Train compartment monitoring method and system based on intelligent agency and freight train

Abstract

The invention discloses a train compartment monitoring method and system based on intelligent agency and a freight train, and the device comprises the steps: receiving real-time data of various sensors in a current compartment and summarized data of a previous compartment sent by an intelligent agency node of the previous compartment, and processing the real-time data of various sensors and the summarized data of the previous compartment to obtain a data packet and a transmission command, and forwarding the data packet and the transmission command to a next compartment / ground server by using millimeter waves. According to the train compartment monitoring system, each module of the intelligent agent node processes data from the current compartment and the previous compartment, so that multi-dimensional monitoring data integration is realized, and the integrated data is transmitted to the next compartment and the ground server by utilizing millimeter waves; real-time monitoring and analysis of each compartment of the freight train are realized, difficulty in integration and analysis of monitoring information of the freight train is weakened, and safety guarantee is provided for operation of the freight train.

Description

Technical Field

[0001] This invention relates to the field of train technology, specifically to a train carriage monitoring method, system, and freight train based on intelligent agents. Background Technology

[0002] Currently, the application of intelligent technology in railway transportation is mostly limited to high-speed passenger transport. Freight trains lack intelligent equipment and rely heavily on manual troubleshooting, resulting in a waste of human resources and time.

[0003] With the increasing speed of freight trains, real-time monitoring and analysis in the field of safety assurance have significant engineering application value. Mature sensor technologies have formed a complete sensing system, overcoming the stringent operating conditions of freight trains. However, the overall monitoring of freight trains faces problems such as information accumulation from multiple sensors and information transmission lag. Furthermore, upgrading each type of monitoring would incur substantial financial costs. Therefore, there is an urgent need for an efficient and low-cost solution to overcome the dilemma of information integration and analysis in freight train monitoring, thereby providing safety assurance for freight train operations. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a train carriage monitoring method, system, and freight train based on intelligent agents.

[0005] An embodiment of the present invention provides a train carriage monitoring method based on intelligent agents, used for freight trains, comprising: Receive real-time data from various sensors in the current carriage and aggregated data from the previous carriage sent by the intelligent agent node of the previous carriage; The real-time data from the various sensors and the aggregated data from the previous carriage are processed to obtain data packets and transmission commands. The data packets and transmission commands are forwarded to the next carriage / ground server using millimeter-wave communication technology.

[0006] The present invention also provides a train carriage monitoring system based on intelligent agents, including an intelligent agent node for each carriage, a receiving antenna, and a ground server, wherein the intelligent agent node includes: A power supply module is used to provide power support for each module of the intelligent agent node; The data receiving module is used to receive real-time data from various sensors in the current carriage and the summary data from the previous carriage sent by the intelligent agent node of the previous carriage. The data fusion module is used to process the real-time data from various sensors and the aggregated data from the previous carriage to obtain data packets and transmission commands. The intelligent agent nodes constituting the data fusion module include carriage-level intelligent agent nodes and vehicle-mounted terminals. The millimeter-wave data transmission module uses millimeter-wave communication technology to forward the data packets and transmission commands to the next carriage (V-band) / ground server (E-band). The power supply module, data receiving module, data fusion module, and millimeter data transmission module are electrically connected in a chain series. If the current carriage is not the last carriage of the train, the data receiving module is simultaneously electrically connected to the data transmission module of the next carriage.

[0007] In summary, for the system, the power supply module, located in each carriage, provides power to the various modules of the intelligent agent node, reducing the mutual interference between carriages during data transmission caused by shared modules. Each module of the intelligent agent node processes data from the current carriage and the previous carriage, achieving multi-dimensional monitoring data integration, reducing data redundancy, and using millimeter-wave communication technology to transmit the processed data packets and transmission commands to the next carriage and the ground server. This enables real-time monitoring and analysis of each carriage of the freight train, mitigating the difficulties of information integration and analysis faced in freight train monitoring at a lower cost, and providing safety assurance for freight train operation.

[0008] The process of processing the real-time data from the various sensors and the aggregated data from the previous carriage to obtain data packets and transmission commands includes: Based on the source of the raw data, the real-time data from various sensors and the aggregated data from the previous carriage are divided into different data processing queues; After the data processing queue is tagged, priority-identified data packets are obtained. The priority identifier data packet is cached according to the transmission command; The priority identifier data packet is reconstructed to obtain a data packet that supports millimeter wave transmission.

[0009] In this way, the intelligent agent nodes at the carriage level and the central on-board terminal at the train level will process real-time data from various sensors and the aggregated data from the previous carriage based on the original data source to obtain data packets and transmission commands, which will facilitate the transmission of real-time operating data of each carriage of the freight train to the next carriage and the ground server.

[0010] The step of obtaining priority identifier data packets after tagging the data processing queue includes: Real-time identification of real-time data from various sensors in different data processing queues and aggregated data from the previous carriage; Each of the data processing queues is assigned a priority identifier, and the priority identifier data packet is obtained; The step of caching the priority identifier data packet according to the transmission command includes: By using tiered caching and differentiated replacement strategies, the lowest priority data is evicted, thereby balancing real-time performance, bandwidth, and storage costs, and providing support for subsequent data reconstruction.

[0011] In this way, real-time identification and labeling of various sensor data in different data processing queues are performed, and the lowest priority data is eliminated by using hierarchical caching and differentiated replacement strategies, thereby balancing real-time performance, bandwidth and storage costs, and providing support for the data reconstruction process in the subsequent real-time monitoring and analysis of each carriage of a freight train.

[0012] The process of reconstructing the priority identifier data packet to obtain a data packet supporting millimeter-wave transmission includes: The dead-zone compression strategy is used to remove data redundancy in the priority identifier data packet, while retaining the data points in the priority identifier data packet that can represent the original data. The data points are downsampled using Fourier transform to extract key frequency domain features, resulting in aggregated data packets. The data in the aggregated data packet is format-converted and encapsulated to obtain a data packet that is easy to transmit at high speed via millimeter waves.

[0013] In this way, by reconstructing the data in the priority identifier data packet through compression, aggregation, and encapsulation, the problems of data redundancy and data accumulation in the data packet are reduced, and the data packet that is easy to transmit at high speed via millimeter wave is obtained, which facilitates data transmission between intelligent agent nodes in different train carriages.

[0014] The step of forwarding the data packet and the transmission command to the next carriage / ground server using millimeter-wave communication technology includes: If the intelligent agent node is a carriage-level intelligent agent node, then the carriage-level intelligent agent node will forward the received data packets and transmission commands to the intelligent agent node of the next carriage using millimeter-wave communication technology.

[0015] Thus, if the intelligent agent node is a carriage-level intelligent agent node, the data packets and transmission commands received by the current carriage will be forwarded to the next carriage using V-band millimeter-wave communication technology, ensuring that all carriages of the freight train have undergone real-time data transmission and fusion, thereby realizing real-time monitoring and analysis of each carriage of the freight train.

[0016] The forwarding of the data packet and the transmission command to the next carriage / ground server includes: If the intelligent agent node is a train-level on-board terminal, the train-level on-board terminal will integrate the data packets and transmission commands received from the intelligent agent nodes of each carriage to obtain the whole vehicle data, and then send the whole vehicle data to the ground server using E-band millimeter wave communication technology.

[0017] Thus, if the intelligent agent node is a train-level onboard terminal, the current carriage integrates data packets and transmission commands from all carriages, and sends the integrated whole-vehicle data to the ground server using E-band millimeter-wave communication technology, ensuring that all carriages of the freight train can ultimately receive real-time monitoring and analysis from the ground server.

[0018] The train carriage monitoring method of the present invention also includes: The ground server evaluates the received data packets and transmission commands, analyzes the real-time status of the train carriages, and issues fault warnings to the train-level onboard terminal.

[0019] In this way, the ground server receives the integrated data of the entire train, ensuring that all carriages of the freight train can be monitored and analyzed in real time. It can also send fault alarms to the freight train, reducing the difficulties of information integration and analysis faced by freight train monitoring at a lower cost, and providing safety assurance for the operation of freight trains.

[0020] The data fusion module in the train carriage monitoring system of the present invention includes: The carriage-level intelligent agent node and the train-level on-board terminal, wherein the carriage-level intelligent agent node includes: The data classification submodule is used to divide the real-time data of the various sensors and the summary data of the previous carriage into different data processing queues according to the source of the original data, and to obtain priority identification data packets after labeling the data processing queues. The data caching submodule is used to cache priority identifier data packets according to the transmission command; The data reconstruction submodule is used to reconstruct the priority identifier data packet to obtain a data packet that supports millimeter wave transmission.

[0021] The present invention also provides a freight train equipped with a train carriage monitoring system based on intelligent agents, wherein the freight train applies the train carriage monitoring method described in any of the preceding claims.

[0022] Additional aspects and advantages of embodiments of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of the invention. Attached Figure Description

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is one of the flowcharts illustrating the train carriage monitoring method of the present invention; Figure 2 This is a schematic diagram of the modules of the train carriage monitoring system of the present invention; Figure 3 This is the second flowchart of the train carriage monitoring method of the present invention; Figure 4 This is the third flowchart of the train carriage monitoring method of the present invention; Figure 5 This is the fourth flowchart of the train carriage monitoring method of the present invention; Figure 6 This is the fifth flowchart of the train carriage monitoring method of the present invention; Figure 7 This is the sixth flowchart of the train carriage monitoring method of the present invention; Figure 8 This is the seventh flowchart of the train carriage monitoring method of the present invention. Detailed Implementation

[0024] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of the present invention, and should not be construed as limiting the embodiments of the present invention.

[0025] With the rapid development of the railway industry, railway freight transport has become an important part of the national economy. Faced with increasing transport demands and the need for more refined services, freight trains urgently require intelligent transformation. Currently, the application of intelligent technology in railway transport is mostly limited to high-speed passenger transport, lacking dynamic monitoring and analysis of freight trains during operation. The intelligent transformation of freight trains has great potential. On the one hand, with the maturity of sensor technology, a complete sensing system has been formed, overcoming the stringent operating conditions of freight trains; on the other hand, the adoption of intelligent technology is relatively low. Compared to passenger trains, freight trains lack intelligent equipment and rely heavily on manual troubleshooting, wasting significant human resources and time. As the speed of freight trains gradually increases, real-time monitoring and analysis in the field of safety assurance have immense engineering application value.

[0026] Because the monitoring of freight trains faces problems such as information accumulation from multiple sensors and information transmission lag, and upgrading each type of monitoring would cause significant financial pressure, there is an urgent need for an efficient and low-cost solution to break through the dilemma of information integration and analysis in freight train monitoring, and to provide safety assurance for freight train operation.

[0027] To address the aforementioned problems, this invention provides a train carriage monitoring method, system, and freight train based on intelligent agents. This method relies on data fusion and millimeter-wave high-capacity high-speed wireless communication technology to solve the current problems of cumbersome dynamic data collection and slow transmission in freight trains. It ensures high data integration and rapid transmission, supporting efficient computing on subsequent ground servers, thereby enabling real-time monitoring and analysis of freight trains and providing a reference for final inspection, early warning, and maintenance.

[0028] Based on the issues mentioned above, please refer to Figure 1 This invention provides a train carriage monitoring method based on intelligent agents for freight trains, comprising: 01: Receive real-time data from various sensors in the current carriage and the summary data from the previous carriage sent by the intelligent agent node of the previous carriage; 02: Process real-time data from various sensors and aggregated data from the previous carriage to obtain data packets and transmission commands; 03: Utilize millimeter-wave communication technology to forward data packets and transmit commands to the next carriage / ground server.

[0029] This invention also provides a train carriage monitoring system based on intelligent agents. The train carriage monitoring method of this invention can be implemented by the train carriage monitoring system of this invention. Please refer to [link to relevant documentation]. Figure 2 The train carriage monitoring system includes intelligent agent nodes, receiving antennas, and a ground server. The intelligent agent node comprises a data receiving module, a data fusion module, and a millimeter-wave data transmission module. The data receiving module receives real-time data from various sensors in the current carriage and aggregated data from the previous carriage sent by the intelligent agent node. The data fusion module processes the real-time data from various sensors and the aggregated data from the previous carriage to obtain data packets and transmission commands. The millimeter-wave data transmission module uses millimeter-wave communication technology to forward data packets and transmission commands to the next carriage (V-band) / ground server (E-band). The intelligent agent node of the train carriage monitoring system in this embodiment also includes a power supply module to provide power to the various modules of the intelligent agent node. The power supply module, data receiving module, data fusion module, and millimeter-wave data transmission module are electrically connected in a chain series. If the current carriage is not the last carriage of the train, the data receiving module is also electrically connected to the data transmission module of the next carriage. The power supply modules for the train carriage monitoring system are present in every carriage. In other words, each carriage has a self-powered intelligent agent node, which reduces the mutual interference of data transmission between carriages due to module sharing.

[0030] This invention utilizes a data fusion module within the intelligent agent node of a train carriage monitoring system to fuse, reconstruct, and store data, achieving preprocessing and integration of multi-source physical data. For real-time collection of multi-dimensional data from train carriages, the data receiving module can establish a physical connection with the current carriage's sensor network using a multi-protocol physical interface. Through this interface channel, it can fully receive real-time data from various sensors in the current carriage, improving the coverage of monitoring data collection from various sensors in the train carriage.

[0031] Specifically, the intelligent agent nodes constituting the data fusion module include carriage-level intelligent agent nodes and onboard terminals. The carriage-level intelligent agent nodes establish communication based on the high-speed, low-latency transmission channel of V-band millimeter waves, processing real-time data from various sensors and aggregated data from the previous carriage to obtain data packets and transmission commands, which are then forwarded to the next carriage, achieving data chaining between carriages and reducing data redundancy. The onboard terminal aggregates and caches data, facilitating rapid understanding of the real-time status of each carriage and enabling cross-carriage data correlation analysis. Furthermore, the onboard terminal can also utilize E-band millimeter waves to transmit data in real-time to a ground server. The ground server receives the data packets and transmission commands transmitted by the onboard terminal, enabling real-time monitoring of the entire train's status.

[0032] In one embodiment of the present invention, the multi-protocol physical interface can be an RS485 / CAN bus. The multi-protocol physical interface used in this embodiment only needs to be capable of receiving real-time data from various sensors in the carriage; the specific choice is not limited here. This embodiment is compatible with existing freight train monitoring sensor networks, establishing a physical connection with the monitoring sensor network through the multi-protocol physical interface to ultimately achieve comprehensive data collection. The collected data includes real-time data from various sensors and summary data from the previous carriage. The real-time data from various sensors specifically includes real-time collected data such as brake shoe temperature, train vibration, brake lever strain stress, and pressure. Simultaneously, the data receiving module is connected to the millimeter-wave data transmission module on the rear panel of the carriage via optical fiber to receive the summary data from the previous carriage.

[0033] In summary, for the system, the power supply module, located in each carriage, provides power to the various modules of the intelligent agent node, reducing the mutual interference of data transmission between carriages caused by shared modules. Each module of the intelligent agent node processes data from the current carriage and the previous carriage, achieving multi-dimensional monitoring data integration, reducing data redundancy, and transmitting the processed data packets and transmission commands to the next carriage and the ground server using millimeter-wave transmission technology. This enables real-time monitoring and analysis of each carriage of the freight train, mitigating the difficulties of information integration and analysis faced by freight train monitoring at a lower cost, and providing safety assurance for freight train operation.

[0034] In some implementations, please refer to Figure 3 Step 02 includes: 021: Based on the source of the raw data, real-time data from various sensors and aggregated data from the previous carriage are divided into different data processing queues; 022: After tagging the data processing queue, priority-identified data packets are obtained; 023: Packet data is identified by priority based on the transmission command; 024: Reconstruct the priority identifier data packet to obtain a data packet that supports millimeter wave transmission.

[0035] The data fusion module of the train carriage monitoring system in this embodiment of the invention further includes a data classification submodule, a data caching submodule, and a data reconstruction submodule. The data classification submodule is used to divide real-time data from various sensors and aggregated data from the previous carriage into different data processing queues based on the source of the original data, and to obtain priority-identified data packets after tagging the data processing queues. The data caching submodule is used to cache priority-identified data packets according to transmission commands. The data reconstruction submodule is used to reconstruct the priority-identified data packets to obtain data packets supporting millimeter-wave transmission.

[0036] Specifically, after the data fusion submodule receives real-time data from various sensors and aggregated data from the previous carriage from the data acquisition submodule, it first uses its own lightweight classification algorithm to automatically divide the raw data stream into different data processing queues according to its source. The received raw data streams of real-time sensor data include real-time sensor readings, and the raw data streams of aggregated data from the previous carriage include video streams, device status codes, etc.

[0037] After dividing the raw data into different data processing queues, the program assigns a priority identifier to each data packet, providing a basis for differentiated processing of data belonging to different priority identifiers. Then, an automatic transfer program transfers all the identified data to the data storage submodule. Simultaneously, it sends caching and reconstruction transmission commands to the data caching submodule, which then caches the priority-identified data packets according to the transmission commands.

[0038] Finally, the data reconstruction submodule is responsible for compressing, aggregating, transforming, and formatting the original data in the cached priority identifier data packets, transforming it from a low-value raw state into high-value, compact, and easily transmitted and analyzed information units, forming data packets with less data redundancy that support millimeter-wave transmission.

[0039] In one embodiment, the carriage-level intelligent agent node can use an edge computing unit, while the onboard terminal, unlike the carriage-level agent node, is a train-level data hub. The carriage-level intelligent agent node mainly performs the core functions of data classification, caching, and reconstruction. The onboard terminal, as the train-level hub, is mainly responsible for aggregating the data of the entire train, including data packets and transmission commands transmitted from the carriage-level agent nodes of each carriage of the freight train.

[0040] In this way, the intelligent agent nodes at the carriage level and the central on-board terminal at the train level will process real-time data from various sensors and the aggregated data from the previous carriage based on the original data source to obtain data packets and transmission commands, which will facilitate the transmission of real-time operating data of each carriage of the freight train to the next carriage and the ground server.

[0041] In some implementations, please refer to Figure 4 Step 022 includes: 0221: Real-time identification of real-time data from various sensors in different data processing queues and summary data from the previous carriage; 0222: Assign a priority identifier to each data processing queue and obtain the priority identifier data packet.

[0042] In some implementations, the data classification submodule is used to identify real-time data from various sensors and aggregated data from the previous carriage in different data processing queues, and to assign a priority identifier to each data processing queue to obtain a priority identifier data packet.

[0043] Specifically, after the raw data is divided into different data processing queues, the program identifies and labels the real-time data from various sensors and the summary data from the previous carriage in different data processing queues in real time according to their content, urgency and business value, and assigns a priority identifier to each data processing queue, providing a basis for subsequent differentiated processing of data belonging to different priority identifiers.

[0044] In some implementations, please refer to Figure 5 Step 023 includes: 0231: Use tiered caching and differentiated replacement strategies to evict the lowest priority data.

[0045] In some implementations, the data caching submodule is used to evict the lowest priority data using tiered caching and differentiated replacement strategies.

[0046] Specifically, the data caching submodule identifies data packets based on the priority of the transmission command. After being tagged, the data belongs to different priority identifiers. Based on the priority identifier, a hierarchical caching and differentiated replacement strategy is used to eliminate the lowest priority data, preventing the long-term retention of old data in the system from causing data accumulation.

[0047] In this way, real-time identification and labeling of various sensor data in different data processing queues are performed, and the lowest priority data is eliminated by using hierarchical caching and differentiated replacement strategies, thereby balancing real-time performance, bandwidth and storage costs, and providing support for the data reconstruction process in the subsequent real-time monitoring and analysis of each carriage of a freight train.

[0048] In some implementations, please refer to Figure 6 Step 024 includes: 0241: Use dead-zone compression to remove data redundancy in priority flag packets, and retain data points in priority flag packets that can represent the original data; 0242: Use Fourier transform to extract key frequency domain features and downsample the data points to obtain aggregated data packets; 0243: Perform format conversion and encapsulation on the data in the aggregated data packet to obtain a data packet that is easy to transmit at high speed via millimeter wave.

[0049] In some implementations, the data reconstruction submodule is used to remove data redundancy in the priority identifier data packet, retain data points in the priority identifier data packet that can represent the original data, downsample the data points to obtain an aggregated data packet, and perform format conversion and encapsulation on the data in the aggregated data packet to obtain a data packet that is easy to transmit at high speed via millimeter waves.

[0050] Specifically, after the data is cached in the form of priority-identified data packets according to the transmission command, the data in the priority-identified data packets will be reconstructed to obtain data packets supporting millimeter-wave transmission. The data reconstruction process first involves further compressing the data packets and removing data redundancy. In one embodiment, a dead-zone compression strategy can be used for time-series data packets, that is, a new value is only recorded when the change in data exceeds the dead-zone threshold (| x t - x t-1 |>ε, where ε represents the dead zone threshold), while using linear interpolation to approximate the trend of a data segment, retaining the data points in the priority identifier data packet that represent the trend of change (such as intervals with excessive slope changes) within a certain error range, so as to represent the original data with fewer data points.

[0051] After limited data compression, to improve the value density of the data in the data packet, data aggregation is performed on the data points that represent the original data. First, the compressed batch of data is recorded and aggregated. For example, 100 vibration readings within one second are decomposed into a superposition of several sine waves using Fourier transform, and a data file containing frequency domain features such as maximum amplitude, minimum amplitude, and average amplitude is extracted. Then, the extracted frequency components corresponding to the maximum, minimum, and average amplitudes are compressed and reconstructed. In one embodiment, the acquired vibration signal is assumed to be a discrete sequence. x 0, x 1,…, x N-1 Where N is the number of sampling points, its discrete Fourier transform expression is: , k =0, 1, ..., N -1 (Formula I) (Formula II) In formula I, X k It is the first k Complex representation of each frequency component; N This represents the number of sampling points; Indicates sampling point x n Complex exponential function; frequency component X k The amplitude is | X k | reflects the intensity of that frequency in the signal; in Formula II, f k For the first k The actual physical frequency corresponding to each frequency component; F S Sampling rate, N These are the sampling points.

[0052] After obtaining the calculation result of the above formula, only the first... k The actual physical frequency corresponding to each frequency component f k Amplitude of frequency component | X k | as the eigenvector matrix ( f k , | X k |) Upload, without having to upload all time-domain data points.

[0053] After the data reconstruction operation, a high-value, compact aggregated data package is formed.

[0054] Furthermore, to facilitate transmission, data encapsulation is required for the aggregated data packets. The data fusion module uses a processor to convert and encapsulate the reconstructed data, transforming it into data packets composed of information units easily transmitted over millimeter waves. Simultaneously, the data fusion module sends transmission commands to the millimeter wave transmission module to forward the data packets. In this way, by reconstructing the data in the priority-identifying data packets through compression, aggregation, and encapsulation, data redundancy and data accumulation are reduced, ultimately resulting in data packets easily transmitted over millimeter waves, facilitating data transmission between intelligent agent nodes in different train carriages.

[0055] In some implementations, please refer to Figure 7 Step 03 includes: 031: If the intelligent agent node is a carriage-level intelligent agent node, the carriage-level intelligent agent node will forward the received data packets and transmission commands to the intelligent agent node of the next carriage using V-band millimeter wave.

[0056] In some implementations, if the intelligent agent node is a carriage-level intelligent agent node, the millimeter data transmission module is used by the carriage-level intelligent agent node to forward the received data packets and transmission commands to the intelligent agent node of the next carriage using V-band millimeter waves.

[0057] Specifically, after determining whether the intelligent agent node is a carriage-level intelligent agent node or an onboard terminal, if it is a carriage-level intelligent agent node, after receiving the data packet and forwarding command from the data fusion module using the data transmission module, it will forward the data to the intelligent agent node in the next carriage through the V-band millimeter-wave transmission channel to continue repeating the processing described in steps 01 and 02. That is, when the current intelligent agent node is a carriage-level intelligent agent node, and the current carriage is an ordinary carriage without an onboard terminal, when the data transmission process reaches the current carriage, there are still other carriages in the freight train that have not uploaded real-time data from various sensors in the carriage or the summarized data from the previous carriage. Therefore, it is necessary to continue receiving and fusing data to obtain data packets and transmission commands from all carriages. At this time, the carriage-level intelligent agent node in the current carriage needs to forward the received data packet and transmission command to the intelligent agent node in the next carriage.

[0058] Thus, if the intelligent agent node is a carriage-level intelligent agent node, the data packets and transmission commands received by the current carriage will be forwarded to the next carriage using V-band millimeter waves, ensuring that all carriages of the freight train ultimately carry out real-time data transmission and fusion, and realizing real-time monitoring and analysis of each carriage of the freight train.

[0059] In some implementations, please refer to Figure 8 Step 03 also includes: 032: If the intelligent agent node is a train-level on-board terminal, the train-level on-board terminal will integrate the data packets and transmission commands received from the intelligent agent nodes in each carriage to obtain the whole vehicle data and send the whole vehicle data to the ground server using E-band millimeter wave.

[0060] In some implementations, if the intelligent agent node is a train-level onboard terminal, the millimeter data transmission module is used to integrate the data packets and transmission commands received from the intelligent agent nodes of each carriage to obtain the whole vehicle data and then use E-band millimeter waves to send the whole vehicle data to the ground server.

[0061] Specifically, the train-level on-board terminal is the data hub for all carriages of the entire freight train. In one embodiment, the train-level on-board terminal can be located at the front of the train. In other embodiments, the train-level on-board terminal can also be located at the rear of the train or in a carriage of the freight train other than the front and rear of the train.

[0062] The train-level onboard terminal receives and integrates vehicle-wide data based on data packets and transmission commands containing data aggregation requests sent by the intelligent agent nodes in the carriages. The onboard terminal receives collected vehicle-wide data, anomaly logs, and discrepancies, aggregates them in its internal large-capacity storage to obtain the vehicle-wide data, and then sends this data to the ground server according to the received transmission commands. The millimeter-wave transmission module of the train-level onboard terminal uses E-band millimeter waves to transmit data to the ground server in real time, enabling the ground server to monitor the vehicle's operational status in real time.

[0063] Thus, if the intelligent agent node is a train-level onboard terminal, the current carriage integrates data packets and transmission commands from all carriages, and sends the integrated whole-vehicle data to the ground server using E-band millimeter wave, ensuring that all carriages of the freight train can ultimately receive real-time monitoring and analysis from the ground server.

[0064] The train carriage monitoring method of this invention further includes: Please see Figure 2 The ground server evaluates the received data packets and transmission commands, analyzes the real-time status of the train carriages, and sends fault warnings to the train-level onboard terminal.

[0065] Specifically, the ground analysis center deploys ground servers corresponding to the received data packets and transmission commands. These ground servers, relying on their own high-performance computing units, can analyze and evaluate the real-time status of the entire freight train and output fault references for maintenance work, thus providing safety assurance for the operation of freight trains.

[0066] In this way, the ground server receives the integrated data of the entire train, ensuring that all carriages of the freight train can be monitored and analyzed in real time. It can also send fault alarms to the freight train, reducing the difficulties of information integration and analysis faced by freight train monitoring at a lower cost, and providing safety assurance for the operation of freight trains.

[0067] In particular, in the train carriage monitoring system of the present invention, the core of the data transmission module is a millimeter-wave communication unit, which includes a radio frequency front-end chip (RFFE), a baseband chip, a millimeter-wave antenna, a storage module, etc. Its hardware performance needs to meet the key technical indicators of millimeter-wave wireless connection, coverage, and networking between carriages and the train body in the freight train environment, so that the data sent by the fusion module can be forwarded to the next carriage or ground base station in a timely manner, and complete the efficient and reliable forwarding and transmission of large-capacity data.

[0068] In some implementations, the radio frequency front-end module (RFFE) mainly includes filters, power amplifiers (PAs), low-noise amplifiers, RF switches, antenna tuners, duplexers, and receivers / transmitters. The filters are responsible for filtering and frequency selection of transmitted and received signals, ensuring that signals are transmitted without interference at different frequencies. The power amplifier amplifies the RF signals in the transmit channel. The low-noise amplifier is mainly used for amplifying small signals in the receive channel; the RF switch is responsible for switching between the receive and transmit channels. The duplexer is responsible for duplex switching and filtering the RF signals in the receive / transmit channels.

[0069] In some implementations, the baseband chip is responsible for signal processing and protocol processing, including modulation / demodulation of data packets and transmission commands of the present invention, channel encoding / decoding, load balancing, encryption and other functions.

[0070] In some implementations, the millimeter-wave antenna is used to transmit and receive data. In this invention, a phased array antenna is used to improve the quality of millimeter-wave signal reception. The phased array antenna with 8×8 or more array elements can be used according to the actual needs of the freight train.

[0071] In some implementations, the storage module caches the information received and transmitted by the millimeter-wave communication module, and its storage space size can be customized according to the actual needs of the freight train.

[0072] The present invention also provides a freight train equipped with a train carriage monitoring system based on intelligent agents, which applies the train carriage monitoring method described in any of the preceding claims.

[0073] Specifically, any server or device included in the intelligent agent-based train carriage monitoring system of the present invention, such as carriage-level intelligent agent nodes and on-board terminals, can be installed on each carriage of the freight train disclosed in the present invention to implement the train carriage monitoring method disclosed in the present invention.

[0074] In the description of this specification, the terms "specifically," "furthermore," "particularly," "that is," etc., refer to specific features, structures, or characteristics described in connection with embodiments or examples that are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0075] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion comprising one or more steps for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0076] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A train carriage monitoring method based on intelligent agents, used for freight trains, characterized in that, include: Receive real-time data from various sensors in the current carriage and aggregated data from the previous carriage sent by the intelligent agent node of the previous carriage; The real-time data from the various sensors and the aggregated data from the previous carriage are processed to obtain data packets and transmission commands. The data packets and transmission commands are forwarded to the next carriage / ground server using millimeter-wave communication technology.

2. The train carriage monitoring method according to claim 1, characterized in that, The process of processing the real-time data from the various sensors and the aggregated data from the previous carriage to obtain data packets and transmission commands includes: Based on the source of the raw data, the real-time data from various sensors and the aggregated data from the previous carriage are divided into different data processing queues; After the data processing queue is tagged, priority-identified data packets are obtained. The priority identifier data packet is cached according to the transmission command; The priority identifier data packet is reconstructed to obtain a data packet that supports millimeter wave transmission.

3. The train carriage monitoring method according to claim 2, characterized in that, The step of obtaining priority identifier data packets after tagging the data processing queue includes: Real-time identification of real-time data from various sensors in different data processing queues and aggregated data from the previous carriage; Each of the data processing queues is assigned a priority identifier, and the priority identifier data packet is obtained; The step of caching the priority identifier data packet according to the transmission command includes: By using tiered caching and differentiated replacement strategies, the lowest priority data is evicted, thereby balancing real-time performance, bandwidth, and storage costs, and providing support for subsequent data reconstruction.

4. The train carriage monitoring method according to claim 2, characterized in that, The process of reconstructing the priority identifier data packet to obtain a data packet supporting millimeter-wave transmission includes: The dead-zone compression strategy is used to remove data redundancy in the priority identifier data packet, while retaining the data points in the priority identifier data packet that can represent the original data. The data points are downsampled using Fourier transform to extract key frequency domain features, resulting in aggregated data packets. The data in the aggregated data packet is format-converted and encapsulated to obtain a data packet that is easy to transmit at high speed via millimeter waves.

5. The train carriage monitoring method according to claim 1, characterized in that, The step of forwarding the data packet and the transmission command to the next carriage / ground server using millimeter-wave communication technology includes: If the intelligent agent node is a carriage-level intelligent agent node, then the carriage-level intelligent agent node will forward the received data packets and transmission commands to the intelligent agent node of the next carriage using V-band millimeter waves.

6. The train carriage monitoring method according to claim 1, characterized in that, The step of forwarding the data packet and the transmission command to the next carriage / ground server using millimeter-wave communication technology includes: If the intelligent agent node is a train-level on-board terminal, the train-level on-board terminal will integrate the data packets and transmission commands received from the intelligent agent nodes of each carriage to obtain the whole vehicle data and send the whole vehicle data to the ground server using E-band millimeter wave.

7. The train carriage monitoring method according to claim 1, characterized in that, Also includes: The ground server evaluates the received data packets and transmission commands, analyzes the real-time status of the train carriages, and issues fault warnings to the train-level onboard terminal.

8. A train carriage monitoring system based on intelligent agents, comprising an intelligent agent node for each carriage, a receiving antenna, and a ground server, characterized in that, The intelligent agent node includes: A power supply module is used to provide power support for each module of the intelligent agent node; The data receiving module is used to receive real-time data from various sensors in the current carriage and the summary data from the previous carriage sent by the intelligent agent node of the previous carriage. The data fusion module is used to process the real-time data from various sensors and the aggregated data from the previous carriage to obtain data packets and transmission commands. The intelligent agent nodes constituting the data fusion module include carriage-level intelligent agent nodes and vehicle-mounted terminals. The millimeter-wave data transmission module uses millimeter-wave communication technology to forward the data packets and transmission commands to the next carriage / ground server; The power supply module, data receiving module, data fusion module, and millimeter data transmission module are electrically connected in a chain series. If the current carriage is not the last carriage of the train, the data receiving module is simultaneously electrically connected to the data transmission module of the next carriage.

9. The train carriage monitoring system according to claim 8, characterized in that, The data fusion module includes: a carriage-level intelligent agent node and a train-level onboard terminal, wherein the carriage-level intelligent agent node includes: The data classification submodule is used to divide the real-time data of the various sensors and the summary data of the previous carriage into different data processing queues according to the source of the original data, and to obtain priority identification data packets after labeling the data processing queues. The data caching submodule is used to cache priority identifier data packets according to the transmission command; The data reconstruction submodule is used to reconstruct the priority identifier data packet to obtain a data packet that supports millimeter wave transmission.

10. A freight train equipped with a train carriage monitoring system based on intelligent agents, characterized in that, The freight train uses the train carriage monitoring method as described in any one of claims 1-7.

Images