which generates a leader location and exit selection plan. A simulation algorithm is proposed to integrate a pedestrian following model and strategic guidance model based on the follower-leader interaction. Finally, a hybrid multiscale approach for modeling guided crowd evacuations is established to evaluate the performance of the guidance strategy. The experimental results show that the required CPU time of the proposed model is much less than that of the microscopic models. Because the number of leaders is minimized and the exit is selected by taking both risk and congestion into account, the obtained guidance strategy can realize the full use of guidance capacity, information confusion reduction and uniform exit usage, all of which contribute to a reduction in evacuation time. [ABSTRACT FROM AUTHOR]

• The coaxial binary composite MnO 2 /polyaniline was prepared. • The coaxial structure is convenient for the ion and electrons transportation. • The MnO 2 /polyaniline composite exhibit excellent capacitive behavior. • The assembled asymmetric supercapacitor displays high performance. • The composite can produce microsupercapacitor combining 3D-printed stamp. Controllable design over the structure and morphology of active material is highly desirable for high performance supercapacitors. Here, coaxial binary composite MnO 2 /polyaniline was prepared by introducing tetraphenyl porphyrin tetrasulfonic acid interlayer (TPPS), as a crosslinking agent and dopant, to fabricate the three-dimensional hollow polyaniline (PANI) nanotube gel and then synthesizing nanoflaky MnO 2 uniformly on the surface of PANI. Moreover, this coaxial structure is convenient for the ion and electrons transportation to exhibit excellent capacitive behavior with the capacitance of 575 F/g at 0.4 A/g. In addition, we assembled an asymmetric supercapacitor using MnO 2 /PANI as positive electrode, displaying high energy density (52.7 Wh/kg), remarkable durability and long cycling life (maintaining 90.2% initial capacitance after 2000 cycles). Using MnO 2 /PANI as inks for 3D-printed stamping is promising to be a novel fabricating method which can rapidly produce flexible microsupercapacitor for wearable micro devices. [ABSTRACT FROM AUTHOR]

• A high-speed train is modeled as an interdependent network. • A robustness metric is proposed based on network theory and practical factors. • Assessment and optimization models of resilience are constructed. • The tabu search algorithm is developed to solve the optimization problem. • The optimal repair strategy of failed units in the high-speed train is analyzed. High-speed trains (HSTs) are used to transport passengers to destinations efficiently and safely. However, when HST units fail, strategies for repairing them are not available. Hence, to optimize repair strategies, an interdependent network is used to model an HST, and a metric of resilience based on network theory is introduced. Based on practical factors, an interdependent machine–electricity–communication network (IMECN) and related cascading failure models are developed. Comprehensive robustness metrics for the IMECN and nodes are proposed considering the topology of networks and functional features of nodes. Accordingly, a resilience optimization model of the IMECN subject to disturbances is formulated and solved using a tabu search (TS) algorithm. Finally, a real-world HST is considered to analyze the optimal repair strategy for different numbers of failed nodes. Analysis results show that the recovery process under the optimal repair strategy has at most three stages. When a few nodes and less than half of the nodes fail, the highest and lowest resilience levels of the IMECN are achieved, respectively. Moreover, the characteristics of the preferentially repaired node in terms of importance, vulnerability, network type, and interdependency are analyzed. Results indicate that the optimal repair strategy is not necessarily determined by topological metrics. [ABSTRACT FROM AUTHOR]

Abstract Traction energy is the main component of railway operation energy, and a timetable that predefines the running time of train operation can be used to determine the traction energy consumption. This study proposes a bi-level model that optimizes timetables to achieve the energy-saving control of railway systems. The upper level of the model ensures the relative stability of the timetable while maintaining railway safety constraints, which makes train operations more convenient for the railway sector as well as passengers; while the lower level of the model optimizes the arrival and departure time among intermediate stations to minimize the energy consumption of each train. Then, a unified iterative optimization algorithm combining particle swarm is developed to solve the model, and a timetable that ensures energy consumption optimizations is thus obtained. A case study using actual operation data from the Beijing-Shanghai high-speed railway is developed to illustrate the proposed method. Results show that the total energy consumption is reduced by more than 7.6%, and the average adjustment time for each distance interval is approximately 1 min, which maintains the stability of the original timetable. Highlights • A bi-level model is developed to optimize railway energy and passenger convenience. • Energy saving, timetable stability, and train tracking constraints are investigated. • Spare time of timetable are optimally distributed by an iterative-loop algorithm. • Train optimal energy-saving operation strategies are proposed and verified by a case. • The model is applied in a real case saving 7.6% energy consumption. [ABSTRACT FROM AUTHOR]

Highlights • Ultrafine porous carbon/MnO 2 nanowires were expected to shorten the ion diffusion path. • The carbon aerogel derived from methyl cellulose provides a cost-consuming system. • The hierarchical structure serves as stable framework to reduce the agglomeration. • The composite can be stamped into arbitrary patterns to be applied in micro-devices. Abstract A green and scaled-up strategy has been successfully employed to fabricate asymmetric supercapacitors (ASC) in conjunction with high energy density as well as highly efficient electrode materials. In this study, a B, N-dual doped porous carbon aerogel negative and a novel ultrafine nanowire positive composite, composed of manganese oxide (MnO 2) core and carbon aerogel shell, were proposed. Methyl cellulose (MC) is spaciously available and the affluent hydroxyl groups on the molecular chains are conductive to trigger supramolecular assemble process to form self-crosslinking hydrogel. And then the hydrogel was dried and carbonized to obtain the B, N-dual doped porous carbon aerogel (BN-MC), with the specific area of 989.8 m2 g−1. Subsequently, a facile and efficient hydrothermal system, introduced to synthesize BN-MC/α-MnO 2 composite, is prone to achieve hybrid nanowires with an average width of ∼15 nm. The prepared 0.1BN-MC/MnO 2 -8 sample (0.1 M NH 4 HB 4 O 7 doping agent and 8 mg BN-MC addition) exhibits highest electrochemical behavior, whose capacitance is up to 338 F g−1 at 1 A g−1. That is ascribed to the well-dispersed α-MnO 2 nanowires and uniformly cladding BN-MC shell. The ASC, assembled with BN-MC/MnO 2 and BN-MC, significantly enhances energy density reaching to 27.75 Wh kg−1 at a power density of 1.35 kW kg−1 operating under 1.8 V voltage window. This ASC also shows superior cycling stability for maintaining 95.6% initial capacitance after 2000 cycles. What's more, the BN-MC/MnO 2 can be stamped into arbitrary patterns in seconds. This provides a new rapid prototyping technology for fabricating flexible and portable micro-supercapacitor devices. [ABSTRACT FROM AUTHOR]

• A high-speed train is modeled based on network theory. • A multi-objective optimization model for identifying critical elements is proposed. • A novel importance metric of critical elements is designed. • The proposed model is verified by a practical high-speed train. • Critical nodes, intra-links and inter-links for a high-speed train are identified. This paper proposes a multi-objective optimization-based approach to identify critical elements, including units and interactions within and between systems, in a high-speed train (HST). In the framework, network theory is used to model the HST as an interdependent machine-electricity-communication network (IMECN) composed of a machine network (MN), an electricity network (EN) and a communication network (CN). Cascading failure models for the subnetworks and IMECN, and topological and functional metrics for robustness are developed. We then formulate a multi-objective optimization model for maximizing the impact of the failure of critical elements on the topological and functional robustness of the IMECN and minimizing their number. We use NSGA-II to solve the optimization problem. Considering a practical HST as a case study, we apply the multi-objective optimization framework to search the groups of critical nodes, intra-links and inter-links. The results show that critical nodes, intra-links and inter-links of the IMECN are within the MN and CN. In particular, end nodes of the critical intra-links and inter-links may also be critical, and the critical elements of subnetworks tend to also be critical for the IMECN. In addition, we find that the critical nodes, intra-links and inter-links are not related to their topological importance. [ABSTRACT FROM AUTHOR]

[Display omitted] • A multi-objective model is proposed to design guidance information system. • The model is designer-friendly due to simultaneous optimization of signage location and information allocation plans. • The model produces cost-effective GIS plans for underground spaces. • The model improves wayfinding efficiency by satisfying potential information demand in underground spaces. Guidance information systems (GISs) are widely used for pedestrian navigation in underground spaces. We investigate GIS design methods in this paper. First, the information demand and potential information locations are abstracted based on the network discretization of underground spaces. Second, a multiobjective GIS design (MOGISD) model that focuses on economic costs and wayfinding is developed, and a dynamic solution evaluation-based heuristic algorithm is proposed to determine the optimal GIS design plan. To obtain a GIS plan that can achieve a perfect match between demand and information, a three-dimensional microscopic pedestrian-information interaction model that considers both human and information factors is formulated. Finally, to evaluate the effectiveness of the MOGISD model, we use the model to design a GIS plan for the underground floor of a multimodal transportation hub. The results show that the proposed method produces cost-effective GIS design plans contributing to high-efficiency wayfinding. The MOGISD model is designer-friendly due to the simultaneous optimization of signage location and information allocation plans. [ABSTRACT FROM AUTHOR]

Rolling bearing as one of the key mechanical components plays a significant role in the field of industrial. To improve the ability of rolling bearing fault diagnosis under multi-rotating situation, this paper proposes a novel rolling bearing fault characteristic product function correntropy (PFC), and employs Least Square Support Vector Machine (LSSVM) to implement intelligent fault identification of rolling bearings under multi-stationary working situations. Firstly, rolling bearing vibration acceleration signal is decomposed by Local Mean Decomposition (LMD) to extract product functions (PF). Secondly, PFC needs to be obtained. PFC is the solution of the correntropy mathematical model of primary signal and PF component that is modified by Correlation Coefficient Entropy (CCE) as amplitude modulation operator. Finally, drawing support from LSSVM, the fault identification is achieved. Through the bearing identification experiments in different rotating situations, it is verified that PFC generates higher diagnosis accuracy than traditional fault features. Meanwhile, it is proved that PFC has more robustness than traditional fault features under cross-mixed roller bearing rolling status. Above all, the higher efficiency and availability of LMD-PFC-LSSVM are confirmed from the experiment consequence. It can be concluded that LMD-PFC-LSSVM is a reliable technology for rolling bearing fault diagnosis online under complicated rolling conditions and possesses the broad application prospect. [ABSTRACT FROM AUTHOR]

The smooth and fast evacuation of a pedestrian group is dependent on guidance services provided by signage systems. This paper investigates the location method of an evacuation signage system in a public space. We proposed a calculation method to determine the guidance efficiency of signage and further present a piece-wise probability function to explain interactions between pedestrians and signage. The interaction between one pedestrian and one signage was extended to the interaction between a pedestrian crowd and a signage system. A location model of the signage system was proposed to determine the minimum number of signs necessary to meet guidance demands. The location model was based on the Cooperative Location Set Cover Problem (CLSCP) and was correspondingly solved by a proposed exponential binary heuristic search algorithm (EBHS), i.e., a combined exponential binary search method and a heuristic search algorithm for solving the Cooperative Maximum Cover Location Problem (CMCLP). Finally, the proposed model was applied to determine the location of an evacuation signage system in a hall. The parameters used in the location model were calibrated based on experimental data. The model results showed that the proposed model can suggest the optimal number and best locations of signs. A sensitivity analysis showed that the guidance capacity of the signage system can be increased by improving the attractiveness of signage and pedestrian trust and familiarity with environment. The same number of signs is suggested for evacuation scenarios wherein crowd following behavior is present. [ABSTRACT FROM AUTHOR]

• The M-LSCSFN for an HSEMU system describes system structure and functions. • The layer reliability is calculated based on MPs using specific operators. • The M-LSCSFN reliability is estimated with the layer reliability. • The HSEMU system reliability is evaluated in terms of the M-LSCSFN reliability. • A practical case is shown to demonstrate the proposed method. This paper proposes a new reliability assessment method for a high-speed electric multiple unit (HSEMU) system based on the expanded multi-state flow network. An HSEMU system is abstracted as a multi-layer multi-source multi-commodity multi-state flow network (M-LSCSFN), where minimum maintenance units (MMUs) and three types of connections form nodes and edges in different layers, and functional characteristics are regarded as network parameters. In the M-LSCSFN, multi layers describe different types of connections and their relationships; multi-source and multi-commodity reflect system features; whereas multi-state flow indicates the dynamic realization of system functions. Considering that an HSEMU system can be divided into multiple independent holistic functions, each holistic function has a corresponding sub-M-LSCSFN, and each layer of this sub-M-LSCSFN is a multi-source multi-commodity multi-state flow network (M-SCSFN). In order to assess system reliability, layer reliability (i.e. , the M-SCSFN reliability) is first defined as the probability that the amounts of each commodity transmitted successfully from source nodes to sink nodes will not be less than the required demands. A corresponding algorithm is proposed to calculate layer reliability in terms of minimal paths (MPs) using the quick inclusion-exclusion technique (QIE). The sub-M-LSCSFN reliability is subsequently calculated based on the layer reliability and the conditional relationships among different layers. Based on these, the HSEMU system reliability is defined as the ability of a system to complete specified multiple functions under stated conditions for a specified period of time, and can be evaluated using the reliability of sub-M-LSCSFNs via the proposed operator. Finally, a case study of the "Fuxing" HSEMU system is analyzed to demonstrate the applicability of the proposed method and algorithm the effectiveness of the proposed method is demonstrated in an HSEMU system reliability assessment case study. [ABSTRACT FROM AUTHOR]

• The high-speed railway network is modeled as a three-layer network. • Network-specific characteristics are considered in performance evaluation instead of pure complex network analysis. • Railway stations are prioritized and compared within different layers of the network. • The current physical rail network in China is generally sufficient to satisfy the requirements of its transportation accessibility. • Management implications are provided to facilitate the rail network planning, transportation organization and operations. High-speed railways, as dominant transportation patterns, constitute a large-scale complex network. Network performance is critical for high-level transportation services in transportation management. However, existing studies generally adopt pure topological indicators to evaluate performance in a single network and without considering the specific characteristics of rail transportation. This paper modeled the high-speed railway as a three-layer network including topological, functional, and service layers and assessed the integrated network performance from the view of transportation accessibility. In addition, the rail stations were prioritized and their performances in different layers were compared to analyze whether the total integrated performance was restricted by its connection abilities, functional capacities, and transportation service level. Finally, a case study of the high-speed rail network in China and comparisons with other traditional evaluation methods were conducted to verify the feasibility and superiority of the model. The findings of this study enable to provide management implications for rail transportation planning, organization, and operations. [ABSTRACT FROM AUTHOR]

With excessive passenger demand and uneven distribution of passenger flow, long-distance subway lines face oversaturated conditions during peak hours, leading to potential safety problems for subway systems. To improve travel efficiency and reduce safety problems, this study proposes a method of constructing a passenger flow control jointly optimized train timetable for an oversaturated subway line under a short-turning operation pattern. Considering the time-dependent and dynamic characteristics of passenger demand, an integer nonlinear programming model is formulated to minimize the total passenger waiting time. To solve the problem, the tabu search algorithm and the IBM ILOG CPLEX solver are combined to seek the approximate optimal solution. Finally, a series of numerical experiments are conducted on the Beijing Fangshan line to demonstrate the effectiveness of the proposed model and approach, which shows that the method of jointly optimizing the train timetable and passenger flow control can drastically decrease the total passenger waiting time and reduce congestion. • A congested subway line under a short-turning operation pattern is considered. • Two types of passenger flow control and boarding strategies are investigated. • Passenger flow control jointly optimized train timetable model is formulated. • Tabu search algorithm with CPLEX solver is designed to search the solutions. • Real-world case study to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: This paper improved the Jøsang's subjective logic through the establishment of rectangular coordinate system, study on subjective logic properties in the two-dimensional image and using fitting line algorithm to depict dynamic trajectory of the opinion. At the same time, the Consensus operator was improved and made it more realistic. On this basis, this paper proposed a network survivability assessment model. It established a set of indicator system that could reflect the survivability and the formal description was given. Using the structures of the network environment for testing, the simulation results show that the indicator system is normative and complete. This model has higher sensitivity and effectiveness, which can better adapt to dynamic changes in the assessment of environment. [Copyright &y& Elsevier]

Abstract: Distributed logic programming languages, which allow both facts and programs to be distributed among different nodes in a network, have been recently proposed and used to declaratively program a wide-range of distributed systems, such as network protocols and multi-agent systems. However, the distributed nature of the underlying systems poses serious challenges to developing efficient and correct algorithms for evaluating these programs. This paper proposes an efficient asynchronous algorithm to compute incrementally the changes to the states in response to insertions and deletions of base facts. Our algorithm is formally proven to be correct in the presence of message reordering in the system. To our knowledge, this is the first formal proof of correctness for such an algorithm. [Copyright &y& Elsevier]

Accurately diagnosing the faults of wheelset bearings in high-speed trains is critical for ensuring the safety of train operation. In recent years, deep learning methods have made significant progress in this area. However, collecting enough labeled data in high-speed operation would be highly impracticable. In addition, the harsh operating environment of wheelset bearings, coupled with fluctuations in loads due to passenger flow, weather conditions, and other factors, makes it challenging to identify the fault signatures submerged in background noise. To address these issues, this paper proposes a semi-supervised fault diagnosis method based on autocorrelation and an improved flow Gaussian mixture model. First, the autocorrelations of raw bearing vibration signals are calculated to reduce the impact of noise while highlighting the discriminative fault signatures. Second, an improved flow Gaussian mixture model is established, preserving its simplicity and interpretability, to handle the constraints of bearing fault diagnosis with limited labeled samples under variable loads. Finally, the proposed method is tested by using a bearing dataset collected from an industrial railway wheelset bearing test rig. The results demonstrate that the proposed method outperforms the five state-of-the-art methods in terms of diagnostic accuracy and robustness. • A semi-supervised method is proposed for high-speed train wheelset bearing fault diagnosis with limited labeled samples under variable loads. • Autocorrelation is used to pre-process the raw vibration signals while providing an input to the model. • An improved flow Gaussian mixture model with a simple structure and interpretability is used to diagnose the bearing faults under variable loads. • The effectiveness of the method is verified on a railway wheelset bearing dataset. [ABSTRACT FROM AUTHOR]

• A new method based on time-frequency domain combined with phase space reconstruction and singular value decomposition is proposed. • The proposed method integrated with wavelet transform, phase space reconstruction and singular value decomposition can lead to fast and efficient results. • Validity of the proposed method proposed is verified by pratical fault diagnosis application of wind turbine. Operation of wind turbines under fault state will directly affect the power output efficiency of wind farms. This paper proposes a new automatic fault diagnosis method for wind turbines. A fault diagnosis system framework is constructed and data of vibration status of wind turbines collected is processed and used for fault diagnosis. Firstly, wavelet coefficients are obtained using a discrete wavelet transform (DWT) for vibration acceleration signals collected from wind turbines. Then, the wavelet coefficients are sequentially subjected to phase space reconstruction (PSR) and singular value decomposition (SVD) to extract the fault features. Finally, an extreme learning machine (ELM) is used to classify the faults. Experimental results show that the proposed method is more effective and accurate than other fault diagnosis methods for wind turbines, such as support vector machine (SVM) and multiscale convolutional neural network (MSCNN). Image, graphical abstract [ABSTRACT FROM AUTHOR]

In order to overcome the limitation of existing weighting methods and mitigate the effect of cascading failures, the harmonic closeness is adopted to define the node weight whose strength is controlled by a weight parameter θ , so that the initial load can be obtained by the node weight. We find that regardless of the average degree < k > in Barabási–Albert (BA networks), Newman–Watts (NW networks), and Erdos–Renyi networks (ER networks), the critical threshold T c achieves the minimum value under optimal θ. In these artificial networks except for NW and ER networks with big tolerance parameter T , the bigger the value of θ , the smaller the value of the normalized avalanche size C F N . Through the comparison of different methods, a key finding is that the value of T c obtained by the method proposed in this paper is significantly smaller than the ones by methods concerning the degree and the betweenness in artificial and real networks. In the range of big T , our method results in the smallest value of C F N in the networks mentioned above compared with previous methods. These results may be helpful for optimizing the distribution of the initial loads in real-life systems, and extending the research on cascading failures in the light of the harmonic closeness. • The harmonic closeness is first adopted to define the node weight. • The critical threshold is minimized under the optimal weight parameter. • The critical threshold with our method is smaller than those with existing methods. • The impact of failed nodes is mitigated by our method under high capacities. [ABSTRACT FROM AUTHOR]

• The main accident causes at the macro level are not applicable to specific scenes. • The distribution of accident causes concentrates with the refinement of the scene. • A data-driven Bayesian network construction approach is proposed. • The significant causes in various specific scenes can be accurately identified. As the freight railway system is a typical complex system, freight railway accidents have various and complex accident scenes. A Data-Driven Bayesian Network (DDBN) with random variables representing scene elements, accident causes, and accident consequences as nodes, and conditional dependencies between nodes as edges is proposed to identify the most significant accident causes in various and complex specific accident scenes. First, an unsupervised-supervised method is designed to define the states of nodes in the DDBN, considering the characteristics of the data involving both continuous and discrete states. Second, a greedy algorithm is proposed to mine the causal sequence between nodes, and the direction of edges in DDBN is established accordingly. Then, an NB-K2-MLE approach is proposed to generate the structure and parameters of the DDBN from data. Finally, a risk calculation function based on DDBN is proposed to calculate the risk of various accident causes in specific scenes. In empirical analysis based on real accident data, the inference accuracy of DDBN reached 87.92 %, with precision and recall exceeding 70 %. More importantly, the research results indicate that the distribution of accident causes will concentrate with the refinement and specific of the scene, and the main accident causes at the macro level cannot be fully applicable to accident prevention in specific scenes. The DDBN constructed in this study can provide data support for the determination of the significant accident causes and the development of targeted accident prevention strategies in various and complex specific scenes. [ABSTRACT FROM AUTHOR]

• A semi-automatic model is developed for hazard and accident cause identification. • We explore and analyze the semantic relationship between the hazards of marine accidents. • The importance of both hazards and accident causes is calculated and ranked. • We gain accident causes distribution in 40 different accident scenarios by using topic model. In the past decades, marine accidents brought the serious loss of life and property and environmental contamination. With the accumulation of marine accident data, especially accident investigation reports, compared with subjective reasoning based on expert experience, data-driven methods for analysis and accident prevention are more comprehensive and objective. This paper aims to develop a content-aware corpus-based model for the analysis of marine accidents to mine the accident semantic features. The general research framework is established to combine accident data, expert prior knowledge, and semi-automated natural language processing (NLP) technology. The NLP models are optimized, fused, and applied to the case study of ship collision accidents. The results show that the proposed model can accurately and quickly extract hazards, accident causes, and scenarios from the accident reports, and perform semantic analysis for the latent relationships between them to extend the accident causation theory. This study can provide a powerful and innovative analysis tool for marine accidents for maritime traffic safety management departments and relevant research institutions. [ABSTRACT FROM AUTHOR]

More and more passengers prefer to travel by subway system, and congestion is inevitable in the routine operation of subway stations. To alleviate congestion, junctions with chamfer zone (JCZs) are built in subway stations. The main intention of this work is to investigate the improvement of walking efficiency by the chamfer zone. Firstly, the specific walking behaviors of pedestrians in JCZs are investigated, and the principle of detour behavior is proposed. Secondly, a walking model considering the detour behavior is developed for pedestrians traversing the JCZ, and model parameters are calibrated by the Response Surface Methodology (RSM). Then, the influences of chamfer geometry on walking speed, walking time and time headways are simulated and analyzed. Results indicate that the walking efficiency will be improved by about 7.1% when the chamfer length is 1 meter. The interference among pedestrians is lower, and most time headways fall within the range of 0.1 ∼ 0.3 s, which means pedestrians walk steadily in the JCZ. Finally, the walking parameters of pedestrians in JCZs and corners are compared in various scenarios, and the matching correlation between chamfer length and turning angle is obtained. [ABSTRACT FROM AUTHOR]

Condition monitoring of railway tracks is essential in ensuring the safety of railway systems. Traditional track recording vehicles (TRV) can only carry out the track inspection in the midnight after all the service vehicles are off duty. One track line can be inspected only once or less in a month. The severe track condition degradation cannot be detected timely and there is potential safety hazard for the railway system. Therefore, the track condition monitoring based on in-service vehicle has been paid more attention in the recent years. In this paper, the track condition monitoring based on the bogie and car body acceleration measurements is presented. The track alignment information which can be derived from the bogie and car body sensor is investigated by using a mathematical model and the frequency response analysis. The track irregularities can be estimated by using the bogie and the car body acceleration sensors. The acceleration signals are preprocessing with a DC filter and a low pass filter. After that, the track alignments are obtained by double integrating the signals. Field tests are carried out in Shang Hai metro Line 1 for demonstrating the effectiveness of the proposed track inspection system. [ABSTRACT FROM AUTHOR]

At present, the interest in non-Pd hydrogen permeation membranes mainly focuses on group 5 B metals (Nb, V, Ta) and their alloys. To solve the problem of easy oxidation and weak dissociation of hydrogen molecules, a "Pd catalytic layer" needs to be deposited on the membrane surface. Due to its limited stability towards sulfur, issues of irreversible poisoning by sulfur compounds must be addressed. To this end, Pd-, Pd 65 Ag 35 -, Pd 90 Cu 10 -, Pd 80 Au 20 -, and Pd 55 Ag 35 Au 10 -coated Nb 30 Hf 35 Co 35 membranes were firstly synthesised by magnetron sputtering, and their hydrogen permeation behaviour in pure H 2 and/or in different H 2 /H 2 S mixtures for up to 1000 ppm at 673 K were evaluated. At these conditions, the permeabilities of the Pd-coated membranes decrease rapidly with increasing the H 2 S concentration and cannot recover after removal of H 2 S from the feed mixtures. In comparison, the Pd–Ag-containing membrane greatly increases the pure H 2 permeability, whereas the membrane doped with Au can effectively improve the response to H 2 S. Specifically, the Pd 65 Ag 35 -coated membrane shows the highest pure H 2 permeability, but it loses maximum permeability value on exposure to H 2 /H 2 S feed mixtures. In contrast, Pd 80 Au 20 -containing membranes exhibit good tolerance of H 2 S poisoning and displayed a high fractional recovery (∼83%) of initial H 2 permeability. The degradation of permeability for Pd- or Pd 65 Ag 35 -coated membranes after the H 2 S exposure can be ascribed to the deterioration of the structure caused by the possible formation of bulk Pd 4 S, which has been corroborated with EDS analysis, XRD, and XPS after testing. Finally, to improve both permeability and chemical tolerance, a Pd 55 Ag 35 Au 10 -coated membrane was successfully fabricated, and it possesses excellent comprehensive performance with respect to the permeability and H 2 S response. At 120 ppm H 2 S in the feed, the H 2 flux equals ca. 43% of its original value at 673 K and can be stable for at least 166 h. Bulk sulfides did not form on the membrane surface after the testing. This paper not only developed a new type of group 5 B alloy membrane with improved resistance to corrosion by H 2 S, but also further confirmed the feasibility of improving H 2 S corrosion resistance by adjusting the catalytic layer type or composition of the outermost coating. Some novel Nb 30 Hf 35 Co 35 self-supported membranes were successfully developed by regulating the type or composition of the outermost catalytic film. Correspondingly, the compositional region suitable for hydrogen permeation with improved resistance to corrosion by H 2 S has been identified. Especially, Pd 55 Ag 35 Au 10 -coated membrane possesses excellent comprehensive performance with respect to the permeability and H 2 S response. At 120 ppm H 2 S in the feed, the H 2 flux equals ca. 43% of its original value at 673 K and can be stable for at least 166 h. Bulk sulfides did not form on the membrane surface after the testing. The first-principles calculations showed that doping alloying elements, especially gold, inhibit the formation of palladium sulphide (e.g., Pd 4 S), thus providing more activating active sites for hydrogen adsorption and dissociation. This could be the main reason for the improvement of hydrogen permeability of PdAuAg-coated composite membranes in an H 2 S atmosphere. [Display omitted] • A new way to design hydrogen permeable alloys with improved H 2 S corrosion resistance is proposed. • A novel Pd 55 Ag 35 Au 10 -coated Nb 30 Hf 35 Co 35 self-supported membrane were successfully developed. • The PdAgAu catalytic layer showed promising resistance to sulfur poisoning. • Adjusting the type or composition of the catalytic layer can alleviate H 2 S poisoning. [ABSTRACT FROM AUTHOR]

Train service plans evaluation remains a longstanding challenge in railway operation. The train service plan evaluation problem based on complex network theory is investigated in this paper. The train service plan is transferred into a train service network, which is a complex network with the typical characteristics of small-world and degree free. Three key indices of a complex network are listed and we relate them with the train service plan. Based on the three indices, we define the convenience degree for traveling, the transferring times, the travel time, and station clustering coefficients to evaluate the train service plan. A computing case is performed to illustrate the applications of the evaluation model and the algorithm. The results show that the evaluating results can be attained via the method proposed effectively, which can offer supporting information for decision makers of the railway operation bureaus. It proposes a novel, practical and valuable tool to evaluate the train service plan. [ABSTRACT FROM AUTHOR]

• The underground space is partitioned into cells with different size. • Guidance demand is treated as heterogeneous over the underground space. • An accurate interaction model for evacuees and signs is proposed. • A location set covering model for EGSS design is proposed. • We combine advantages of analytical and simulation methods to design EGSS. Emergency guidance signage systems (EGSSs) are widely used to guide crowd evacuation in underground spaces. However, the absence of signs often reduces the evacuation efficiency. Current methods for EGSS design suffer from challenges associated with complex evacuee-signage interactions, guidance demand determination and numerous potential sign locations. The present article investigates a discrete approach to EGSS design. To this end, a cell-based discretization method is proposed to abstract the guidance demand and potential sign locations. To design an EGSS that can efficiently convey guidance information to evacuees, an interaction model, which is determined by the cognitive capabilities of moving evacuees and the characteristics of signs, is proposed for analyzing the guidance relationship between evacuees and signs with different locations and orientations. An EGSS design model is formulated to recommend the optimal number, location, and installation orientation of signs, and a wayfinding simulation model is proposed to validate the obtained design plan. Finally, a real-world case study is used to demonstrate the effectiveness and advantages of the proposed methods. [ABSTRACT FROM AUTHOR]

Split pins (SPs), as essential fasteners in the pantograph-catenary system, play a critical role in maintaining the structural stability of the catenary support device (CSD) in high-speed railroad. Unfortunately, impact and vibration generated by the interaction between catenary system and high-speed running vehicles will progressively damage SPs, including but not limited to the loose, breakage, and missing of SPs. Current image processing technologies have limited accuracy. Traditional convolutional neural network (CNN)-based systems that directly learn deep features are inefficient in detecting subtle defects (such as fractured and loose SPs). This study presents a novel SP inspection system based on an innovative destruction and reconstruction learning (DRL). In the DRL, a novel cross stage partial network (CSPNet)-based classifier is firstly developed to alleviate the impact of redundant semantic information and optimize gradient regression process. Then, "destruction" in the DRL stream means these partitioned sub-regions in an input image are shuffled by a region shuffling mechanism (RSM); this guides the classifier to pay more attention to the distinguishable regions in the destroyed and original images for a purpose of correct recognition. For "reconstruction", a newly developed position reconstruction network (PRN) is utilized to restore the original spatial distribution of the sub-regions by modeling the semantic correlation among the sub-regions. The proposed DRL can injects more discriminative details to the classifier by jointly training the classification network and PRN. Finally, comprehensive experiments are conducted on CSD datasets captured from real railroad lines, and the results demonstrate that the proposed approach outperforms existing state-of-the-art models yet keeps a reasonable inference speed. [ABSTRACT FROM AUTHOR]

• Accident semantic network framework and a conceptual model of accident risk control chain can provide great help for the study of accidents. • A method for acquiring and constructing key contents based on semantic analysis is elaborated. • A management matrix is established to serve safety decision. • It provides suggestions for the actual operation management of urban rail transit. Although the occurrence of urban rail transit operation accidents threatens the safety of people's lives and property, there is little research on how to systematically mine the knowledge behind accident data in China, and there is also lack of methods to support the safety management strategy of urban rail transit operation. To address these shortcomings, this study proposes a comprehensive analysis method that combines qualitative and quantitative analysis of urban rail operation accident factors and safety management strategies based on accident data, which can provide reference for systematic analysis of such accidents in the future and will help urban rail safety researchers and decision makers to take effective safety management measures. The new contributions of the method are as follows: (1) a semantic network framework for accidents and a conceptual model of accident risk control chain are proposed; (2) a method for acquiring and constructing key contents based on semantic analysis is elaborated, and indicators for evaluating the influence degree of key item and the influence possibility of entries are proposed; (3) a management matrix is established to serve safety decision making based on indicators for the degree of attention that key item of accidents should receive. In order to explain the application of the comprehensive analysis method, 910 urban rail transit operation safety accident cases are analyzed in detail, and suggestions for the actual operation management of urban rail transit are proposed in the light of the actual situation of urban rail transit operation in China. [ABSTRACT FROM AUTHOR]

Greening of the railway energy supply chain is an irreversible trend, and photovoltaics (PVs) provide the most suitable type of renewable energy to integrate with railways. The integration of variable and uncertain PV power generation with the dynamic loads on a railway increases the flexibility needed to maintain load-generation balance. A new evolutionary model of a railway energy supply system (RESS) for railway PV integration systems (RPISs) is proposed by constructing a three-in-one "traction-storage-information integrated station" (TSIIS). A flexibility evaluation approach is proposed to estimate the system flexibility of both the individual TSIIS and the banded featured microgrid group through energy storage capacity configuration and dynamic scheduling via a multileader multifollower Stackelberg game. Finally, the Lanzhou-Xinjiang Railway is taken as an example to calculate the emission reduction and flexibility. The results show that through application of the storage configuration and scheduling strategy to the TSIIS network, the flexibility index of the Lanzhou-Xinjiang Railway increased by 8.40%, which promoted a corresponding decrease in carbon dioxide emissions and increase in PV permeability. In addition, with the gradual construction of PV and energy storage systems, an RPIS can basically achieve self-consistent energy utilization, offering environmental benefits and strong flexibility. [ABSTRACT FROM AUTHOR]

Corridor with corner may be the potential bottleneck in subway station, and walking behavior of passengers around corner is random. In this study, walking process of passenger around corner is divided into five phases according to field survey, i.e., pre-turning straight-going, ready-turning, turning, distance-adjusting and post-turning straight-going. An advanced social force model is proposed to model the turning behavior of passenger around corner, taking visual angle, walking preference and steering force into consideration. To improve walking efficiency, the influences of geometry of facility, walking preference of passengers and arrival rate of passenger flow are investigated. Results show when angle of corner is less than 120°, the walking speed slows down sharply and walking time extends. To improve the walking efficiency, passengers can be guided to walk in lines orderly. The maximum of throughput can be obtained with proper arrival rate. Once the volume of passengers exceeds the facility capacity, effects of "stop-and-go" and "faster is slower" will occur around corner. • Analyzing and demonstrating the walking process of passengers around corner. • Modeling movement of turning passengers with advanced social force model. • Verifying the proposed model with controlled experiments. • Quantifying arrival rate and corner angle influencing on walking efficiency. [ABSTRACT FROM AUTHOR]

To segment railway track regions in real-time for intrusion detection and improving security, this paper proposes an efficient railway track region segmentation network (ERTNet) based on the encoder–decoder architecture. Firstly, to ensure the lightweight of the encoder, depthwise convolution and the channel shuffle are utilized to construct sandglass-type feature extraction unit. Secondly, a feature-matching-based cross-fusion decoder is utilized to fuse deep and shallow feature maps. Thirdly, the knowledge distillation is employed with large-scale Deeplab v3+ as the teacher model to improve performance. Additionally, a loss function is proposed to penalize pixel points with large offsets. Finally, the ERTNet is validated on the self-built dataset, achieving an MIoU (Mean Intersection over Union) of 92.4% , which is 5.22% improvement over the benchmark model. ERTNet achieves a balance between segmentation accuracy and computational efficiency, requiring only 0.5 M parameters and 0.92 G FLOPs (Floating Point Operations). [Display omitted] • Visual perception is utilized to identify foreign object intrusion into railroad. • Deep learning based ERTNet model is proposed for track region pixel extraction. • ERTNet is built with lightweight neural network and cross-fusion decoder. • Optimizing training modes with knowledge distillation and new loss function. • ERTNet fits real-time needs and helps to improve the safety of rail construction. [ABSTRACT FROM AUTHOR]

• A novel combined method for bogie diagnosis is proposed. • A novel loss function called Imbalance-weighted cross-entropy is proposed to deal with data imbalance. • The proposed method is robust to noise and different operating conditions and imbalanced data. The condition monitoring of bogies plays an important role in the safe operation of rail transit systems. In practice, it is difficult to diagnose bogies under variable operating conditions for traditional methods. In this paper, deep residual learning is used to deal with this problem by utilizing its powerful feature self-learning characteristics and fast kurtogram is used to further enhance the robustness of the method to variable operating conditions and noises. Besides, considering the performance of data-driven deep learning method is easily to be affected by data imbalance which often occurs in practical data collection, a novel loss function called imbalance-weighted cross-entropy (IWCE) is proposed in this paper for networks training to solve the data imbalance problem. Finally, the robustness to variable operating conditions, noises and data imbalance of the proposed method is verified by two case studies. [ABSTRACT FROM AUTHOR]

During the merging process in a Y-shaped passage, the walking direction of pedestrians changes, which can cause crowding and decrease the efficiency of walking facilities. Based on an analysis of interactions among pedestrians, this study conducts a series of merging simulations with an improved social force model (ISFM) to investigate the influences of various walking parameters of pedestrians in a Y-shaped passage. First, we construct an ISFM by adding a centripetal force to analyze pedestrian turning behaviors based on the basic social force model (BSFM). Second, we define simulation scenes with different angles (30°, 60°and 90°) and branch passage widths (2 m, 3 m and 4 m). Walking behavior parameters such as the walking speed, traversal time and population density in the branch passages are analyzed. The results show that as the width of the branch passage increases, the traversal time decreases and throughput increases. The turning angle has little influence on the results, but as the angle decreases, pedestrian speeds sharply fluctuate in the merging area, thereby increasing the blockage probability and prolonging the traversal time. Finally, relationships between the branch passage width and traversal time and the merging angle and speed fluctuations are proposed. • Addressing the merging process with an improved social force model. • Establishing simulation scenes with different angles and branch passage widths. • Exploring the relationships among the pedestrian walking behavior parameters. • The main factor that affects the traversal time is the width of the branch passage. • The main factor affecting the pedestrian speed is the merging angle. [ABSTRACT FROM AUTHOR]

• A new modeling method for safe region is proposed to handle cases where only normal data are available. • A novel optimization method determines the kernel parameter for the safe region model. • A new indicator derived from the safe region model tracks degradation evolution throughout the whole life cycle. • The proposed methods are applied to anomaly detection and degradation evaluation of wheelset bearings. This paper presents a modeling method, an optimization method, and two applications for the safe region concept that has been developed for the health assessment of wheelset bearings of high-speed trains. The proposed safe region model uses support vector data description to handle cases in high-speed trains where only normal data are available. The optimization process is designed with a strong physical interpretation of the relationship between the kernel parameter and the model performance. A new indicator derived from the safe region model is proposed for degradation evaluation throughout the whole life cycle. Experimental data on wheelset bearings with naturally generated faults are used to demonstrate that the proposed approach has competitive advantages in several performance aspects for anomaly detection and provides monotonic labels in assessment of increasing degradation levels. [ABSTRACT FROM AUTHOR]

Abstract The railway track fasteners play a critical role in fixing the track on the ballast bed. Achieving full automation of the fastener defect detection is significant in terms of ensuring track safety, and reducing maintains cost. In this paper, innovative and intelligent methods using image processing technologies and deep learning networks are proposed. In the first part, the traditional fastener positioning method based on image processing is reconsidered. In addition, a novel fastener defect detection and identification method using Dense-SIFT features is proposed which can achieve a better performance than the methods available in the literature. In the second part, VGG16 is trained for fastener defect detection and recognition. The result demonstrates that it is possible to carry out the defect detection of fasteners with CNN. Finally, Faster R-CNN is used for fastener defect detection to advance detection rate and efficiency. The fastener positioning and recognition can be carried out simultaneously. The time for the defect detection and classification is only one-tenth of the other methods mentioned above. Highlights • The Dense-SIFT, spatial pyramid decomposition and BOVW techniques are applied to the classification and this method achieves classification accuracy of 99.26%. • The DCNN method uses only one network for fastener defects detection and achieves a sound classification accuracy of 97.14%. • Faster R-CNN based method realizes the positioning and classification of the fasteners simultaneously and the time spent by Faster R-CNN is only about 10% of the time spent by the other methods. [ABSTRACT FROM AUTHOR]

Novel and flexible Mn 3 O 4 nanosheets/reduced graphene oxide nanosheets (Mn 3 O 4 NSs/rGO NSs) paper-like electrodes have been developed by a feasible method. The rGO works as a flexible substrate and an electron conductor. The synergetic effect of interconnected nanosheet structure can effectively depress self-agglomeration of rGO, which displays eminent conductivity. Moreover, the crumpled surface with more active sites are more convenient for the ion and electron insertion/desertion in the reversible reactions. So flexible Mn 3 O 4 NSs/rGO NSs electrode exhibited a high specific capacitance of 409 F g −1 at a current density of 0.5 A g −1 , superior cycling stability of 92% after 3000 charge-discharge cycles. More importantly, the composite material was first applied in 3D printing and screen printing of micro-patterns because of the desirable processability. It is anticipated that the high-performance Mn 3 O 4 NSs/rGO NSs composites are a promising electrode material in printable energy storage and wearable electronic equipment. [ABSTRACT FROM AUTHOR]

With the widespread implementation of smart cards and with more travel details being recorded, travel patterns can be studied more precisely and deeply. Although considerable attention has been paid to travel patterns, the relationship between travel patterns and the functional structure of a city is yet to be well understood. In this research, we study this relationship by analysing metro passenger trip data (in Beijing and Shenzhen in China and London in the United Kingdom), and we present two unprecedented findings. First, through averaging, a linear relationship is found to exist between individual travel distance and the distance between the origin and the city centre. The underlying mechanism is a travel pattern we call “ring aggregation”, i.e., the daily movement of city passengers is aggregated into a ring (with approximately equal distances to the city centre). Then, for commuting trips, the daily travel pattern can be regarded as switching between the outer residential ring and the inner work ring. Second, this linear relationship and the ring aggregation pattern seem to be exclusive characteristics of metro systems (and may also fit other moderate- and long-distance transportation modes) but do not apply to short-distance transportation modes, such as bicycles and taxis. This finding implies that the ring aggregation pattern is a product of the relationship between travel patterns and the city functional structure at a large scope. [ABSTRACT FROM AUTHOR]

• A new GDM method is developed based on incomplete PFPRs. • The sufficient conditions for the missing elements to be expressible are presented. • The missing elements in unrepresentable incomplete PFPRs are estimated. • The weights of decision-makers are determined based on the consensus degree. Completing the Pythagorean fuzzy preference relations (PFPRs) based on additive consistency may exceed the defined domain. Therefore, we develop a group decision-making (GDM) method with incomplete PFPRs. Firstly, sufficient conditions for the expressibility of estimated preference values in PFPRs based on additive consistency are presented. Next, the correction algorithm is developed to correct the inexpressible elements in incomplete PFPRs. Then, a GDM method based on incomplete PFPRs is proposed to determine the objective weights of decision-makers. Finally, an example of subway station safety management during COVID-19 is selected to illustrate the applicability of the developed GDM method. The results show that the developed GDM method effectively identifies the crucial risk factor in subway station safety management and has better performance in terms of computational time complexity than the multiplicative consistency method. [ABSTRACT FROM AUTHOR]

The condition monitoring of railway track line is one of the fundamental tasks to ensure the safety of the railway transportation system. Railway track line is mainly made up of tracks, fasteners, bolts, backing plates, and so on. Given the requirements for rapid and accurate inspection, an innovative and intelligent method for multi-component identification and common defect detection of railway track line is investigated based on instance segmentation. More specifically, a railway track line image (RTL-I) dataset is constructed and annotated manually in this paper. After that, based on the work of YOLACT and YOLACT++, combined with prior knowledge, a railway track line image segmentation model (RTLSeg for short) is proposed. Firstly, taking the characteristics of the objects in the RTL-I dataset, preset anchors are redesigned and a feature enhanced module is introduced in the prediction head to improve the detection and segmentation accuracy of the model. Secondly, to strengthen the internal information propagation within the model, PaFPN (path aggregation feature pyramid network) is applied instead of FPN in RTLSeg. Thirdly, with the help of CoordConv, Coord-Protonet is presented to add position awareness explicitly to the model for more robust and higher quality prototype masks. Finally, to further improve the model performance, the attention mechanism is explored and a novel spatial attention-guided bounding box branch is employed in the enhanced prediction head. Both quantitative and qualitative experimental results show that the proposed method is feasible in detecting and segmenting multi-component and common defects of railway track line, and outperforms the compared baseline models. In particular, RTLSeg is able to achieve 91.35 bbox mAP and 91.60 mask mAP with the customized dataset. Meanwhile, the average inference speed reaches 13.07 fps. The average detection accuracy and recall are 100% and 99.83%, respectively. Furthermore, the effectiveness of each optimized part of the proposed RTLSeg model is demonstrated by additional ablation study. • Pixel-level multi-component inspection of railway track line based on image instance segmentation is investigated for the first time. • Based on YOLACT and YOLACT++, a railway track line image segmentation model (RTLSeg for short) is proposed, which achieves better detection and segmentation performance than the compared baseline models, and is more suitable for the scenario of the railway track line. • The proposed RTLSeg model enables accurate and fast detection of railway track line components and common defects at the pixel level. [ABSTRACT FROM AUTHOR]

Considering the complexity of the urban rail transit system (URTS), it is necessary to systematically identify the critical factors and their relationships to avoid or prevent urban rail transit accidents. In response to the problems few previous studies have modeled networks for the whole system components of URTS and have focused too much on the identification of specific component states or behaviors without quantitative analysis of inter-component relationships. This paper proposed a risk network construction and analysis method based on complex network theory to identify the critical accident causative factors and their relationships of the URTS system. Firstly, the accident is defined as the combination of a sequence of risk incidents occurring in a certain order and their resulting consequences, and the definitions of risk point and risk incident were given. Secondly, a risk network construction and analysis method based on the combination of accident reports and complex network theory was proposed to obtain the critical factors and their relationships in the process of the accident. At last, this paper constructed the urban rail transit risk point set containing 39 risk points, built the risk network based on 201 accident reports of URTS in China and analyzed the critical risk points and risk relationships. The results show that this study can provide a new perspective for identifying the critical causative factors of URTS accidents and their relationships for practical application in risk analysis and accident prevention. • The new definitions of risk points and incidents are put forward. • A critical risk analysis framework are proposed by complex network theory. • Critical risk points and relationships are identified in the urban rail transit system by the risk network. [ABSTRACT FROM AUTHOR]

• The improved brute-force search transforms the energy-saving operation problem into optimal turning point solution. • Considering the speed limit, a more reasonable energy-saving turning point is obtained by adding a constant speed section. • The energy-saving effect of improved brute-force search is better than GA and PSO algorithms in Guangzhou Metro Line 7. Rail train operation energy consumption mainly focuses on train traction energy consumption. Reducing train traction energy consumption in rail transit operation is significant to developing a green and low-carbon economy and reducing operation costs. The rail train operation energy-saving optimization framework is developed considering the utilization of regenerative braking energy. Firstly, three objectives of punctual arrival, fixed-point parking and minimum energy consumption are provided by train operation strategy analysis. Secondly, the improved brute-force search is developed to solve train operation energy-saving multi-objective problems. The running time, speed, distance, power, and energy consumption of operation intervals are calculated. Finally, Guangzhou Metro Line 7 is taken as an example to verify the effectiveness of the developed optimization model. The results show that the improved brute-force search method effectively finds a more energy-saving turning point under constant interval operation time and has a better energy-saving effect than two other heuristic algorithms. [ABSTRACT FROM AUTHOR]

Potential safety hazards (PSHs) along the track needs to be inspected and evaluated regularly to ensure a safe environment for high-speed railroad operations. Other than track inspection, evaluating potential safety hazards in the nearby areas often requires inspectors to patrol along the track and visually identify potential threads to the train operation. The current visual inspection approach is very time-consuming and may raise safety concerns for the inspectors, especially in remote areas. Using the unmanned aerial vehicle (UAV) has great potential to complement the visual inspection by providing a better view from the top and ease the safety concerns in many cases. This study develops an automatic PSH detection framework named YOLARC (You Only Look at Railroad Coefficients) using UAV imagery for high-speed railroad monitoring. First, YOLARC is equipped with a new backbone having multiple available receptive fields to strengthen the multi-scale representation capability at a granular level and enrich the semantic information in the feature space. Then, the system integrates the abundant semantic features at different high-level layers by a light weighted feature pyramid network (FPN) with multi-scale pyramidal architecture and a Protonet with residual structure to precisely predict the track areas and PSHs. A hazard level evaluation (HLE) method, which calculates the distance between identified PSH and the track, is also developed and integrated for quantifying the hazard level. Experiments conducted on the UAV imagery of high-speed railroad dataset show the proposed system can quickly and effectively turn UAV images into useful information with a high detection rate and processing speed. [ABSTRACT FROM AUTHOR]

Turning junction of subway station is bottleneck being prone to congestion. Speeds of passengers fluctuate in these zones, which will diminish walking efficiency of the passengers and level of service (LOS) of facility. This paper explores the correlation between junction width and walking efficiency of passengers in subway station. Firstly, walking trajectories of passengers at junctions are abstracted from video footages with object detection and tracking algorithm. The parameters of walking behaviors of passengers in conditions with various junction widths are measured. With the calibrated turning social force model (TSFM), walking processes of passengers are simulated, and the optimal widths of junctions are estimated. Results show that the farther away from the inner turning boundary edge (TBE) is, the earlier the individual starts to turn. When junction width is more than 3 m, the walking efficiency will not be higher with junction widening during peak hours. The optimal width of junction is 5.41 m in the passenger-merging scenario with mean arrival rate of 1.67 ped ⋅ s−1. • Abstracting passengers' trajectories based on deep learning algorithm. • Turning trajectory is symmetrical to the turning center. • Exploring the correlation between junction width and walking efficiency. • Turning angular speed of passenger is calibrated to be 29.6 degree/s. • In merging flow scenario with 1.67 ped/s, the passage should be about 5.41 m wide. [ABSTRACT FROM AUTHOR]

• We provide a rail wheel tread defect detection framework based on improved YOLOv3 model to classify four tread defect types. • The improved YOLOv3 model transforms into the regression problem of the prediction frame and carries out multi-scale prediction for wheel tread defect detection. • Compared with the other detection methods, the improved YOLOv3 model maintains the detection speed, improves the detection accuracy and meets the real-time requirement. The rail wheel is the key device of the rail vehicle. When the vehicle running conditions worsen or emergency braking, the rail train will have severe sliding friction and collision on the track, resulting in scratches, peeling, and scratches on the wheel tread. The traditional wheel defect detection technology can not realize automatic classification and accurate defect positioning. Thus, an improved YOLOv3 framework for rail wheel surface defect detection is developed. The classification detection of four types of wheel tread defects is realized, and the average accuracy of mean average precision (mAP) is 0.92. Compared with the other detection methods, the experimental results show that the improved YOLOv3 model not only maintains the detection speed but also improves the detection accuracy and meets the real-time requirement. [ABSTRACT FROM AUTHOR]

Tramway has the advantages of reliable operation, energy-saving, and environmental protection, which is an important way to alleviate traffic pressure. Energy consumption optimization of tramway operation is significant in reducing the operation cost. Therefore, the energy consumption optimization model of tramway operation based on the improved PSO algorithm is developed. Firstly, the energy consumption model of the tramway is established based on the energy-saving operation strategy. Then, the competition mechanism-based particle swarm optimization (CM-PSO) algorithm is developed to improve energy-saving performance. Finally, the Guangzhou Haizhu tramway is taken as an example to illustrate the effectiveness of the developed method. The results show that the developed model effectively reduces the tramway operation energy consumption and performs better than the previous methods. • The energy-saving effect of the CM-PSO algorithm is better than other methods. • The operation strategy reduces energy consumption under operation time constraints. • The turning points between different work conditions are found. [ABSTRACT FROM AUTHOR]

Prediction of passenger flow distribution in urban rail transit stations can provide important data support for passenger flow organization and passenger travel guidance. However, complex station space structure and simulation-based passenger flow data bring challenges to accurate analysis and prediction of the passenger flow inside the station. This paper proposes a temporal graph attention convolutional neural network model (TGACN) to predict the passenger flow volume and density in key areas of the station. Firstly, considering the topological structure of key areas and the characteristics of passenger flow and flow trend in the station, a multi-graph generation method for continuous space in stations is designed, including geographic neighborhood graph and semantic neighborhood graph, to represent the static and dynamic correlation between nodes. Secondly, a new method of spatio-temporal feature fusion is proposed, which takes multi-graph as input to optimize the extraction and expression of spatial and temporal correlation. Finally, the TGACN is verified by passenger flow data set, which is constructed based on real-time video monitoring data of a transit station in Guangzhou. Experiments demonstrate that the TGACN can obtain the spatio-temporal correlation from passenger flow data, and the prediction results are better than the existing baseline models. • A prediction model for multipoint passenger flow in station is proposed. • Geographical graph is constructed to measure the spatial correlation between nodes. • Semantic graph is constructed to measure the correlation of passenger flow state. • Effective spatial information screening is realized by dimensionality reduction. [ABSTRACT FROM AUTHOR]

Long-term axle temperature prediction plays a significant role in train condition assessment and daily maintenance. However, most of methods make predictions for short-term conditions. In this paper, a method named GA-GRGAT is introduced that uses GAT and GAN to forecast the long-term axle temperature. In the proposed method, a GRGAT framework is used as a spatiotemporal fusion in temperature prediction. The GAN network with the GRGAT framework is used to construct a temporal conditional sequence after analyzing the periodic variation of the axle temperature, which can fuse the historical axle temperature information to improve the long-term prediction accuracy of the GA-GRGAT model. Our method in Python software and using the actual high-speed trains datasets in spring and summer. We using MAE, RMSE, MAPE, PCC to evaluate the accuracy of prediction. The accuracy of the GA-GRGAT is more than 90% on long-term predictions (1 day), and more than 80% on super long-term predictions (2 week). The GA-GRGAT method outperforms and is more accurate than the classical forecasting methods such as GRU, GOAMLP, DCNN, SVR and HA. In addition, the cost time of the proposed method is less than 5 min, which meets the requirements of high accuracy and long time. [ABSTRACT FROM AUTHOR]

With the rapid development of high-speed rail (HSR) systems, the security and safety of these huge systems are becoming the primary concerns for passengers. HSR infrastructure plays an important role in HSR systems, making the maintenance of security and safety of the HSR infrastructure especially important. Meanwhile, sensor network technologies allow the realization of real-time and all-weather monitoring of HSR infrastructure. This paper analyzes the application requirements and characteristics of infrastructure health monitoring sensor network (IHMSN) through construction of a three-layer IHMSN which is composed of end devices, repeater points, and access points. The physical topology optimization goal of IHMSN is to set the optimal number of network nodes (namely, minimum cost) as well as the best physical connections. Given types and amount of the end devices, a multiple knapsack model is established which converts the physical topology optimization problems into multiple knapsack problems. Based on the different needs of practical application, three different cases (basis case, adding devices case and weight-based case) are proposed, and the corresponding models are built. Some artificial intelligence algorithms and a traditional dynamic programming algorithm are presented to solve the problems. In addition, a general algorithmic finite state machine is proposed to describe the solving process. After comparing these algorithms in execution time, memory, and optimal results, the genetic algorithm and particle swarm optimization algorithm stand out when used to solve the basic case as well as the extended cases. The numerical results show that these proposed models and algorithms can effectively solve the physical topology optimization problem of IHMSN for HSR systems. Moreover, these methods can effectively reduce network costs and provide a theoretical basis for network communication link optimization. [ABSTRACT FROM AUTHOR]