21 results about "Train communication network" patented technology

The invention discloses a train communication network fault diagnosis method and device based on a multi-view hypergraph convolutional network, and the method comprises the following steps: firstly, collecting various high-order heterogeneous fault data of a train communication network under multiple working conditions, multiple time periods and multiple communication modes, then carrying out the analysis and preprocessing of the collected original data, and carrying out the fault diagnosis of the train communication network. The method comprises the steps of preprocessing data, dividing the preprocessed data into a training set and a test set according to a certain proportion, then fully mining potential relations among nodes from multiple dimensions through multi-view joint modeling, then building a train communication network fault diagnosis MVHGNN model, and finally performing training and parameter adjustment on the MVHGNN model through the training set. And sending the test set into the trained MVHGNN model to identify various network faults. According to the method, a complex heterogeneous high-order structure in the train communication network is analyzed through multiple views, the classification capability of the model on various types of space-time faults is improved, and intelligent diagnosis of the train communication network can be realized.

This invention provides a method for planning a wireless network within a high-speed train, belonging to the field of train communication. The method extracts key internal structural factors affecting radio wave propagation within the train by establishing a geometric model of the target scenario, and then constructs a segmented path loss model. It establishes a communication system model and a channel model based on the asymmetrical structural features inside the carriages, deriving mathematical expressions for wireless network coverage performance indicators. Combining the communication quality requirements within the train and relevant model formulas, it determines the maximum coverage distance of radio waves, determines the initial deployment locations of base stations based on the actual layout within the train, optimizes the size of cell overlap areas using wireless network coverage performance indicators, adjusts base station locations, and obtains the final deployment plan. This invention solves the problem that wireless communication technologies with random fluctuation characteristics cannot meet the high communication quality requirements of next-generation train communication networks.

The invention provides an automatic networking method for a train communication network and a storage medium. The method comprises the following steps: a management center reads a networking planning file and constructs a logical topology; based on a link layer discovery protocol, the management center interacts with managed equipment to construct a physical topology; if the physical topology is matched with the logical topology, the management center establishes network configuration protocol connection with the managed equipment; the management center distributes automatic networking configuration data to the managed equipment; and based on a publishing and subscribing mode, the management center performs state monitoring on the managed equipment.

The invention relates to a time synchronization system and method of a TSN train communication network. The system comprises a backbone network clock domain, a marshalling network clock domain and an intermediate clock domain formed by connecting backbone network nodes and marshalling network nodes, the two backbone network clock domains receive the same external clock source through each autonomous clock node to realize time synchronization, and provide a unified time reference for each marshalling network clock domain; under the condition that the time synchronization of the clock domain of the backbone network is abnormal or invalid, the backbone network node sends a synchronization abnormal signal to the marshalling network node of the intermediate clock domain where the backbone network node is located; the marshalling network clock domain is used for realizing time synchronization of equipment supporting the TSN in the marshalling network; under the condition that the synchronization abnormal signal is received, the marshalling network node of the middle clock domain does not enable the time synchronization function, so that the backbone network clock domain is isolated from the marshalling network clock domain where the marshalling network node is located. The problem that large-area synchronization abnormity of the whole vehicle may be caused once the time synchronization of the backbone network fails is solved.

The application provides a motor train city area braking software integrated test system, and relates to the motor train city area braking software test technical field.The system comprises: a plurality of electronic brake control unit cabinets to be tested, a plurality of sub-simulation nodes, a main simulation node and a test service node; the sub-simulation nodes are electrically connected with the corresponding electronic brake control unit cabinets respectively, are used for simulating the external input signals of the electronic brake control unit cabinets and collecting the output signals of the electronic brake control unit cabinets; the main simulation node is used for controlling the signal interaction between the sub-simulation nodes and the electronic brake control unit cabinets by managing the train communication network and simulating the train bus control instructions; the test service node is used for uniformly configuring, parameter issuing and scheduling controlling the main simulation node and the sub-simulation nodes, and analyzing the output signals to generate a test report.The application realizes software configuration instead of physical replacement, significantly shortens the test preparation time, improves the test stability and prolongs the hardware life.

The application provides a train communication network service bearing capacity calculation method and system, belongs to the train communication network traffic management technical field, establishes a TRDP data throughput model and other protocol data throughput model; based on the established TRDP data throughput model and other protocol data throughput model, a total throughput model of the train communication network is established, and the network service bearing capacity is calculated and evaluated. The application models the TRDP multi-service flow based on the train communication network, including the arrival time interval, frequency and throughput model of the TRDP process data, and the throughput model of other protocol small flow data, finally establishes a network total throughput model, the established total throughput model better simulates the network throughput condition under the real train network environment, and can reserve bandwidth in advance according to the composition of the network service type, so that the network performance is ensured.

The invention provides a train network security protection method, a train network security protection system and a computer readable storage medium. The train network security protection method comprises the following steps: carrying out identity authentication on an external system, and identifying an equipment category and a corresponding port of the external system; in response to the fact that the equipment category of the external system is maintenance equipment, starting a bastion host function; and in response to the fact that the equipment category of the external system is communication equipment, based on the configuration of the corresponding port, starting a security audit and intrusion detection function or a firewall function, the starting of the security audit and intrusion detection function or the firewall function comprising analysis of a message of a train-specific protocol and control data thereof. According to the train network safety protection method, multiple safety protection functions can be achieved at the same time, the number of safety devices and the size of a train communication network are reduced, analysis of a train special protocol is achieved, and therefore more efficient protection is achieved in a rail transit vehicle-mounted network.

The invention belongs to the technical field of train communication network scheduling. The invention provides a train-mounted communication network scheduling method and system. The method comprises the following steps: acquiring a communication data flow in a train communication network; the data stream is divided into a periodic data stream and a non-periodic data stream, and the periodic data stream is divided into a first-level periodic data stream and a second-level periodic data stream; dividing each basic time slot into a first-level time slice, a second-level time slice and a third-level time slice, and respectively transmitting the three data streams; according to an inter-time-slot load balancing algorithm, the periodic data streams are distributed, starting and stopping gating time of each time slice is set, three-level time slices are reserved for the non-periodic data streams, and a gating list of communication scheduling is obtained; and when the aperiodic data stream is present, scheduling optimization is carried out on the aperiodic data stream by adopting a trust mechanism to obtain final scheduling configuration. According to the method, important and real-time data are ensured to be transmitted preferentially, and real-time response and long-term scheduling balance of aperiodic data streams are realized.

The invention discloses a layered fault diagnosis method and device for a train communication network, and the method comprises the following steps: firstly, building a train communication network topology model, carrying out the parameter definition of each network node device and link in the model, then configuring a communication network and a network fault in the model, configuring the network fault into the communication network, and carrying out the fault diagnosis of the train communication network. Then collecting network performance index data in terminal equipment, preprocessing the index data, randomly dividing an index data set into a training set and a test set according to a proportion, then constructing a train communication network fault diagnosis model, and finally sending the training set to the fault diagnosis model for training and testing. And inputting the test set into the trained fault diagnosis model to identify different fault types. According to the method, efficient collection and effective detection of Ethernet train communication network fault data can be realized, and the problems of difficulty in data collection and insufficient diagnosis precision of the train communication network under a complex topological structure are effectively solved.

The invention belongs to the technical field of train communication network security, and relates to an Ethernet intrusion detection method and system, and the method comprises the steps: respectively obtaining physical domain data and information domain data from a train communication network, carrying out the normalization processing of a primary side network flow through employing a Z value standardization method, and obtaining a time domain feature; the basic flow, the communication protocol, the connection behavior and the like are converted into frequency domain signals through short-time Fourier transform, and frequency domain features are obtained; calculating the weight of each time domain feature and the weight of each frequency domain feature by using a grey correlation analysis method; respectively modeling by adopting a structured state space sequence model and a convolutional neural network to obtain two abstract feature representations; fusing the two abstract feature representations, and weighting the weight of each feature through an attention mechanism to obtain a fused feature representation; and analyzing the two features by using a target intrusion detection model, and identifying different intrusion attack events. According to the invention, the integrity of intrusion detection and the accuracy of detection are improved.

This invention provides an asynchronous scheduling method for TSN-5G train communication networks based on multi-agent reinforcement learning. The method includes: constructing a Markov model using a graph neural network based on the topology of a heterogeneous TSN-5G train scenario; defining the state space, action space, and reward function of the Markov model; inputting the data flow information to be processed in the heterogeneous TSN-5G network into a multi-agent reinforcement learning asynchronous scheduling algorithm; treating each base station and switch as an independent agent; updating the state space and action space of the Markov model through iterative computation; and outputting a scheduling scheme for the data flow to be processed. This invention proposes a comprehensive scheme that combines the utilization of wired and wireless channels and end-to-end jitter performance in TSN-5G to optimize data transmission, thereby achieving optimal control of end-to-end latency and jitter of the entire train network.

The application discloses a train communication network construction method and device, equipment and storage medium, applied to the train communication field, including: obtaining a wired train communication network composed of a backbone network node and a marshalling network node when a train is initially running, the backbone network node and the marshalling network node both have wired functions and wireless functions; numbering each marshalling network node by using a wired marshalling network in the wired train communication network; constructing a wireless marshalling network according to the numbering; and connecting the wireless marshalling network to the train backbone network. After the wired train communication network is initialized, the marshalling network node is numbered based on the wired marshalling network, the wireless marshalling network is constructed according to the numbering, the initialization of the wireless marshalling network is completed, the existing wired network architecture is compatible, the hardware change is small, and the stability is high, so that the maintenance cost is reduced, and the maintenance efficiency is improved.

The invention relates to the technical field of rail transit communication, provides a rail transit vehicle deterministic communication and control network architecture based on a time-sensitive network, and provides a train network deterministic data transmission scheduling control method based on the architecture. Then establishing a system model, constructing a TSN traffic scheduling problem constraint, ensuring that the end-to-end delay is less than the maximum delay requirement, and formulating a forwarding strategy of all scheduling traffic in the network on the basis of meeting jitter and delay; and an application verification scheme example and a mapping scheme of a data category and a time-sensitive network priority in the train communication network are given. According to the method, low-time-delay low-jitter high-reliability transmission of train time-sensitive flow is ensured, the real-time performance and reliability of train communication network data transmission are remarkably improved, and the method is particularly suitable for scenes such as high-speed railways and urban rail transit which have extremely high requirements on the real-time performance and reliability of data transmission.

The invention discloses a train door control method and device. The current speed of the target train is obtained through the train communication network in the target train, and a key speed data basis is provided for subsequent train door control. A zero-speed signal is generated under the condition that the current speed is smaller than the preset speed, and whether the train is in a low-speed state or not can still be accurately judged when the head train braking control unit breaks down. And the centralized control unit receives the zero-speed signal and generates a vehicle door centralized control enable signal only in a low-vehicle-speed state, so that the door is allowed to be controlled, and the safety of vehicle door control is enhanced. And finally, the train door centralized control enable signal is issued to the train door control unit in each train to carry out train door control, so that the whole train door can be accurately controlled when the head train brake control unit fails, and the stability and the safety of the train are enhanced.

The invention discloses a train communication network fault diagnosis method and device based on space-time diagram convolution, and the method comprises the following steps: collecting the communication data of each communication node during the operation of a train communication network, carrying out the preprocessing and analysis of the collected communication data, splicing the feature sets of each communication node into a feature set according to the column, and carrying out the recognition of the feature set; calculating a feature similarity relationship among the nodes for the features among the nodes in the training set, constructing a node similarity adjacency matrix through the feature similarity relationship, and then constructing a node static adjacency matrix according to the size of communication traffic among the nodes; an MS-STGCN model of train communication network fault diagnosis is constructed, spatial relations in different scales in a feature set are extracted, high accuracy is achieved through training and fault recognition, and the train communication network fault diagnosis method is used for achieving train communication network fault diagnosis. According to the method, time and space information of a train communication network is fused, and the feature extraction capability of the model is improved by using a multi-scale receptive field.

This invention discloses an intrusion detection method for train communication networks based on multi-scale residual networks, comprising the following steps: establishing a train communication network topology model and a traffic model for mixed transmission of normal traffic and intrusion traffic; collecting traffic data packets in the link as raw data samples in the intrusion detection dataset and preprocessing them to obtain data samples containing statistical features; constructing a train communication network intrusion detection MSRNet-GLAM model, checking and evaluating the performance of the MSRNet-GLAM model to enable it to learn key feature information of data samples at multiple scales; training and optimizing the MSRNet-GLAM model using an improved focus loss function, and performing intrusion detection on the traffic data of the train communication network to be detected to obtain the traffic detection results of the train communication network. This invention can effectively solve the problems of low security and difficulty in detecting intrusion traffic under the trend of large-scale data communication in train communication networks, ensuring the security of train networks.

The embodiment of the application discloses a train communication network topology structure and a data transmission method based on ECN and ETBN, which are used for improving the reliability and stability of train communication. The structure comprises: a first ECN, a second ECN, a third ECN and a fourth ECN arranged in each carriage of a train; a first ETBN, a second ETBN, a third ETBN and a fourth ETBN arranged in a backbone network head carriage of the train; the first ETBN, the second ETBN, the third ETBN and the fourth ETBN form an ETBN network layer; the first ECN and the second ECN of each carriage in the train form a first ECN network layer, and the third ECN and the fourth ECN of each carriage in the train form a second ECN network layer; and the ETBN network layer is located between the first ECN network layer and the second ECN network layer.

This invention relates to a method, equipment, and medium for coordinated control of main circuit breakers in electric locomotive formations. The method includes: determining the master locomotive in the formation; determining the node sequence of each locomotive in the formation based on the master locomotive; collecting overhead contact line parameters in real time; dynamically calculating the closing time interval of the main circuit breakers of each slave locomotive based on the grid parameters and the node sequence; prioritizing the closing of the main circuit breaker of the master locomotive; after the master locomotive's main circuit breaker closes first, sending a closing start signal to all slave locomotives through the train communication network; and ensuring that each slave locomotive, upon receiving the closing start signal, closes its own main circuit breaker according to its node sequence, after delaying the closing time interval. By combining a multi-level, high-fault-tolerant automatic formation mode identification mechanism with a dynamically adaptive main circuit breaker coordinated control strategy, precise, safe, and efficient control of the high-voltage main circuit of electric locomotives with different formation configurations is achieved.

This application belongs to the field of train communication network security technology, and relates to an Ethernet intrusion detection method and system. The method includes: acquiring physical domain data and information domain data from the train communication network; normalizing primary network flows using Z-value standardization to obtain time-domain features; converting basic traffic, communication protocols, and connection behaviors into frequency-domain signals using short-time Fourier transform to obtain frequency-domain features; calculating the weights of each time-domain feature and each frequency-domain feature using grey relational analysis; modeling using a structured state-space sequence model and a convolutional neural network to obtain two abstract feature representations; fusing the two abstract feature representations and weighting the weights of each feature using an attention mechanism to obtain a fused feature representation; and analyzing the two features using a target intrusion detection model to identify different intrusion attack events. This invention improves the completeness and accuracy of intrusion detection.

The invention relates to the technical field of rail transit, in particular to a fault diagnosis method for MVB-PHM, which comprises the following steps: firstly, acquiring historical MVB signal data, dividing the data into sub-intervals after normalization, counting the frequency of fault data, selecting the sub-interval with the maximum frequency as an initial seed, determining a fault threshold interval by variable step length growth, and determining the fault threshold interval; dividing the indexes into a first category of indexes of a single index and a second category of indexes needing multi-index coupling, then establishing a tree-shaped classifier, classifying a first hierarchy by using the first category of indexes, selecting a highest-purity index as a top-layer condition, selecting a second hierarchy by using the second category of indexes, and performing combined classification through information entropy sorting until the purity requirement is met; and finally, collecting MVB signal data in real time and extracting features, and obtaining a fault type by using the trained tree classifier and a mode discrimination method. According to the method, the tree classifier constructed in a data driving mode is adopted, a quantitative feedback mechanism is introduced, and accuracy and adaptivity of train communication network fault classification are achieved.

Provided in the present invention are an automatic networking method for a train communication network, and a storage medium. The method comprises: a management center reading a networking planning file, and constructing a logical topology; on the basis of a link layer discovery protocol, the management center interacting with a managed device, so as to construct a physical topology; if the physical topology matches the logical topology, the management center establishing a network configuration protocol connection with the managed device; the management center distributing automatic networking configuration data to the managed device; and on the basis of a publish-subscribe mode, the management center monitoring the state of the managed device.