High-speed railway infrastructure maintenance decision-making method and system based on knowledge engineering

Abstract

The invention discloses a high-speed railway infrastructure maintenance decision-making method and system based on knowledge engineering, a storage medium and electronic equipment. The method comprises the following steps: acquiring historical detection data of an infrastructure; a corresponding trend prediction algorithm is adopted to carry out trend prediction on different types of historical detection data, a corresponding trend prediction result is obtained, a maintenance task and a maintenance scheme are determined based on the trend prediction result, and the maintenance task comprises a corresponding disease type, a disease level and maintenance time; based on the maintenance task and the maintenance scheme, the maintenance process in the maintenance knowledge base is read, and a corresponding maintenance plan is generated to perform maintenance operation based on the maintenance process and the maintenance plan; and after the maintenance operation is completed, updating the maintenance ledger. According to the invention, intelligent maintenance decision and management combining state maintenance, prevention maintenance and periodic maintenance can be realized, the quality and efficiency of infrastructure maintenance are greatly improved, and fine distribution of maintenance resources is realized.

Description

Technical Field

[0001] The present application relates to the technical field of infrastructure maintenance, and in particular to a knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, system, storage medium and electronic equipment. Background Art

[0002] Infrastructure such as rails, bridges, tunnels, and roadbeds are the basis for train operation and are crucial to the safe operation of high-speed railways. With the rapid growth of my country's high-speed railway operating mileage, the demand for infrastructure maintenance and repair has increased significantly.

[0003] Traditional infrastructure maintenance and repair mainly adopts a maintenance strategy that combines condition-based maintenance and periodic maintenance. Periodic inspections detect defects, and condition-based maintenance performs repairs before failures occur by monitoring equipment status in real time, effectively improving maintenance efficiency and reducing maintenance costs. However, condition-based maintenance relies on accurate condition monitoring and fault prediction, and still has certain limitations in practical applications. For example, issues such as the accuracy of condition monitoring data, the reliability of fault prediction models, and the dynamic allocation of maintenance resources may affect the actual effectiveness of condition-based maintenance. Periodic maintenance, on the other hand, involves regular preventive maintenance based on the equipment's service life and maintenance experience. Although it can ensure the basic reliability of the equipment, it lacks specificity and can easily lead to over- or under-maintenance, resulting in increased maintenance costs and reduced equipment utilization. Summary of the Invention

[0004] The embodiments of the present application provide a high-speed railway infrastructure maintenance decision-making method, system, storage medium and electronic equipment based on knowledge engineering, which can improve the intelligence level of infrastructure maintenance, achieve optimal allocation of maintenance resources and effective control of maintenance costs.

[0005] The present application provides a knowledge engineering-based decision-making method for high-speed railway infrastructure maintenance, including: Obtain historical infrastructure inspection data; Use the corresponding trend prediction algorithm to perform trend prediction on different types of historical detection data to obtain the corresponding trend prediction results; Determine maintenance tasks and maintenance plans based on the trend prediction results, wherein the maintenance tasks include corresponding disease types, disease levels, and maintenance time; Based on the maintenance task and the maintenance plan, read the maintenance process in the maintenance knowledge base and generate a corresponding maintenance plan; Perform maintenance work based on the maintenance process and the maintenance plan; After the maintenance work is completed, the maintenance record is updated.

[0006] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, the historical inspection data includes track status data of the rails; The corresponding trend prediction algorithm is used to perform trend prediction on different types of historical detection data to obtain corresponding trend prediction results, including: Input the time series historical data corresponding to the orbital status data into the Prophet model to obtain the predicted future series data; Calculate the rail degradation rate based on the rail TQI value; Calculating a prediction deviation value based on the future sequence data and the corresponding actual detection value; The future sequence data, the predicted deviation value, and the degradation rate are input into a Transformer model for processing to obtain a temporal variation pattern of track smoothness.

[0007] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, the Transformer model includes a linear layer, a multi-head attention mechanism, a fully connected layer, an encoder, and a decoder; The processing of the Transformer model includes: After the input data is fed into the linear layer, it is passed through the multi-head attention mechanism to obtain the first feature; Inputting the first feature into a fully connected layer and sequentially performing nonlinear transformation, offset processing, activation processing, and nonlinear mapping to obtain a second feature; The second feature is input into the encoder for encoding, and then input into the decoder for decoding and prediction to obtain the change pattern of track smoothness over time.

[0008] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, the processing process in the fully connected layer is expressed by the following formula:

[0009] in, Input data to the previous layer of the fully connected layer; is the weight matrix of the first fully connected layer, which linearly transforms the input data; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation; ReLU is the rectified linear unit activation function, which performs nonlinear mapping on the input data, converting negative values ​​to 0 and keeping non-negative values ​​unchanged; is the weight matrix of the second fully connected layer, which is used to perform linear transformation on the result obtained after the ReLU function is activated; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation.

[0010] Furthermore, in the above-mentioned knowledge engineering-based maintenance decision-making method for high-speed railway infrastructure, the historical inspection data includes the coating corrosion and spalling area of ​​the bridge and the corresponding service time; The method further comprises: using a corresponding trend prediction algorithm to perform trend prediction on different types of historical detection data to obtain corresponding trend prediction results; A GM (1, 1) model is established based on the coating corrosion and spalling area of ​​the bridge and the corresponding service time; The coating life prediction formula is obtained based on the GM (1, 1) model; The time when the coating area reaches a preset layer area threshold is calculated based on the life prediction formula.

[0011] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, the process of predicting the coating life based on the coating corrosion and spalling area of ​​the bridge and the corresponding service time includes: make represents the corrosion area of ​​the surface coating. Assume that the original sequence of the coating corrosion area change is: , the corresponding time series is , the cumulative sequence of the original sequence is , and satisfies and ; Among them, the parameters a and u are obtained by the least squares method Sure; in,

[0012]

[0013] The solution of the equation is ; Sure The value of , and then we can restore the series, ; set up , This is the prediction formula for the change in coating corrosion area; Among them, k is the kth corrosion, a is the development coefficient, u is the gray action, A and B are constants; Since the GM(1,1) model is based on the equidistant detection data, it is assumed that the time corresponding to the first data of the equidistant sequence is , the equal time interval is , then at any moment , the coating life prediction formula can be obtained as .

[0014] Furthermore, in the above-mentioned knowledge engineering-based maintenance decision-making method for high-speed railway infrastructure, the step of reading the maintenance process in the maintenance knowledge base based on the maintenance task and the maintenance plan and generating a corresponding maintenance plan includes: Allocate different maintenance tasks to appropriate dates based on priority, time, equipment, and manpower consumption constraints; Based on the time loss of switching between different tasks, the daily maintenance task schedule is obtained, and the corresponding annual, monthly, weekly and daily plans are generated.

[0015] Furthermore, the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, wherein corresponding annual, monthly, weekly, and daily plans are generated by an intelligent decision-making algorithm based on mixed integer programming, includes: Determine the known variables: Simulation days d 、Type of machinery n and the total number of tasks t ; Task list: contains task date constraints, location, and time window , Use of machinery and the time from the workshop to the maintenance location ; Task association list: including predecessor tasks, successor tasks, and time window and machine usage; Model variables: Each skylight transport task selection ; The start time of each task ; The start and end nodes of the daily skylight selection ; Constraints: Constraint 1: Each day can only have one start and end time node:

[0016] in, is the start node or the end node; Constraint 2: The start and end time nodes of each day cannot be repeated nodes:

[0017] in, is the starting node, is the end node; Constraint 3: Flow conservation constraint:

[0018] in, Indicates whether the event after event j in day d is i; Constraint 4: The node is not accessible:

[0019] Constraint 5: The start time of the start node is the time from the workshop to the job location:

[0020] in, is the node start time, Indicates whether the first task after leaving the workshop on day d is i, is the time from workshop to task i, Represents a very large number. If the brackets are correct, it is 1, otherwise it is 0; Constraint 6: Time is a positive number:

[0021] Constraint 7: The direction of the edge determines the time sequence within the skylight:

[0022] in, 、 、 、 They represent the start time of task i, the duration of task i, the transportation time from task i to j, and the start time of task j respectively; Constraint 8: Each task must be completed within a specified timeframe:

[0023] in, Is the event after event i in day k j? Is the event after event j in day k i? Constraint 9: Each task cannot be completed multiple times:

[0024] Input the determined known variables into the model, solve the model based on the constraints, and obtain the corresponding annual, monthly, weekly, and daily plans.

[0025] Input the determined known variables into the model, solve the model based on the constraints, and obtain the corresponding annual, monthly, weekly, and daily plans.

[0026] Furthermore, in the above-mentioned knowledge engineering-based maintenance decision-making method for high-speed railway infrastructure, the maintenance records include a disease record, an early warning record, and a repair record; The disease record includes the disease type, location information, detection time, level information and processing time limit information; the early warning record includes statistical information on diseases that need to be processed in the near future; the maintenance record includes the maintenance plan, maintenance time, maintenance quality and maintenance completion status of the disease.

[0027] The present application also provides a knowledge engineering-based high-speed railway infrastructure maintenance decision-making system, including: The maintenance decision module is used to obtain historical inspection data of infrastructure and use corresponding trend prediction algorithms to perform trend prediction on different types of historical inspection data to obtain corresponding trend prediction results; Maintenance knowledge base, used to store maintenance processes, maintenance plans and all inspection data; The maintenance plan module is used to generate corresponding maintenance plans based on maintenance tasks and maintenance plans; The ledger management module is used to store disease ledgers, early warning ledgers and maintenance ledgers.

[0028] Furthermore, in the above-mentioned high-speed railway infrastructure maintenance decision-making system based on knowledge engineering, the maintenance knowledge base includes a maintenance assistant, a maintenance rule knowledge base and a maintenance process knowledge base; The maintenance assistant includes: A data collection unit, used to collect data from various data sources; The knowledge extraction unit is used to perform text parsing on the collected data and vectorize the parsed text fields into the knowledge base; The knowledge storage unit is used to store the extracted knowledge into the local knowledge base according to the predefined knowledge graph structure; The knowledge updating unit is used to regularly obtain the latest data from the data source and update the knowledge base content.

[0029] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making system, the maintenance assistant is further configured to receive user questions, provide intelligent answers, and extract knowledge, including: Receive user questions and vectorize the questions to obtain question vectors; Acquire knowledge text from the knowledge document and perform text parsing to obtain text fields, vectorize the text fields to obtain text vectors, and store the text vectors in a vector database; Based on the question vector, a query is performed in the vector database, and a Top K ranking analysis is performed through similarity comparison to obtain a question library; Input the question library into the deep learning model to obtain preliminary prediction results; The preliminary prediction results are input into the LLM large language model for intelligent answering and knowledge extraction.

[0030] Furthermore, in the above-mentioned high-speed railway infrastructure maintenance decision-making system based on knowledge engineering, the maintenance rule knowledge base is used to store the name information, description information, inspection cycle, level classification rules and disease processing time limit rules of infrastructure diseases.

[0031] Furthermore, in the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making system, the maintenance process knowledge base is used to store new technical solutions for status repair and preventive repair of infrastructure diseases, including: maintenance operation category, operation purpose, operation conditions, quality standards, maintenance process steps, personnel allocation, tool configuration, maintenance guidance videos and pictures.

[0032] An embodiment of the present application also provides a computer-readable storage medium, in which a plurality of instructions are stored. The instructions are suitable for being loaded by a processor to execute any of the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making methods.

[0033] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used for the steps in any of the above-mentioned knowledge engineering-based high-speed railway infrastructure maintenance decision-making methods.

[0034] This application provides a knowledge-based engineering-based maintenance decision-making method, system, storage medium, and electronic device for high-speed railway infrastructure. This application uses corresponding trend prediction algorithms to perform trend predictions on different types of historical detection data, obtains corresponding trend prediction results, and determines maintenance tasks and plans based on the trend prediction results. This application can implement intelligent maintenance decision-making and management that combines condition-based maintenance, preventive maintenance, and periodic maintenance, significantly improving the quality and efficiency of infrastructure maintenance and repair, and achieving refined allocation of maintenance resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The following detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings will make the technical solutions and other beneficial effects of the present application apparent.

[0036] Figure 1 A flowchart of a high-speed railway infrastructure maintenance decision-making method based on knowledge engineering provided in an embodiment of the present application.

[0037] Figure 2 A flowchart of track smoothness prediction provided in an embodiment of the present application.

[0038] Figure 3 A schematic diagram of the structure of the Transformer model provided in the embodiment of this application.

[0039] Figure 4 A schematic diagram of the structure of a high-speed railway infrastructure maintenance decision-making system based on knowledge engineering provided in an embodiment of the present application.

[0040] Figure 5 A flowchart of the maintenance assistant provided in an embodiment of the present application performing intelligent answering and knowledge extraction.

[0041] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0043] Embodiments of the present application provide a knowledge engineering-based high-speed railway infrastructure maintenance decision-making method, system, storage medium, and electronic device. The knowledge engineering-based high-speed railway infrastructure maintenance decision-making system provided in embodiments of the present application can be integrated into an electronic device, such as a terminal or server. The terminal can include a tablet computer, laptop computer, personal computer (PC), microprocessor box, or other device.

[0044] See also Figure 1 , Figure 1 A flowchart of a knowledge engineering-based high-speed railway infrastructure maintenance decision-making method provided in an embodiment of the present application, which is applied to electronic equipment, includes the following steps: S1, obtain historical detection data of infrastructure.

[0045] Among them, infrastructure includes tracks, bridges, tunnels, roadbeds, etc. Historical monitoring data include track status data of rails, coating corrosion and peeling area of ​​bridges and corresponding service time, deformation data and settlement data of tunnels, geometric deformation data of roadbeds, etc.

[0046] S2, using corresponding trend prediction algorithms to perform trend prediction on different types of historical detection data to obtain corresponding trend prediction results.

[0047] In one embodiment, the historical inspection data includes rail condition data. This track condition data, also known as monthly dynamic inspection data, includes track irregularity indicators (such as gauge, level, height, and track orientation), track quality index (TQI), and vehicle vibration response data.

[0048] Figure 2 The flow chart of track smoothness prediction provided in the embodiment of this application is as follows: Figure 2 As shown, step S2 includes the following steps: S21, input the time series historical data corresponding to the orbital status data into the Prophet model to obtain the predicted future series data.

[0049] Specifically, the orbital status data is divided into two parts according to time, one part is used as the time series historical data, and the other part is used as the time series future data. The time series historical data is input into the Prophet model to obtain the predicted future series data, and the time series future data is the actual detection value of the predicted future series data.

[0050] S22: Calculate the rail degradation rate based on the rail TQI value.

[0051] Specifically, the rail TQI value degradation rate Calculated using linear regression fitting.

[0052] S23, calculating a prediction deviation value based on the future sequence data and the corresponding actual detection value.

[0053] The specific calculation formula is:

[0054] in, is the actual detection value, For the predicted future sequence data, is the prediction deviation value, The larger the value, the worse the prediction effect of the Prophet model, and vice versa.

[0055] S24, input the future sequence data, predicted deviation value and degradation rate into the Transformer model for processing to obtain the change pattern of track smoothness over time.

[0056] Figure 3 A schematic diagram of the structure of the Transformer model provided in the embodiment of the present application is shown in FIG. Figure 3 As shown in the figure, the Transformer model includes a linear layer, a multi-head attention mechanism, a fully connected layer, an encoder, and a decoder.

[0057] The processing of the Transformer model includes: S241, after inputting the input data into the linear layer, the first feature is obtained through the multi-head attention mechanism.

[0058] S242: Input the first feature into a fully connected layer and perform nonlinear transformation, offset processing, activation processing, and nonlinear mapping in sequence to obtain a second feature.

[0059] The processing in the fully connected layer is expressed by the following formula:

[0060] in, Input data to the previous layer of the fully connected layer; is the weight matrix of the first fully connected layer, which linearly transforms the input data; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation; ReLU is the rectified linear unit activation function, which performs nonlinear mapping on the input data, converting negative values ​​to 0 and keeping non-negative values ​​unchanged; is the weight matrix of the second fully connected layer, which is used to perform linear transformation on the result obtained after the ReLU function is activated; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation.

[0061] S243 , input the second feature into the encoder for encoding, and then input it into the decoder for decoding and prediction, to obtain a temporal variation pattern of track smoothness.

[0062] In one embodiment, the historical inspection data includes the coating corrosion and spalling area of ​​the bridge and the corresponding service time.

[0063] Step S2 further includes the following steps: S25, establishing a GM(1,1) model based on the coating corrosion and spalling area of ​​the bridge and the corresponding service time.

[0064] S26, based on the GM (1, 1) model, the coating life prediction formula is obtained.

[0065] S27, calculating the time when the coating area reaches a preset layer area threshold based on the life prediction formula.

[0066] Specifically, let represents the corrosion area of ​​the surface coating. Assume that the original sequence of the coating corrosion area change is: , the corresponding time series is , the cumulative sequence of the original sequence is , and satisfies and ; Among them, the parameters a and u are obtained by the least squares method Sure; in,

[0067]

[0068] The solution of the equation is ; Sure The value of , and then we can restore the series, ; set up , This is the prediction formula for the change in coating corrosion area; Among them, k is the kth corrosion, a is the development coefficient, u is the gray action, A and B are constants; Since the GM(1,1) model is based on the equidistant detection data, it is assumed that the time corresponding to the first data of the equidistant sequence is , the equal time interval is , then at any moment , the coating life prediction formula can be obtained as The above formula can easily predict the time when the coating corrosion area reaches S%, which provides a reference for the maintenance of steel bridge anti-corrosion coatings.

[0069] This embodiment uses grey theory to establish a coating corrosion life prediction model. The detection data required by the prediction model is easy to obtain and the calculation is simple. It is very convenient to perform on-site life prediction of the anti-corrosion coating and provide a basis for maintenance decisions.

[0070] S3, based on the trend prediction results, determine the maintenance tasks and maintenance plans. The maintenance tasks include the corresponding disease type, disease level, and maintenance time.

[0071] S4, based on the maintenance task and maintenance plan, read the maintenance process in the maintenance knowledge base and generate a corresponding maintenance plan.

[0072] In one embodiment, step S4 includes: S41, assign different maintenance tasks to appropriate dates according to priority, time, equipment, and manpower consumption constraints; S42, based on the time loss of switching between different tasks, obtain the daily maintenance task schedule and generate corresponding annual, monthly, weekly and daily plans.

[0073] Specifically, determine the known variables: simulation days d 、Type of machinery n and the total number of tasks t ; Task list: contains task date constraints, location, and time window , Use of machinery and the time from the workshop to the maintenance location ; Task association list: including predecessor tasks, successor tasks, and time window and machine usage; Model variables: Each skylight transport task selection ; The start time of each task ; The start and end nodes of the daily skylight selection ; Constraints: Constraint 1: Each day can only have one start and end time node:

[0074] in, is the start node or the end node; Constraint 2: The start and end time nodes of each day cannot be repeated nodes:

[0075] in, is the starting node, is the end node; Constraint 3: Flow conservation constraint:

[0076] in, Indicates whether the event after event j in day d is i; Constraint 4: The node is not accessible:

[0077] Constraint 5: The start time of the start node is the time from the workshop to the job location:

[0078] in, is the node start time, Indicates whether the first task after leaving the workshop on day d is i, is the time from workshop to task i, Represents a very large number. If the brackets are correct, it is 1, otherwise it is 0; Constraint 6: Time is a positive number:

[0079] Constraint 7: The direction of the edge determines the time sequence within the skylight:

[0080] in, 、 、 、 They represent the start time of task i, the duration of task i, the transportation time from task i to j, and the start time of task j respectively; Constraint 8: Each task must be completed within a specified timeframe:

[0081] in, Is the event after event i in day k j? Is the event after event j in day k i? Constraint 9: Each task cannot be completed multiple times:

[0082] Input the determined known variables into the model, solve the model based on the constraints, and obtain the corresponding annual, monthly, weekly, and daily plans.

[0083] S5: Perform maintenance work based on the maintenance process and maintenance plan.

[0084] S6: After the maintenance work is completed, the maintenance record is updated.

[0085] Specifically, the disease record includes the disease type, location information, detection time, level information and processing time limit information, etc. The early warning record includes statistical information on diseases that need to be dealt with in the near future; the maintenance record includes the maintenance plan, maintenance time, maintenance quality and maintenance completion status of the disease.

[0086] According to the method described in the above embodiment, this embodiment will be further described from the perspective of a high-speed railway infrastructure maintenance decision-making system based on knowledge engineering. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering can be implemented as an independent entity or integrated into an electronic device, which can be a terminal, server or other device. The terminal may include a tablet computer, a laptop computer, a personal computer (PC), a micro processing box, or other devices.

[0087] See also Figure 4 , Figure 4 The present invention specifically describes a high-speed railway infrastructure maintenance decision-making system based on knowledge engineering provided in an embodiment of the present invention, which is applied to electronic equipment. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering may include: The maintenance decision module is used to obtain historical inspection data of infrastructure and use corresponding trend prediction algorithms to perform trend prediction on different types of historical inspection data to obtain corresponding trend prediction results; Maintenance knowledge base, used to store maintenance processes, maintenance plans and all inspection data; The maintenance plan module is used to generate corresponding maintenance plans based on maintenance tasks and maintenance plans; The ledger management module is used to store disease ledgers, early warning ledgers and maintenance ledgers.

[0088] In one embodiment, the maintenance knowledge base includes a maintenance assistant, a maintenance rule knowledge base, and a maintenance process knowledge base; Maintenance assistants include: Data collection unit, used to collect data from various data sources (such as input text files such as Word, Excel, PDF, web pages, databases, etc.); The knowledge extraction unit is used to perform text parsing on the collected data and vectorize the parsed text fields into the knowledge base; The knowledge storage unit is used to store the extracted knowledge into the local knowledge base according to the predefined knowledge graph structure; The knowledge update unit is used to regularly obtain the latest data from the data source and update the knowledge base content.

[0089] Figure 5 The flowchart of the maintenance assistant provided in the embodiment of the present application for intelligent answering and knowledge extraction is as follows: Figure 5 As shown, in one embodiment, the maintenance assistant is further configured to receive user questions and provide intelligent answers and knowledge extraction, including: Receive user questions and vectorize them to obtain question vectors; Obtain knowledge text from knowledge documents and perform text parsing to obtain text fields. Vectorize the text fields to obtain text vectors, and store the text vectors in a vector database. Based on the question vector, query the vector database and perform Top K ranking analysis by similarity comparison to obtain the question library; Input the question library into the deep learning model to obtain preliminary prediction results; The preliminary prediction results are input into the LLM large language model for intelligent answering and knowledge extraction.

[0090] As a specific embodiment, the maintenance assistant is a deepseek maintenance assistant, which is locally deployed based on deepseek technology.

[0091] In one embodiment, the maintenance rule knowledge base is used to store the name information, description information, inspection cycle, level classification rules and disease processing time limit rules of infrastructure diseases.

[0092] In one embodiment, the maintenance process knowledge base is used to store new technical solutions for status repair and preventive repair of infrastructure defects, including: maintenance operation category, operation purpose, operation conditions, quality standards, maintenance process steps, personnel configuration, tool configuration, maintenance guidance videos and pictures.

[0093] During specific implementation, the above modules and / or units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above modules and / or units can refer to the previous method embodiments. The specific beneficial effects that can be achieved can also be found in the beneficial effects in the previous method embodiments, which will not be repeated here.

[0094] In addition, embodiments of the present application further provide an electronic device, which may be a computer, tablet computer, or other device. This electronic device can implement the steps of any embodiment of the knowledge engineering-based high-speed railway infrastructure maintenance decision-making method provided in embodiments of the present application, and thus can achieve the beneficial effects achieved by any knowledge engineering-based high-speed railway infrastructure maintenance decision-making method provided in embodiments of the present application. For details, please refer to the previous embodiments and will not be repeated here.

[0095] Figure 6 This figure shows a block diagram of the specific structure of an electronic device provided in an embodiment of the present invention. This electronic device can be used to implement the knowledge engineering-based high-speed railway infrastructure maintenance decision-making method provided in the above-mentioned embodiment. The electronic device 500 can be a terminal, server, or other device. The terminal can include a tablet computer, laptop computer, personal computer (PC), microprocessor box, or other device.

[0096] RF circuit 510 is used to receive and transmit electromagnetic waves, converting them into electrical signals, thereby enabling communication with a communications network or other devices. RF circuit 510 may include various existing circuit components for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, memory, and the like. RF circuit 510 can communicate with various networks, such as the Internet, an intranet, or a wireless network, or with other devices via a wireless network. These wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The wireless networks may utilize various communication standards, protocols, and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE802.11g, and / or IEEE802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging, and short messaging, and any other suitable communication protocols, including those currently undeveloped.

[0097] The memory 520 can be used to store software programs and modules, such as the corresponding program instructions / modules in the above-mentioned embodiments. The processor 580 executes various functional applications and data processing by running the software programs and modules stored in the memory 520, that is, realizing functions such as taking pictures with the front camera, processing the captured images, and switching the display color of the displayed content on the display screen. The memory 520 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 520 may further include a memory remotely located relative to the processor 580, and these remote memories may be connected to the electronic device 500 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0098] The input unit 530 may be used to receive input digital or character information, and generate a keyboard and a mouse related to user settings and function control. The display unit 540 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces. These graphical user interfaces can be composed of graphics, text, icons, videos, or any combination thereof. The display unit 540 may include a display panel 541. Optionally, the display panel 541 can be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like.

[0099] Audio circuit 560, speaker 561, and microphone 562 provide an audio interface between the user and electronic device 500. Audio circuit 560 converts received audio data into electrical signals and transmits them to speaker 561, which then converts them into sound signals for output. Microphone 562, on the other hand, converts collected sound signals into electrical signals, which are then received by audio circuit 560 and converted into audio data. The audio data is then processed by output processor 580 and transmitted via RF circuit 510 to, for example, another terminal. Alternatively, the audio data may be output to memory 520 for further processing. Audio circuit 560 may also include an earphone jack to allow communication between external headphones and electronic device 500.

[0100] Electronic device 500, through a transmission module 570 (e.g., a Wi-Fi module), can help users receive requests, send information, and so on, providing users with wireless broadband Internet access. Although the figure shows transmission module 570, it is understood that it is not a required component of electronic device 500 and can be omitted as needed without changing the essence of the invention.

[0101] Processor 580 is the control center of electronic device 500. It connects all components of the phone using various interfaces and circuits. By running or executing software programs and / or modules stored in memory 520 and accessing data stored in memory 520, it executes various functions of electronic device 500 and processes data, thereby providing overall monitoring of the electronic device. Optionally, processor 580 may include one or more processing cores. In some embodiments, processor 580 may integrate an application processor and a modem processor. The application processor primarily handles the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 580.

[0102] Electronic device 500 also includes a power supply 590 (e.g., a battery) for powering various components. In some embodiments, the power supply can be logically connected to processor 580 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. Power supply 590 can also include any components, such as one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, and a power status indicator.

[0103] Although not shown, the electronic device 500 also includes a camera (such as a front camera and a rear camera), a Bluetooth module, etc., which will not be described in detail here. Specifically, in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors. The one or more programs include instructions for performing the following operations: Obtain historical infrastructure inspection data; Use the corresponding trend prediction algorithm to perform trend prediction on different types of historical detection data to obtain the corresponding trend prediction results; Determine maintenance tasks and maintenance plans based on the trend prediction results, wherein the maintenance tasks include corresponding disease types, disease levels, and maintenance time; Based on the maintenance task and the maintenance plan, read the maintenance process in the maintenance knowledge base and generate a corresponding maintenance plan; Perform maintenance work based on the maintenance process and the maintenance plan; After the maintenance work is completed, the maintenance record is updated.

[0104] In specific implementation, the above modules can be implemented as independent entities, or can be arbitrarily combined and implemented as the same or several entities. The specific implementation of the above modules can be found in the previous method embodiments and will not be repeated here.

[0105] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be accomplished through instructions, or by controlling related hardware through instructions. The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. To this end, an embodiment of the present invention provides a storage medium storing a plurality of instructions that can be loaded by a processor to execute the steps of any of the embodiments of the knowledge engineering-based high-speed railway infrastructure maintenance decision-making method provided in the embodiment of the present invention.

[0106] The computer-readable storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0107] Since the instructions stored in the storage medium can execute the steps in any embodiment of the high-speed railway infrastructure maintenance decision-making method based on knowledge engineering provided in the embodiments of the present invention, the beneficial effects that can be achieved by any high-speed railway infrastructure maintenance decision-making method based on knowledge engineering provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0108] The above is a detailed introduction to a high-speed railway infrastructure maintenance decision-making method, system, storage medium and electronic device based on knowledge engineering provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A high-speed railway infrastructure maintenance decision-making method based on knowledge engineering, characterized by: include: Obtain historical infrastructure inspection data; Use the corresponding trend prediction algorithm to perform trend prediction on different types of historical detection data to obtain the corresponding trend prediction results; Determine maintenance tasks and maintenance plans based on the trend prediction results, wherein the maintenance tasks include corresponding disease types, disease levels, and maintenance time; Based on the maintenance task and the maintenance plan, read the maintenance process in the maintenance knowledge base and generate a corresponding maintenance plan; Perform maintenance work based on the maintenance process and the maintenance plan; After the maintenance work is completed, the maintenance record is updated.

2. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 1 is characterized in that: The historical detection data includes track status data of the rails; The corresponding trend prediction algorithm is used to perform trend prediction on different types of historical detection data to obtain corresponding trend prediction results, including: Input the time series historical data corresponding to the orbital status data into the Prophet model to obtain the predicted future series data; Calculate the rail degradation rate based on the rail TQI value; Calculating a prediction deviation value based on the future sequence data and the corresponding actual detection value; The future sequence data, the predicted deviation value, and the degradation rate are input into a Transformer model for processing to obtain a temporal variation pattern of track smoothness.

3. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 2 is characterized in that: The Transformer model includes a linear layer, a multi-head attention mechanism, a fully connected layer, an encoder, and a decoder; The processing of the Transformer model includes: After the input data is fed into the linear layer, it is passed through the multi-head attention mechanism to obtain the first feature; Inputting the first feature into a fully connected layer and sequentially performing nonlinear transformation, offset processing, activation processing, and nonlinear mapping to obtain a second feature; The second feature is input into the encoder for encoding, and then input into the decoder for decoding and prediction to obtain the change pattern of track smoothness over time.

4. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 3 is characterized in that: The processing in the fully connected layer is expressed by the following formula: in, Input data to the previous layer of the fully connected layer; is the weight matrix of the first fully connected layer, which linearly transforms the input data; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation; ReLU is the rectified linear unit activation function, which performs nonlinear mapping on the input data, converting negative values ​​to 0 and keeping non-negative values ​​unchanged; is the weight matrix of the second fully connected layer, which is used to perform linear transformation on the result obtained after the ReLU function is activated; It is the bias term of the first fully connected layer, which offsets the result of the linear transformation.

5. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 1 is characterized in that: The historical inspection data includes the bridge coating corrosion and peeling area and the corresponding service time; The method further comprises: using a corresponding trend prediction algorithm to perform trend prediction on different types of historical detection data to obtain corresponding trend prediction results; A GM (1, 1) model is established based on the coating corrosion and spalling area of ​​the bridge and the corresponding service time; The coating life prediction formula is obtained based on the GM (1, 1) model; The time when the coating area reaches a preset layer area threshold is calculated based on the life prediction formula.

6. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 5 is characterized in that: The process of predicting the life of the coating based on the coating corrosion and spalling area of ​​the bridge and the corresponding service time includes: make represents the corrosion area of ​​the surface coating. Assume that the original sequence of the coating corrosion area change is: , the corresponding time series is , the cumulative sequence of the original sequence is , and satisfies and ; Among them, the parameters a and u are obtained by the least squares method Sure; in, The solution of the equation is ; Sure The value of , and then we can restore the series, ; set up , This is the prediction formula for the change in coating corrosion area; Among them, k is the kth corrosion, a is the development coefficient, u is the gray action, A and B are constants; Since the GM(1,1) model is based on the equidistant detection data, it is assumed that the time corresponding to the first data of the equidistant sequence is , the equal time interval is , then at any moment , the coating life prediction formula can be obtained as .

7. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 1 is characterized in that: The method of reading the maintenance process in the maintenance knowledge base based on the maintenance task and the maintenance plan and generating a corresponding maintenance plan includes: Allocate different maintenance tasks to appropriate dates based on priority, time, equipment, and manpower consumption constraints; Based on the time loss of switching between different tasks, the daily maintenance task schedule is obtained, and the corresponding annual, monthly, weekly and daily plans are generated.

8. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 7 is characterized in that: Generate corresponding annual, monthly, weekly, and daily plans through an intelligent decision-making algorithm based on mixed integer programming, including: Determine the known variables: Simulation days d 、Type of machinery n and the total number of tasks t ; Task list: contains task date constraints, location, and time window , Use of machinery and the time from the workshop to the maintenance location ; Task association list: including predecessor tasks, successor tasks, and time window and machine usage; Model variables: Each skylight transport task selection ; The start time of each task ; The start and end nodes of the daily skylight selection ; Constraints: Constraint 1: Each day can only have one start and end time node: in, is the start node or the end node; Constraint 2: The start and end time nodes of each day cannot be repeated nodes: in, is the starting node, is the end node; Constraint 3: Flow conservation constraint: in, Indicates whether the event after event j in day d is i; Constraint 4: The node is not accessible: Constraint 5: The start time of the start node is the time from the workshop to the job location: in, is the node start time, Indicates whether the first task after leaving the workshop on day d is i, is the time from workshop to task i, Represents a very large number. If the brackets are correct, it is 1, otherwise it is 0; Constraint 6: Time is a positive number: Constraint 7: The direction of the edge determines the time sequence within the skylight: in, 、 、 、 They represent the start time of task i, the duration of task i, the transportation time from task i to j, and the start time of task j respectively; Constraint 8: Each task must be completed within a specified timeframe: in, Is the event after event i in day k j? Is the event after event j in day k i? Constraint 9: Each task cannot be completed multiple times: Input the determined known variables into the model, solve the model based on the constraints, and obtain the corresponding annual, monthly, weekly, and daily plans.

9. The high-speed railway infrastructure maintenance decision-making method based on knowledge engineering according to claim 1 is characterized in that: The maintenance records include disease records, early warning records and repair records; The disease record includes the disease type, location information, detection time, level information and processing time limit information; the early warning record includes statistical information on diseases that need to be processed in the near future; the maintenance record includes the maintenance plan, maintenance time, maintenance quality and maintenance completion status of the disease.

10. A high-speed railway infrastructure maintenance decision-making system based on knowledge engineering, used to implement the high-speed railway infrastructure maintenance decision-making method based on knowledge engineering as claimed in any one of claims 1 to 9, characterized in that: include: The maintenance decision module is used to obtain historical inspection data of infrastructure and use corresponding trend prediction algorithms to perform trend prediction on different types of historical inspection data to obtain corresponding trend prediction results; Maintenance knowledge base, used to store maintenance processes, maintenance plans and all inspection data; The maintenance plan module is used to generate corresponding maintenance plans based on maintenance tasks and maintenance plans; The ledger management module is used to store disease ledgers, early warning ledgers and maintenance ledgers.

11. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering according to claim 1 is characterized in that: The maintenance knowledge base includes a maintenance assistant, a maintenance rule knowledge base and a maintenance process knowledge base; The maintenance assistants include: A data collection unit, used to collect data from various data sources; The knowledge extraction unit is used to perform text parsing on the collected data and vectorize the parsed text fields into the knowledge base; The knowledge storage unit is used to store the extracted knowledge into the local knowledge base according to the predefined knowledge graph structure; The knowledge updating unit is used to regularly obtain the latest data from the data source and update the knowledge base content.

12. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering according to claim 11 is characterized in that: The maintenance assistant is also used to receive user questions and provide intelligent answers and knowledge extraction, including: Receive user questions and vectorize the questions to obtain question vectors; Acquire knowledge text from the knowledge document and perform text parsing to obtain text fields, vectorize the text fields to obtain text vectors, and store the text vectors in a vector database; Based on the question vector, a query is performed in the vector database, and a Top K ranking analysis is performed through similarity comparison to obtain a question library; Input the question library into the deep learning model to obtain preliminary prediction results; The preliminary prediction results are input into the LLM large language model for intelligent answering and knowledge extraction.

13. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering according to claim 11 is characterized in that: The maintenance rule knowledge base is used to store the name information, description information, inspection cycle, level classification rules and disease treatment time limit rules of infrastructure diseases.

14. The high-speed railway infrastructure maintenance decision-making system based on knowledge engineering according to claim 11 is characterized in that: The maintenance process knowledge base is used to store new technical solutions for condition-based repair and preventive repair of infrastructure defects, including: maintenance operation categories, operation purposes, operation conditions, quality standards, maintenance process steps, personnel configuration, tool configuration, maintenance guidance videos and pictures.

15. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a plurality of instructions, which are suitable for being loaded by a processor to execute the high-speed railway infrastructure maintenance decision-making method based on knowledge engineering as described in any one of claims 1 to 9.

16. An electronic device, characterized in that: It includes a processor and a memory, the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used to execute the steps in the high-speed railway infrastructure maintenance decision-making method based on knowledge engineering as described in any one of claims 1 to 9.

Images