Rail transit intelligent operation and maintenance method and device based on 5G-A network and medium

Abstract

The invention discloses a rail transit intelligent operation and maintenance method and device based on a 5G-A network and a medium. The method comprises the following steps: acquiring an actual operation state data set associated with each subsystem through a 5G-A network; dividing each actual operation state data set according to professional categories and data types to obtain actual operation state data subsets; key parameters are extracted from each actual operation state data subset according to a parameter extraction mode, and an actual operation state data key parameter set is formed; inputting the actual operation state data key parameter set into a target fault recognition model to obtain an actual fault corresponding to each professional category; and determining a target maintenance decision according to the fault type, the fault severity and the fault influence range of the actual fault, the pre-acquired maintenance resource distribution condition and the operation state data of other professional categories. According to the invention, the effects of high efficiency, precision, intelligence and collaboration of rail transit operation and maintenance are realized.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a rail transit intelligent operation and maintenance method, device and medium based on a 5G-A network. Background Art

[0002] In the existing rail transit operation and maintenance system, due to the limited bandwidth of the data transmission network of some rail transit lines, data congestion and delay are prone to occur when a large amount of data is transmitted concurrently, which makes it impossible for the intelligent operation and maintenance system to receive the latest status information of the equipment in time, affecting the rapid response and processing of faults. In addition, most analyses still remain at the level of simple statistics and threshold judgment, and fail to make full use of advanced algorithms such as machine learning and deep learning to explore the complex relationships and potential laws hidden in the data, making it difficult to achieve accurate fault prediction and equipment performance evaluation. In addition, the human resource dependence is high. The rail transit operation and maintenance work is highly dependent on professional and technical personnel, but the number and skill level of personnel are limited. When faced with large-scale operation and maintenance tasks and complex faults, human resource shortages are prone to occur, affecting the timeliness and quality of operation and maintenance work. In addition, for some equipment distributed in remote areas or difficult to reach locations, on-site maintenance is difficult. The existing remote maintenance technology is not mature enough to achieve comprehensive remote monitoring, diagnosis and repair of equipment, which increases the cost and difficulty of operation and maintenance. Summary of the invention

[0003] The present invention provides a method, device and medium for intelligent operation and maintenance of rail transit based on 5G-A network, which realizes the effects of efficient, precise, intelligent and coordinated operation and maintenance of rail transit.

[0004] According to one aspect of the present invention, a rail transit intelligent operation and maintenance method based on a 5G-A network is provided, which is applied to an intelligent operation and maintenance system; the method comprises:

[0005] Obtaining a set of actual operating status data associated with each subsystem included in the rail transportation transportation system through the 5G-A network; wherein the subsystem uses the 5G-A network to establish a communication connection with the intelligent operation and maintenance system;

[0006] Dividing each of the actual operation status data sets according to professional categories and data types to obtain actual operation status data subsets of different data types and professional categories;

[0007] Extracting key parameters related to the rail transportation traffic situation from each of the actual operation status data subsets in a parameter extraction method matching the data type to form a key parameter set of the actual operation status data corresponding to each professional category;

[0008] Input the key parameter set of the actual operation status data into the target fault identification model matching the professional category to obtain the corresponding fault identification model of each professional category, wherein the subsystem uses the 5G-A network to establish a communication connection with the intelligent operation and maintenance system;

[0009] The target maintenance decision is determined according to the fault type, fault severity and fault impact scope of the actual fault, the distribution of maintenance resources acquired in advance and the operating status data of other professional categories.

[0010] According to another aspect of the present invention, a rail transit intelligent operation and maintenance device based on a 5G-A network is provided, which is applied to an intelligent operation and maintenance system; the device comprises:

[0011] An acquisition module is used to acquire a set of actual operation status data associated with each subsystem included in the rail transportation transportation system through a 5G-A network; wherein the subsystem uses the 5G-A network to establish a communication connection with the intelligent operation and maintenance system;

[0012] A partitioning module, used to partition each of the actual operation status data sets according to professional categories and data types, to obtain actual operation status data subsets of different data types and professional categories;

[0013] An extraction module, used to extract key parameters related to the rail transportation traffic situation from each of the actual operation status data subsets according to a parameter extraction method matching the data type, to form a key parameter set of the actual operation status data corresponding to each professional category;

[0014] An evaluation module, used for inputting the key parameter set of the actual operation status data into a target fault identification model matching the professional category to obtain the actual fault corresponding to each professional category;

[0015] The determination module is used to determine the target maintenance decision according to the fault type, fault severity and fault impact scope of the actual fault, the distribution of maintenance resources acquired in advance and the operating status data of other professional categories.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the intelligent operation and maintenance method of rail transit based on the 5G-A network as described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the rail transit intelligent operation and maintenance method based on the 5G-A network as described in any embodiment of the present invention when executed.

[0021] According to another aspect of the present invention, a computer program product is provided, which includes a computer program, and when the computer program is executed by a processor, it implements the intelligent operation and maintenance method of rail transit based on 5G-A network as described in any embodiment of the present invention.

[0022] The technical solution of the embodiment of the present invention obtains the actual operation status data set associated with each subsystem in the rail transportation transportation system through the 5G-A network, thereby improving the data transmission efficiency; and, divides each actual operation status data set according to the professional category and data type, to obtain actual operation status data subsets of different data types and professional categories, thereby achieving the effect of clear division of data of different professional categories; and, for data of different data types, different parameter extraction methods are used to complete the extraction of key parameters in the corresponding actual operation status data subset, thereby ensuring the effect of high reliability of parameter extraction; and, the key parameters of the actual operation status data are input into the target fault identification model that matches the professional category, thereby ensuring that the actual fault output by the model has higher accuracy and precision; and, according to the fault type, fault severity and fault impact range of the actual fault, the distribution of maintenance resources acquired in advance and the operation status data of other professional categories, the target maintenance decision is determined, thereby achieving the effect of providing decision support from different professional dimensions, thereby ensuring the high availability of maintenance decisions.

[0023] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 It is a flow chart of a rail transit intelligent operation and maintenance method based on a 5G-A network provided by an embodiment of the present invention;

[0026] Figure 2 is a flow chart of another rail transit intelligent operation and maintenance method based on 5G-A network provided by an embodiment of the present invention;

[0027] Figure 3 This is an overall architecture diagram of a rail transit intelligent operation and maintenance system based on a 5G-A network provided by an embodiment of the present invention;

[0028] Figure 4 It is a schematic diagram of implementing data collection and parameter extraction provided by an embodiment of the present invention;

[0029] Figure 5 It is a schematic diagram of building and implementing a fault identification model provided by an embodiment of the present invention;

[0030] Figure 6 It is a structural schematic diagram of a rail transit intelligent operation and maintenance device based on a 5G-A network provided by an embodiment of the present invention;

[0031] Figure 7 It is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The rail transit operation and maintenance system in the prior art has the following disadvantages:

[0034] 1) Complex rail transit system: There are more than a dozen urban rail transit transportation organization and operation and maintenance management systems, such as electricity, signaling, communications, water supply and drainage, ventilation and even air conditioning, and many subsystems are distributed under different systems, including a variety of system equipment models. In addition, there are many and scattered rail transit stations, and various facilities and equipment are densely distributed in the stations. The entire network is huge, and it is becoming increasingly difficult to implement comprehensive and effective monitoring and overall management and control.

[0035] 2) Difficulty in operation and maintenance management: Due to the complexity of the rail transit system, the types of urban rail transit faults are diverse. Different systems will occupy different maintenance resources and need to be managed with different operation and maintenance methods and plans, which greatly increases the difficulty of operation and maintenance management. In the past, most of the operation and maintenance management was passive manual inspection and lacked intelligent system support. This is a change that needs to be made in the operation and maintenance management of urban rail transit under the 5G technology system.

[0036] 3) Inefficient emergency response: The operational safety of rail transit is closely related to the safety of passengers. Once an emergency occurs, such as a sudden large passenger flow at a station, a fire in a station, or equipment failure in a section, the key to emergency response is to quickly resume operations and minimize the impact on transportation. The current management methods have limited information transmission and low response efficiency, which affects the effectiveness of personnel evacuation and resource allocation.

[0037] 4) Issues of operation and maintenance personnel and operation and maintenance costs: Due to the rapid development of rail transit systems and the complexity and diversity of different rail transit subsystems, the training of relevant professional talents is difficult, and the operation and maintenance costs are also increasing day by day. In addition, due to the continuous increase in labor costs, increasing the number of professional talents or extending working hours will inevitably increase operating costs. At the same time, affected by physical fitness, external environment, working hours, etc., the work results of personnel are somewhat unstable. Reducing the workload of operation and maintenance personnel and reducing operation and maintenance costs are a major challenge for rail transit operation and maintenance management.

[0038] The present invention aims to provide a rail transit intelligent operation and maintenance method based on 5G-A technology, so as to comprehensively solve the many defects existing in the existing rail transit operation and maintenance technology, and realize the efficient, precise, intelligent and coordinated operation and maintenance of rail transit, which is mainly reflected in data transmission optimization, data analysis and processing upgrade, fault diagnosis and early warning enhancement, remote management and synergistic efficiency improvement.

[0039] In order to facilitate the understanding of the plan, the rail transit intelligent operation and maintenance system and 5G-A technology are explained.

[0040] The rail transit intelligent operation and maintenance system is an intelligent and automated operation and maintenance management system for rail transit equipment based on technologies such as the Internet of Things, big data, and artificial intelligence. The system can monitor the operating status of rail transit equipment in real time, predict failure risks, provide intelligent diagnosis and intelligent decision-making, and help companies improve equipment maintenance efficiency and reduce operating costs. It is mainly used for equipment maintenance of rail transit lines such as subways and light rails; real-time monitoring of key equipment such as trains, signals, communications, and electricity; intelligent management of other equipment such as station facilities and environments; and remote monitoring and command of operation and maintenance personnel.

[0041] 5G-A technology is based on the evolution and enhancement of 5G network in terms of function and coverage. It is a key information technology that supports the 3D and cloud-based Internet industry, the intelligent interconnection of all things, the integration of communication perception, and the flexibility of intelligent manufacturing. The transmission rate of 5G-A has been greatly improved compared with 5G, which can better meet the needs of services such as VR and AR that have extremely high bandwidth requirements, and bring users a smoother visual experience. Its low latency characteristics are more prominent, and can reach sub-millisecond latency. It is crucial for scenarios with strict real-time requirements such as industrial automation and intelligent transportation. It can ensure rapid response and precise control between devices, improve production efficiency and traffic safety. 5G-A supports more devices to connect at the same time, and can support millions of devices per square kilometer. It meets the needs of massive device interconnection in the era of the Internet of Things and promotes the development of fields such as intelligent transportation.

[0042] The "capability triangle" of 5G has been enhanced and expanded to a "capability hexagon", with new capabilities such as inter-sensory integration, inter-computing and intelligence integration, and ground-ground integration. (1) Inter-sensory integration: 5G-A integrates radar-like perception functions, which can realize the perception capabilities of positioning, ranging, and speed measurement of surrounding objects, while traditional wireless communication technologies such as 4G and Wi-Fi do not have this function. (2) Ground-ground integration: 5G-A realizes three-dimensional ubiquitous coverage of air, ground, and space, deeply integrates ground networks with satellite networks, breaks the limitations of traditional wireless communication technologies in geographical coverage, and can provide communication services in special scenarios such as remote areas, at sea, and in the air, ensuring seamless communication at any time and any place. (3) Computing and network integration: It can be closely integrated with cloud computing, edge computing and other technologies, bringing computing resources closer to users and data sources, realizing rapid data processing and analysis, improving the intelligence level of the network, and better supporting the development of applications such as artificial intelligence and big data.

[0043] In one embodiment, Figure 1This is a flowchart of a method for intelligent operation and maintenance of rail transit based on 5G-A network provided by an embodiment of the present invention. This embodiment can be applied to the situation where intelligent operation and maintenance of rail transit is performed by comprehensively considering status data of multiple professional categories. The method can be executed by an intelligent operation and maintenance device for rail transit based on 5G-A network. The intelligent operation and maintenance device for rail transit based on 5G-A network can be implemented in the form of hardware and / or software. The intelligent operation and maintenance device for rail transit based on 5G-A network can be configured in an intelligent operation and maintenance system. Figure 1 As shown, the method includes:

[0044] S110. Obtain a set of actual operating status data associated with each subsystem included in the rail transit transportation system through the 5G-A network.

[0045] Each subsystem uses a 5G-A network to establish a communication connection with the intelligent operation and maintenance system. In one embodiment, the 5G-A network is deployed in the operating environment of rail transit by combining distributed micro base stations and macro base stations; wherein the micro base stations are deployed in one of the following areas where personnel and equipment are concentrated: stations and vehicle depots; and the macro base stations are deployed in the rail lines and are set at intervals. In an embodiment, each subsystem can be connected to the intelligent operation and maintenance system through a micro base station or a macro base station.

[0046] For example, micro base stations can be deployed in key locations such as platforms, halls, escalators, train control rooms, and station machine rooms. Multiple macro base stations can be deployed on each rail line and set up according to pre-configured deployment intervals. For example, the deployment interval can be 5 kilometers, that is, a macro base station is deployed at 5 kilometers intervals on each rail line to ensure the communication quality and data transmission efficiency between each subsystem on the rail line (for example, rainfall detection system, engineering system, disaster prevention system, power monitoring system, etc.) and the intelligent operation and maintenance system.

[0047] The rail transportation system refers to an overall system that uses rails as vehicle travel paths for the transportation of goods or passengers. The rail transportation system may include some interrelated subsystems, for example, the rail transportation system may include but is not limited to one of the following: power monitoring system, signal system, integrated monitoring system, closed-circuit television monitoring system (referred to as CCTV system), engineering system, PIS system, AFC / ACC system, disaster prevention system, rainfall detection system and vehicle system, etc.

[0048] Among them, the actual operation status data set refers to the set of actual operation status data of each device or sensor in different subsystems at the current moment. Different subsystems have different actual operation status data contained in the corresponding actual operation status data set. Specifically, the power monitoring system is used to monitor and control the rail transit power supply system in real time. For example, the power monitoring system can monitor the voltage, current, power and other parameters of the power supply line, as well as the operating status of the power supply equipment, such as the working conditions of the transformer, switch cabinet and contact network, that is, the actual operation status data set associated with the power monitoring system may include but is not limited to at least one of the following: the voltage, current and power of the power supply line, as well as the operating status of the power supply equipment. The signal system is a system used to ensure the safe and efficient operation of rail transit. It can provide control information for the train by collecting relevant data such as the signal light status, signal transmission quality, and working parameters of the communication equipment in the signal system to ensure that each train can maintain a safe interval distance. It can be understood that the actual operation status data set associated with the signal system may include but is not limited to at least one of the following: the signal light status, signal transmission quality, and working parameters of the communication equipment. The integrated monitoring system is used to integrate the data of multiple subsystems (such as power monitoring system, signal system, rainfall monitoring system, etc.) in the rail transportation system to uniformly monitor and manage the operation status of rail transportation; the CCTV system is used to conduct video monitoring of stations, carriages and sections of rail transportation, that is, the actual operation status data set associated with the CCTC system is the image information collected from each place. The engineering system is used to construct, maintain and maintain tracks, bridges, tunnels and roadbeds to ensure the safe operation of trains. It can be understood that the actual operation status data set associated with the engineering system may include: the geometric state and unevenness of the track, the damage data of the rail head, rail waist, rail bottom and other parts, and the stress, deformation, temperature change of the track. The passenger information system (PIS system) is used to release information to passengers through the display terminals (such as display screens or broadcasts) set up in stations and carriages, that is, the actual operation status data set associated with the PIS system may include: relevant information released to passengers (such as train timetables, train arrival times, station layout maps, etc.). The automatic ticket collection system (AFC system) is used to manage the ticket sales, ticket checking and passenger flow information of rail transit; the ticket clearing system (ACC system) is used to clear and settle the ticket business of the entire rail transit, that is, the actual operation status data set associated with the AFC / ACC system may include but is not limited to at least one of the following: ticket sales information, ticket checking information and passenger flow information. The disaster prevention system is a system for responding to disaster events that occur during the operation of rail transit. Correspondingly, the actual operation status data set associated with the disaster prevention system may include but is not limited to: environmental information around rail transit transportation.The rainfall detection system is used to monitor the rainfall along the rail transit line. Correspondingly, the actual operation status data set associated with the rainfall detection system may include but is not limited to: the rainfall and other weather conditions around the rail transit. The vehicle system is used to inspect and maintain the vehicle itself and the associated vehicle inspection and depot facilities. Correspondingly, the actual operation status data set associated with the vehicle system may include but is not limited to: the use of the internal facilities of the vehicle, etc.

[0049] The actual operation status data in the actual operation status data set associated with each subsystem may be acquired through one of the following methods: dedicated monitoring software, dedicated monitoring interface, associated monitoring sensor, etc.

[0050] S120. Divide each actual operation status data set according to professional category and data type to obtain actual operation status data subsets of different data types and professional categories.

[0051] Among them, professional categories refer to different categories divided based on factors such as technical fields, equipment types or functional uses associated with the operating status data. Exemplarily, professional categories may include but are not limited to at least one of the following: track facility status, train operating status, passenger flow status, signal communication status, power supply status and environmental status. Correspondingly, the data associated with professional categories may include but are not limited to at least one of the following: track facility status data, train operating status data, passenger flow status data, signal communication status data, power supply status data and environmental status data.

[0052] Among them, track facility status data: through track inspection vehicles, rail flaw detectors, and various sensors installed on the track, such as strain sensors, displacement sensors, temperature sensors, etc., the geometric status and unevenness of the track, the damage data of the rail head, rail waist, rail bottom and other parts, as well as the stress, deformation, temperature change, etc. of the track are collected in real time. Train operation status data: through various sensors on the train, such as speed sensors, acceleration sensors, axle temperature sensors, vibration sensors, etc., the train operation parameters are obtained, including the train speed, acceleration, axle temperature, vibration, etc. At the same time, the train location information, operation schedule and other data are obtained through the train control system. Signal communication status data: collect relevant data of the signal system, including the status of signal lights, signal transmission quality, working parameters of communication equipment, etc. Power supply status data: through the power monitoring system, the voltage, current, power and other parameters of the power supply line are monitored, as well as the operating status of the power supply equipment, such as the working conditions of transformers, switch cabinets, contact networks, etc. Environmental data: Through the disaster prevention system, external meteorological system, and the collection of environmental information around rail transportation, such as weather conditions (rainfall, temperature, wind speed, etc.), geological conditions (seismic activity, soil moisture, etc.). Passenger flow status data: Through the automatic ticket vending and checking system, sorting system and closed-circuit television monitoring system, the passenger flow distribution of different time periods and different lines and video information of key locations such as stations are obtained, combined with intelligent image analysis technology, accurate statistics of passenger flow in and out of the station and the number of people waiting on the platform are obtained.

[0053] Data types refer to categories divided according to different data formats. For example, data types may include but are not limited to at least one of the following: text format, time series format, voice format, image format, and cross-section format; the actual operation status data subset includes a collection of operation status data of different data types and different professional categories. In an embodiment, after the actual operation status data sets associated with each subsystem are classified and integrated according to professional categories, the actual operation status professional data sets corresponding to each professional category can be obtained; and then each actual operation status professional data set is divided according to the data type to obtain actual operation status data subsets of different data types in each professional category.

[0054] S130. Extract key parameters related to the rail transit transportation situation from each actual operation status data subset in accordance with a parameter extraction method that matches the data type, to form a key parameter set of actual operation status data corresponding to each professional category.

[0055] The parameter extraction method refers to the implementation method of obtaining key parameters in each actual operation status data subset; the key parameters refer to the parameters associated with the rail transportation traffic situation in each actual operation status data subset. In one example, the parameter extraction method may include but is not limited to at least one of the following: one-hot encoding + fully connected neural network (FCNN); recurrent neural network algorithms such as long short-term memory neural network (LSTM); convolutional neural network (CNN); advanced algorithms such as transformer that integrate multi-head attention mechanism.

[0056] In the embodiment, key parameters related to the rail transportation traffic situation are extracted from the actual operation status data subsets of different professional categories and different data types. Among them, for the cross-sectional format data, unique hot encoding + fully connected neural network (FCNN) is used for parameter extraction; for time series data, long short-term memory neural network (LSTM) and other recurrent neural network algorithms are used for parameter extraction; for image format data, convolutional neural network (CNN) and other algorithms are used for parameter extraction; for text and voice data, transformer and other advanced algorithms that integrate multi-head attention mechanism are used for parameter extraction. Different neural networks have strong feature learning capabilities and processing capabilities for time series data, and can effectively mine complex patterns and potential laws in rail transportation traffic data. For track facilities, features such as track geometry and rail wear are extracted; for train operation, train delays and train punctuality are extracted; in terms of signal communication, signal equipment status and communication stability features are extracted; for passenger flow, peak values, valley values ​​and change trends of passenger flow are extracted to form key parameter sets of actual operation status data of different professional categories.

[0057] S140, inputting the key parameter set of the actual operation status data into the target fault identification model matching the professional category to obtain the actual fault corresponding to each professional category.

[0058] The target fault identification model is a large model for fault identification of each professional category; the actual fault is used to characterize the faults existing in each professional category. In the embodiment, since the operating status data corresponding to different professional categories are different, in order to improve the accuracy and precision of fault identification, a matching target fault identification model can be constructed for each professional category.

[0059] In one embodiment, the process of creating a target fault identification model includes: obtaining a training operation status data set corresponding to each professional category; inputting the training operation status data set into an initial fault identification model, and iteratively training the structure and parameters of the initial fault identification model using a hyperparameter search method until the output predicted fault matches the historical real fault, thereby obtaining a corresponding target fault identification model.

[0060] Among them, the training running state data set refers to the set of running state data used as training samples. In the embodiment, a training running state data set matching a professional category is identified and extracted from the total set of training running state data, and the training running state data set is used as an input parameter and input into the initial fault identification model to optimize the model structure and parameters of the initial fault identification model to determine the optimal parameters and structure of the fault identification model, so that the predicted fault output by the model matches the historical real fault, and the fault identification model composed of the parameters and structure that minimize the error is used as the final target fault identification model.

[0061] Of course, in order to improve the effectiveness and efficiency of training, the initial fault recognition model corresponding to each professional category is the same. Then the training operation status data sets of different professional categories are used to iteratively train the initial fault recognition model, and the target fault recognition models corresponding to different professional categories are obtained, thereby achieving the model customization effect of different professional categories and ensuring the accuracy of each target fault recognition model in fault recognition of the associated professional category.

[0062] S150: Determine a target maintenance decision based on the actual fault type, fault severity, and fault impact range, the pre-acquired maintenance resource distribution, and other professional categories of operating status data.

[0063] Among them, the fault type refers to the fault classification to which the actual fault belongs, and the fault classification can be carried out according to professional categories. For example, for the power supply system of the train, by learning its historical voltage, current, power and other data, a fault identification model corresponding to the power supply system is established. When abnormal fluctuations in real-time data are monitored through the fault identification model, the fault type can be quickly determined, such as inverter failure, contactor adhesion, etc.

[0064] The severity of a fault can be comprehensively considered and quantitatively evaluated from different angles. For example, the severity of an actual fault can be quantitatively evaluated based on factors such as the nature of the fault, the scope of fault impact, duration, and difficulty of maintenance. The scope of fault impact refers to the area, equipment, personnel, and business involved in a fault. The distribution of maintenance resources refers to the distribution and inventory of maintenance personnel, spare parts, maintenance equipment and other resources in the rail transit transportation system. Target maintenance decisions can include target maintenance time and target maintenance plan.

[0065] In an embodiment, it can be determined whether the actual fault can be remotely repaired based on at least one of the fault type, fault severity and fault impact scope of the actual fault, and the target maintenance time and target maintenance plan can be determined based on the distribution of maintenance resources and operating status data of other professional categories to obtain a corresponding target maintenance decision.

[0066] The technical solution of this embodiment obtains the actual operation status data set associated with each subsystem in the rail transportation transportation system through the 5G-A network, thereby improving the data transmission efficiency; and, each actual operation status data set is divided according to the professional category and data type, so as to obtain the actual operation status data subsets of different data types and professional categories, thereby achieving the effect of clear division of data of different professional categories; and, for data of different data types, different parameter extraction methods are used to complete the extraction of key parameters in the corresponding actual operation status data subset, thereby ensuring the effect of high reliability of parameter extraction; and, the key parameters of the actual operation status data are input into the target fault identification model that matches the professional category, thereby ensuring that the actual faults output by the model are more accurate and precise; and, fault identification achieves the effect of providing decision support from different professional dimensions, thereby ensuring the high availability of maintenance decisions.

[0067] In one embodiment, Figure 2 This is a flowchart of another rail transit intelligent operation and maintenance method based on 5G-A network provided by an embodiment of the present invention. This embodiment is based on the above embodiment and further explains the process of obtaining the actual operation status data set, the process of determining the actual operation status data subset, the process of determining the actual operation status data key parameter set, the process of determining the actual fault, and the process of determining the maintenance decision. Figure 2 As shown, the method includes:

[0068] S210. Using a unified data interface through the 5G-A network, obtain a set of initial operating status data associated with each subsystem included in the rail transit transportation system.

[0069] Among them, the unified data interface can be a data interface configured using a predefined set of universal data interface standards, and the data interface standards can include parameter types, data formats, and communication protocols of all subsystems; the initial operating status data set refers to a set of unprocessed operating status data obtained from each subsystem. In an embodiment, the unified data interface can connect data acquisition systems and data sources corresponding to different subsystems, can obtain initial operating status data from each subsystem included in the rail transit transportation system in real time and / or periodically, and construct a corresponding initial operating status data set with the initial operating status data associated with each subsystem.

[0070] In the embodiment, by developing a unified data interface and connecting data acquisition systems and data sources in different professional fields, it is possible to ensure that data from the track facility monitoring system, train operation control system, signal system, power supply system, and automatic ticket vending and checking system can be obtained stably and efficiently, and real-time transmission and timed synchronization of initial operating status data can be achieved, ensuring that data from each data source can be updated to the evaluation system in a timely manner, providing a basis for data fusion.

[0071] S220: Perform data preprocessing operations on the initial operating status data in the initial operating status data set to obtain a corresponding actual operating status data set.

[0072] Among them, the data preprocessing operation may include but is not limited to at least one of the following: noise reduction, removal of invalid values, removal of duplicate values, supplementation of missing values, data smoothing, data filtering, text segmentation, speech recognition, text vectorization, data labeling and data enhancement and other operations. In an embodiment, at least one of the above-mentioned data preprocessing operations is performed on the initial operating state data in the initial operating state data set to obtain the corresponding actual operating state data, and constitute the actual operating state data set associated with each subsystem. Exemplarily, the collected initial operating state data can be denoised to remove noise data caused by sensor failure, transmission interference and other reasons. For example, for sensor data that obviously deviates from the normal range, it is screened and eliminated by setting a reasonable threshold. For example, identify and process missing values, and use data interpolation, mean filling and other methods to supplement the missing data to ensure the integrity of the data.

[0073] S230. Associate and match each actual operation status data set according to the professional category to obtain an actual operation status professional data set that matches the professional category.

[0074] In an embodiment, the actual operation status professional data contained in the actual operation status professional data set corresponding to the same professional category can come from the actual operation status data sets associated with multiple subsystems. All matching actual operation status data can be extracted from the actual operation status data set associated with each subsystem according to the professional category, and then all the actual operation status data associated with a certain professional category are integrated to form an actual operation status professional data set matching the professional category. Based on key information such as time and geographical location, the actual operation status data from data sources corresponding to different subsystems are associated and matched according to the transportation situation of different professional categories. Exemplarily, based on the train operation status, the train operation speed, train operation time data and train energy consumption data are integrated to form a complete train operation status professional data set as the actual operation status professional data set corresponding to the train operation status.

[0075] S240. Divide each actual operation status professional data set according to data type to obtain actual operation status data subsets of different data types.

[0076] Based on each actual operation status professional data set, it is divided into a text format data set, a time series format data set, a voice format data set, an image format data set and a cross-sectional format data set according to different data types. It can be understood that each actual operation status professional data set includes a text format data set, a time series format data set, a voice format data set, an image format data set and a cross-sectional format data set.

[0077] S250: Acquire a parameter extraction method that matches the data type of each actual operation status data subset.

[0078] In an embodiment, a mapping relationship may be established between a data type and a parameter extraction method, and a matching parameter extraction method may be searched based on the data type of each actual running status data subset and the mapping relationship between the two.

[0079] S260. Extract key parameters of each professional category from the actual operation status data subset according to the parameter extraction method to obtain a key parameter set of actual operation status data corresponding to each professional category.

[0080] Key parameters related to the rail transit transportation situation can be extracted from actual operation status data subsets of different professional categories and different data types, and all key parameters corresponding to a professional category can be integrated to obtain the key parameter set of actual operation status data of the corresponding professional category.

[0081] For cross-sectional format data, one-hot encoding + fully connected neural network (FCNN) is used for parameter extraction. For time series data, recurrent neural network algorithms such as long short-term memory neural network (LSTM) are used for parameter extraction. For image format data, convolutional neural network (CNN) is used for parameter extraction. For text and voice data, advanced algorithms such as transformer that integrate multi-head attention mechanism are used for parameter extraction.

[0082] Exemplarily, one-hot encoding + fully connected neural network (FCNN) is used to extract track geometry and rail wear status from a cross-sectional format dataset, and advanced algorithms such as transformer that integrate a multi-head attention mechanism are used to extract the number of overhauls from a text format dataset. The track geometry, rail wear status and number of overhauls are then combined into a key parameter set of actual operating status data corresponding to the professional category of track facility status.

[0083] S270. Select a matching target fault identification model according to the professional category of the key parameter set of the actual operating status data.

[0084] It should be noted that the target fault identification model corresponding to each professional category is different. In an embodiment, the key parameters of the actual operation status data are divided according to the professional category, and a mapping relationship can be established between the professional category and the target fault identification model. A matching target fault identification model can be found based on the professional category of each set of key parameters of the actual operation status data and the mapping relationship between the two.

[0085] S280: Input the key parameter set of the actual operation status data into the target fault identification model to obtain the corresponding actual fault.

[0086] By inputting the actual operating status data parameter set into the pre-trained target fault identification model, the actual faults of the rail transit of this professional category can be output in real time.

[0087] S290: Determine whether to adopt a remote maintenance method based on at least one of the fault type, fault severity, and fault impact scope of the actual fault.

[0088] In an embodiment, whether to adopt a remote maintenance method may be determined based on at least one of the fault type, fault severity, and impact range. In an embodiment, when the fault type, fault severity, and fault impact range all meet the requirements for remote maintenance, remote maintenance is preferred; when the fault type of the actual fault meets the requirements for remote maintenance, remote maintenance is preferred; when the fault severity and fault impact range of the actual fault meet the requirements for remote maintenance, but the fault type does not meet the requirements for remote maintenance, on-site maintenance is preferred; when at least one of the fault type, fault severity, and fault impact range does not meet the requirements for remote maintenance, on-site maintenance is preferred.

[0089] S2100. Use the 5G-A network to obtain the distribution of maintenance resources in the rail transit transportation system from a pre-created maintenance resource database.

[0090] In an embodiment, the 5G-A network can be used to intelligently manage and deploy various resources in the rail transportation transportation system, such as maintenance personnel, spare parts, and maintenance equipment. A maintenance resource database is established to update the location, status, inventory, and other information of maintenance resources in real time. Through data interaction between the 5G-A network and other subsystems, maintenance resources are dynamically deployed according to operation and maintenance needs. For example, when a train suddenly breaks down and requires specific spare parts, the intelligent resource management and deployment system can quickly query the spare parts inventory in nearby warehouses and arrange the fastest transportation method to deliver the spare parts to the fault site. At the same time, according to the skill level, workload, and other conditions of the maintenance personnel, maintenance tasks are reasonably allocated to improve resource utilization efficiency and reduce operation and maintenance costs.

[0091] S2110. If remote maintenance is adopted, the actual fault maintenance site conditions are obtained in real time through the 5G-A network, and the target maintenance time and target maintenance plan are determined based on the distribution of maintenance resources and the operating status data of other professional categories as the target maintenance decision.

[0092] In the case of remote maintenance, remote expert collaboration can be achieved with the help of the high speed and low latency characteristics of the 5G-A network. Maintenance personnel can communicate with remote experts in real time on site through high-definition video calls and AR (augmented reality) technology. Remote experts can intuitively see the maintenance site and provide maintenance guidance through AR annotations in the maintenance personnel's field of vision, greatly improving maintenance efficiency and quality, especially for the handling of some complex faults, effectively shortening maintenance time by more than 30%.

[0093] S2120: If remote maintenance is not adopted, the target maintenance time and target maintenance plan are determined directly based on the distribution of maintenance resources and the operating status data of other professional categories as the target maintenance decision.

[0094] If remote maintenance is not used and a signal system failure is detected in a certain section, the maintenance plan can be quickly determined by analyzing the fault location, train operation plan, and the situation of nearby maintenance personnel and spare parts. For example, the nearest maintenance team can be dispatched to the fault location with the required spare parts, and the maintenance time can be reasonably arranged to minimize the impact on train operation.

[0095] In one embodiment, Figure 3 This is an overall architecture diagram of a rail transit intelligent operation and maintenance system based on a 5G-A network provided by an embodiment of the present invention. Figure 3 As shown, the rail transit intelligent operation and maintenance system can include: intelligent perception and data collection platform, intelligent data analysis and diagnosis platform, intelligent strategy and remote operation and maintenance system, and intelligent resource management and deployment system.

[0096] First, a comprehensive 5G-A network is built on the lines, stations, depots and trains of the rail transit intelligent operation and maintenance system. A combination of distributed micro base stations and macro base stations is used to ensure that network signals are covered without dead spots in complex rail transit environments. Micro base stations are deployed in densely populated and equipment-intensive areas such as stations and depots, while macro base stations are set up at intervals along the track lines to ensure stable network connection during high-speed train travel. Using 5G-A's carrier aggregation technology, multiple frequency band resources are integrated to greatly increase network bandwidth and meet the high-speed transmission requirements of massive operation and maintenance data. For example, when the train is running at high speed, it can still ensure that high-definition video surveillance data and a large amount of sensor data collected by on-board equipment are transmitted to the operation and maintenance center in real time, with a transmission rate of up to several Gbps. At the same time, through 5G-A's network slicing technology, an independent dedicated network slice is created for rail transit operation and maintenance, isolating it from other business networks such as passenger communications, ensuring high reliability and low latency of the operation and maintenance network, and the latency of the operation and maintenance network slice can be controlled within 1 millisecond, effectively avoiding the impact of network congestion and data transmission delay on operation and maintenance work.

[0097] For the intelligent perception and data acquisition platform: Based on the low latency and high reliability characteristics of the 5G-A network, the platform integrates the monitoring data of sensors in various key parts of rail transit, such as trains, tracks, power supply systems, and signal systems. Realize real-time and accurate collection of the operating status of equipment. For example, the vibration sensor installed on the train running gear can collect vibration data at microsecond intervals and transmit it to the data processing center in real time through the 5G-A network. The data processing center can analyze the health status of the train running gear in real time and promptly discover potential fault hazards, such as wheel tread scratches and bearing wear. Image sensors collect high-definition images of track surfaces, tunnel structures, etc. The image data transmitted through the 5G-A network can be used to detect safety hazards such as foreign objects on the track and cracks in the tunnel. The clarity and real-time performance of its image transmission far exceed traditional operation and maintenance systems.

[0098] For the intelligent data analysis and diagnosis platform: relying on the powerful data transmission capability of the 5G-A network, the collected massive operation and maintenance data are aggregated to the intelligent data analysis and diagnosis platform. The platform adopts a combination of big data processing technology and artificial intelligence algorithms. First, using big data storage and management technology, the massive operation and maintenance data is efficiently stored and quickly retrieved, and a unified data fusion framework is built to effectively solve the problem of inconsistent data formats, standards and semantics of various subsystems of rail transit. Seamless integration and sharing of data from different sources and types is achieved. Then, machine learning algorithms are used to learn and model the normal operation mode and fault characteristics of the equipment. By comparing and analyzing the real-time collected data with the model, accurate diagnosis and prediction of equipment failures can be achieved. For example, for the power supply system of the train, by learning its historical voltage, current, power and other data, a power supply system health model is established. When the real-time data fluctuates abnormally, the fault type can be quickly determined, such as inverter failure, contactor adhesion, etc., and the development trend of the fault can be predicted, and maintenance plans can be formulated in advance, effectively reducing the equipment failure rate and maintenance costs.

[0099] For intelligent decision-making and remote collaborative operation and maintenance system: Based on the results of the intelligent data analysis and diagnosis platform, the intelligent decision-making and remote collaborative operation and maintenance module comes into play. This module uses intelligent algorithms to make the best maintenance decisions based on the severity of the fault, the scope of impact, and the distribution of maintenance resources. For example, when a signal system fault is detected in a certain section, the module quickly determines the maintenance plan by analyzing the fault location, train operation plan, and the situation of nearby maintenance personnel and spare parts, such as dispatching the nearest maintenance team to carry the required spare parts to the fault location, and reasonably arranging the maintenance time to minimize the impact on train operation. At the same time, with the high rate and low latency characteristics of the 5G-A network, remote expert collaboration is realized. Maintenance personnel can communicate with remote experts in real time on site through high-definition video calls, AR (augmented reality) technology, etc. Remote experts can intuitively see the maintenance site situation and provide maintenance guidance through AR annotations in the maintenance personnel's field of vision, greatly improving maintenance efficiency and quality, especially for the handling of some complex faults, effectively shortening the maintenance time by more than 30%.

[0100] For the intelligent resource management and deployment system: the system uses the 5G-A network to intelligently manage and deploy various resources in rail transit operation and maintenance, such as maintenance personnel, spare parts, and maintenance equipment. Establish a resource database to update the location, status, inventory and other information of resources in real time. Through data interaction with other subsystems through the 5G-A network, resources are dynamically deployed according to operation and maintenance needs. For example, when a train suddenly breaks down and requires specific spare parts, the intelligent resource management and deployment system can quickly query the spare parts inventory in nearby warehouses and arrange the fastest transportation method to deliver the spare parts to the fault site. At the same time, according to the skill level and workload of the maintenance personnel, the maintenance tasks are reasonably allocated to improve resource utilization efficiency and reduce operation and maintenance costs.

[0101] In this embodiment, the real-time performance of data can be guaranteed by high bandwidth. 5G-A technology has the characteristics of ultra-high speed, and its bandwidth is greatly improved compared with traditional communication technology. This enables a large amount of equipment status data in the rail transit system, such as high-definition video monitoring data on trains, massive detection data collected by sensors along the track, etc., to be quickly transmitted to the operation and maintenance center in a short time. For example, in the traditional operation and maintenance mode, it may take several minutes to transmit high-definition video of the running gear of the train, but the present invention uses 5G-A technology to shorten the transmission time to a few seconds, realizing real-time data transmission, and the operation and maintenance personnel can obtain the latest status of the equipment in the first place.

[0102] In addition, the low latency of the 5G-A network (which can reach sub-millisecond level) ensures the timeliness of data interaction by improving response speed through low latency. When an abnormality occurs in rail transit equipment, such as a failure in the train braking system, the data collected by the sensor can be transmitted to the operation and maintenance center in near real time, and the operation and maintenance center can also quickly send control instructions back to the equipment. This greatly shortens the time interval from fault discovery to response. Compared with existing technologies, the fault response time can be shortened by more than 30%, effectively reducing the impact of faults on rail transit operations.

[0103] In addition, massive connections support system scalability. The rail transit system contains many devices. 5G-A technology can support the simultaneous connection of massive devices and accommodate millions of connections per square kilometer. This facilitates the expansion of the rail transit system in the future. Whether adding new trains, more track monitoring equipment or IoT devices in station facilities, they can easily access the network without network congestion, ensuring the sustainable development of the operation and maintenance system.

[0104] Moreover, all-round perception of equipment status: With the help of 5G-A network, the present invention can integrate a large number of high-precision sensors distributed in various corners of trains, tracks, power supply systems and station equipment. These sensors can comprehensively collect equipment operating parameters, such as temperature, pressure, vibration, current and other data, to achieve all-round perception of equipment status.

[0105] In addition, intelligent diagnosis can warn of faults in advance: using big data analysis platforms and artificial intelligence algorithms, combined with real-time data transmitted by 5G-A, in-depth analysis of the equipment's operating status can be performed. Compared with existing technologies, the accuracy of fault prediction can be improved by about 40%, and early warning can be given at the budding stage of faults, giving maintenance personnel more preparation time and effectively avoiding further deterioration of equipment faults.

[0106] Finally, by utilizing the high speed and low latency characteristics of 5G-A technology, maintenance personnel can collaborate with remote experts in real time on site through high-definition video calls and augmented reality (AR) technology. Remote experts can intuitively see the situation on the maintenance site and provide maintenance guidance through AR annotations in the maintenance personnel's field of vision. This remote collaboration method can effectively solve the problem of complex fault maintenance, especially for some maintenance work that requires special skills or experience, which can ensure the quality of maintenance while reducing the time and cost of experts going to the site.

[0107] In one embodiment, Figure 4 FIG. 1 is a schematic diagram of implementing data collection and parameter extraction provided by an embodiment of the present invention. Figure 4 As shown, the implementation process of data collection and parameter extraction includes the following steps:

[0108] Step 1: Collect initial operating status data.

[0109] In an embodiment, initial operation status data of a track facility monitoring system, a train operation control system, a signal system, a power supply system, an automatic ticket vending and checking system, etc. are obtained to construct an initial operation status data set.

[0110] Step 2: preprocess the initial operation status data.

[0111] Data preprocessing includes at least one of the following operations: noise reduction, invalid value extraction, duplicate value deletion, missing value supplementation, data smoothing, filtering, text segmentation, speech recognition, text vectorization, data annotation, and data enhancement. The initial running state data set is preprocessed to obtain the actual running state data set.

[0112] Step 3: divide the actual operation status data set according to professional categories to obtain a professional data set of actual operation status that matches each professional category.

[0113] In an embodiment, the actual operation status professional data set obtained according to professional categories includes the following sets: track facility status data, train operation status data, passenger flow status data, signal communication status data, power supply status data and environment status data.

[0114] Step 4: divide each actual operation status professional data set according to the data type to obtain actual operation status data subsets of different data types.

[0115] The actual operation status professional data set of each professional category includes data types. Each actual operation status professional data set is divided to obtain actual operation status data subsets of different data types.

[0116] Step 5: extract key parameters of each professional category from the actual operation status data subset using parameter extraction method to obtain the key parameter set of actual operation status data corresponding to each professional category.

[0117] Exemplarily, the professional category is rail transit status, and the corresponding key parameter set of actual operation status data includes: track geometry status, rail wear status and number of overhauls; the professional category is train operation status, and the corresponding key parameter set of actual operation status data includes: train delay rate, average recovery time of train delays, train operation punctuality rate and capacity resource utilization status; the professional category is passenger flow status, and the corresponding key parameter set of actual operation status data includes: temporal and spatial distribution of passenger flow in the station, passenger congestion and passenger travel characteristics; the professional category is signal communication status, and the corresponding key parameter set of actual operation status data includes: signal equipment status, communication stability, signal alarm rate and communication alarm rate; the professional category is power supply status, and the corresponding key parameter set of actual operation status data includes: contact network status, foreign object intrusion rate and track circuit stability; the professional category is environmental status, and the corresponding key parameter set of actual operation status data includes: line condition risk, geographical environment risk, climate condition risk and natural disaster risk rate.

[0118] In this embodiment, by adopting a parameter extraction method that matches the professional category to extract key parameters, high reliability of parameter extraction is ensured.

[0119] In one embodiment, Figure 5 It is a schematic diagram of constructing and implementing a fault identification model provided by an embodiment of the present invention.

[0120] According to the task requirements of rail transit transportation situation risk assessment, a target fault identification model for rail transit professional situation based on an adaptive large model is constructed with different data features as input parameters. A special interface is designed at the input layer of the model to receive multi-source data of different professional categories and convert it into a vector representation suitable for model processing through the embedding layer. In the middle layer of the model, multiple hidden layers are set to learn high-level features in the data through nonlinear transformation of neurons. In the output layer, actual faults of different professional categories are used as output results.

[0121] Based on the key parameters of the data obtained by the multi-source data acquisition and processing module, the model is trained by cross-validation, and the model structure and parameters (number of hidden layers, number of neurons, learning rate, optimizer, Dropout rate, etc.) are optimized by hyperparameter search to determine the optimal parameters and structure of the model so that the predicted faults of the model match the actual historical faults.

[0122] like Figure 5 As shown in the figure, the model processing process is explained by taking the fault identification of rail facilities as an example:

[0123] The track geometry history state, rail wear history state and historical overhaul times in the training operation status data set are used as input parameters and input into the initial fault identification model. The number of hidden layers, neurons, learning rate, optimizer and Dropout rate in the initial fault identification model are optimized and adjusted until the predicted fault of the output track facility status matches the corresponding historical real fault. Then, the target fault identification model corresponding to the track facility status is obtained and the model training process is completed. If the error is greater than the pre-configured set value, the number of hidden layers, neurons, learning rate, optimizer and Dropout rate are returned and adjusted, and the training is continued until the output predicted fault matches the historical real fault.

[0124] In this embodiment, by creating target fault identification models of different professional categories and using the target fault identification models that match the professional categories to perform fault identification, the accuracy of fault identification is improved.

[0125] In one embodiment, Figure 6 Schematic diagram of a rail transit intelligent operation and maintenance device based on a 5G-A network provided by an embodiment of the present invention. Figure 6 As shown, the device includes: an acquisition module 610, a division module 620, an extraction module 630, an evaluation module 640 and a determination module 650.

[0126] The acquisition module 610 is used to acquire a set of actual operation status data associated with each subsystem included in the rail transportation transportation system through the 5G-A network; wherein the subsystem uses the 5G-A network to establish a communication connection with the intelligent operation and maintenance system;

[0127] A partitioning module 620 is used to partition each actual operation status data set according to professional categories and data types to obtain actual operation status data subsets of different data types and professional categories;

[0128] An extraction module 630 is used to extract key parameters related to the rail transportation traffic situation from each subset of actual operation status data in accordance with a parameter extraction method matching the data type, so as to form a key parameter set of actual operation status data corresponding to each professional category;

[0129] Evaluation module 640, used to input the key parameter set of actual operation status data into the target fault identification model matching the professional category to obtain the actual fault corresponding to each professional category;

[0130] The determination module 650 is used to determine the target maintenance decision according to the fault type, fault severity and fault impact scope of the actual fault, the distribution of maintenance resources acquired in advance and other professional categories of operating status data.

[0131] In one embodiment, the acquisition module 610 includes:

[0132] An acquisition unit, used to acquire an initial operation status data set associated with each subsystem included in the rail transit transportation system through a 5G-A network using a unified data interface;

[0133] The preprocessing unit is used to perform data preprocessing operations on the initial operating status data in the initial operating status data set to obtain a corresponding actual operating status data set.

[0134] In one embodiment, the partitioning module 620 includes:

[0135] A classification unit, used to associate and match each actual operation status data set according to a professional category, to obtain an actual operation status professional data set that matches the professional category;

[0136] The partitioning unit is used to partition each actual operation status professional data set according to the data type to obtain actual operation status data subsets of different data types.

[0137] In one embodiment, the extraction module 630 includes:

[0138] An acquisition unit, used for acquiring a parameter extraction method matching a data type of each actual operation status data subset;

[0139] The extraction unit is used to extract the key parameters of each professional category from the actual operation status data subset according to the parameter extraction method, and obtain the key parameter set of the actual operation status data corresponding to each professional category.

[0140] In one embodiment, the evaluation module 640 includes:

[0141] A selection unit, used to select a matching target fault identification model according to the professional category of the key parameter set of the actual operation status data;

[0142] The evaluation unit is used to input the key parameter set of the actual operating status data into the target fault identification model to obtain the corresponding actual fault.

[0143] In one embodiment, the determination module 650 includes:

[0144] A first determination unit, configured to determine whether to adopt a remote maintenance method based on at least one of a fault type, a fault severity, and a fault impact range of the actual fault;

[0145] An acquisition unit, configured to acquire the distribution of maintenance resources in the rail transit transportation system from a pre-created maintenance resource database using a 5G-A network;

[0146] The second determination unit is used to obtain the maintenance site situation of the actual fault in real time through the 5G-A network if a remote maintenance method is adopted, and determine the target maintenance time and target maintenance plan based on the distribution of maintenance resources and the operating status data of other professional categories as a target maintenance decision;

[0147] The third determination unit is used to determine the target maintenance time and the target maintenance plan as the target maintenance decision directly based on the distribution of the maintenance resources and the operation status data of other professional categories if the remote maintenance method is not adopted.

[0148] In one embodiment, the process of creating a target fault identification model includes:

[0149] Obtain the training running status data set corresponding to each professional category;

[0150] The training operation status data set is input into the initial fault identification model, and the structure and parameters of the initial fault identification model are iteratively trained using a hyperparameter search method until the output predicted fault matches the historical real fault, thereby obtaining the corresponding target fault identification model.

[0151] In one embodiment, the 5G-A network is deployed in the operating environment of rail transit by combining distributed micro base stations and macro base stations; wherein the micro base stations are deployed in one of the following areas where personnel and equipment are concentrated: stations and depots; and the macro base stations are deployed in the rail lines and are set at intervals.

[0152] The rail transit fault identification device based on 5G-A network provided in an embodiment of the present invention can execute the rail transit fault identification method based on 5G-A network provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0153] In one embodiment, Figure 7is a structural block diagram of an electronic device provided by an embodiment of the present invention, such as Figure 7 As shown, a schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0154] like Figure 7 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0155] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0156] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as the intelligent operation and maintenance method of rail transit based on the 5G-A network.

[0157] In some embodiments, the intelligent operation and maintenance method for rail transit based on the 5G-A network may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the intelligent operation and maintenance method for rail transit based on the 5G-A network described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the intelligent operation and maintenance method for rail transit based on the 5G-A network by any other appropriate means (e.g., by means of firmware).

[0158] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0159] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0160] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0161] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0162] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0163] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0164] An embodiment of the present invention also provides a computer program product, including a computer program, which, when executed by a processor, can implement a rail transit intelligent operation and maintenance method based on a 5G-A network as provided in any embodiment of the present application.

[0165] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present application, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or similar programming language. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0166] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0167] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A rail transit intelligent operation and maintenance method based on 5G-A network, characterized in that: Applied to intelligent operation and maintenance system; the method comprises: Obtaining a set of actual operating status data associated with each subsystem included in the rail transportation transportation system through the 5G-A network; wherein the subsystem uses the 5G-A network to establish a communication connection with the intelligent operation and maintenance system; Dividing each of the actual operation status data sets according to professional categories and data types to obtain actual operation status data subsets of different data types and professional categories; Extracting key parameters related to the rail transportation traffic situation from each of the actual operation status data subsets in a parameter extraction method matching the data type to form a key parameter set of the actual operation status data corresponding to each professional category; Inputting the key parameter set of the actual operating status data into a target fault identification model matching the professional category to obtain the actual fault corresponding to each professional category; The target maintenance decision is determined according to the fault type, fault severity and fault impact scope of the actual fault, the distribution of maintenance resources acquired in advance and the operating status data of other professional categories.

2. The method according to claim 1, characterized in that The actual operation status data set associated with each subsystem included in the rail transportation system is obtained through the 5G-A network, including: Using a unified data interface through the 5G-A network to obtain the initial operating status data set associated with each subsystem included in the rail transportation system; A data preprocessing operation is performed on the initial operating status data in the initial operating status data set to obtain a corresponding actual operating status data set.

3. The method according to claim 1, characterized in that Each of the actual operation status data sets is divided according to professional categories and data types to obtain actual operation status data subsets of different data types and professional categories, including: Associating and matching each of the actual operation status data sets according to professional categories to obtain actual operation status professional data sets matching the professional categories; Each of the actual operation status professional data sets is divided according to data type to obtain actual operation status data subsets of different data types.

4. The method according to claim 1, characterized in that: The key parameters related to the rail transportation traffic situation are extracted from each of the actual operation status data subsets in a parameter extraction method matching the data type to form a key parameter set of the actual operation status data corresponding to each professional category, including: Acquire a parameter extraction method that matches the data type of each of the actual operation status data subsets; The key parameters of each professional category are extracted from the actual operation status data subset according to the parameter extraction method to obtain the key parameter set of the actual operation status data corresponding to each professional category.

5. The method according to claim 1, characterized in that The step of inputting the key parameter set of the actual operation status data into a target fault identification model matching the professional category to obtain the actual fault corresponding to each professional category includes: Selecting a matching target fault identification model according to the professional category of the key parameter set of the actual operating status data; The key parameter set of the actual operating status data is input into the target fault identification model to obtain the corresponding actual fault.

6. The method according to claim 1, characterized in that The target maintenance decision is determined according to the fault type, fault severity and fault impact scope of the actual fault, the distribution of maintenance resources acquired in advance and the operation status data of other professional categories, including: Determining whether to adopt a remote maintenance method based on at least one of the fault type, fault severity and fault impact scope of the actual fault; Using the 5G-A network, obtaining the distribution of maintenance resources in the rail transit transportation system from a pre-created maintenance resource database; If remote maintenance is adopted, the maintenance site conditions of the actual fault are obtained in real time through the 5G-A network, and the target maintenance time and target maintenance plan are determined based on the distribution of maintenance resources and the operating status data of other professional categories as the target maintenance decision; If the remote maintenance method is not adopted, the target maintenance time and the target maintenance plan are directly determined based on the distribution of the maintenance resources and the operating status data of other professional categories as the target maintenance decision.

7. The method according to any one of claims 1 to 6, characterized in that: The process of creating the target fault identification model includes: Obtain the training running status data set corresponding to each professional category; The training operation status data set is input into the initial fault identification model, and the structure and parameters of the initial fault identification model are iteratively trained using a hyperparameter search method until the output predicted fault matches the historical real fault, thereby obtaining the corresponding target fault identification model.

8. The method according to any one of claims 1 to 6, characterized in that: The 5G-A network adopts a combination of distributed micro base stations and macro base stations and is deployed in the operating environment of rail transit; the micro base stations are deployed in one of the following areas where personnel and equipment are concentrated: stations and vehicle depots; the macro base stations are deployed in the rail lines and are set at intervals.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the rail transit intelligent operation and maintenance method based on the 5G-A network as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the rail transit intelligent operation and maintenance method based on a 5G-A network as described in any one of claims 1 to 8 when executed.

Images