Railway tunnel operation and maintenance monitoring information intelligent processing and alarm system
By combining data acquisition, preprocessing, anomaly prediction, simulation, and early warning modules, and utilizing knowledge graphs and physical models, the problems of time-consuming and labor-intensive monitoring and delayed early warning in railway tunnels have been solved, enabling efficient and accurate identification and early warning of defects in railway tunnels.
Patent Information
- Application Number
- CN202511255038.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-04
- Publication Date
- 2025-11-14
AI Technical Summary
Traditional railway tunnel monitoring methods are time-consuming and labor-intensive, lack the ability to comprehensively analyze multi-dimensional data, resulting in delayed response of early warning systems and affecting the ability to respond quickly to emergencies.
The system employs data acquisition and preprocessing modules, anomaly prediction modules, simulation modules, and early warning modules, combined with knowledge graphs and physical models, to achieve multi-dimensional data analysis and early warning for railway tunnels.
It improves the efficiency and safety of tunnel operation and maintenance, can accurately identify defects and provide timely warnings, and reduce potential risks.
Smart Images
Figure CN120946408A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of railway operation and maintenance monitoring technology, and more specifically to an intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information. Background Technology
[0002] With the increasing number of railway tunnels, tunnel operation and maintenance monitoring has become increasingly important. Traditional monitoring methods mainly rely on manual inspections and simple monitoring equipment, which suffer from problems such as being time-consuming and labor-intensive, and having missed or false detections. At the same time, existing equipment lacks the ability to comprehensively analyze multi-dimensional data, cannot adapt to the processing needs of large-scale data, and lacks intelligent analysis tools, resulting in delayed response of early warning systems, which in turn affects the ability to respond quickly to emergencies.
[0003] Therefore, how to achieve multi-dimensional data analysis and early warning of railway tunnels, thereby improving tunnel safety and operation and maintenance efficiency, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0004] In view of this, the present invention provides an intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information, which can realize accurate identification and timely early warning of tunnel defects.
[0005] To achieve the above objectives, the present invention adopts the following technical solution:
[0006] An intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information includes: a data acquisition and preprocessing module, an anomaly prediction module, a simulation module, and an early warning module;
[0007] The data acquisition and preprocessing module is used to acquire various structural status data from different monitoring points in the tunnel in real time and perform preprocessing.
[0008] The anomaly prediction module is used to predict the structural state changes of each monitoring point in the future based on the monitoring data of each monitoring point in the current time period, and to identify abnormal monitoring points.
[0009] The simulation module is used to construct a physical model of the tunnel and associate the physical model with a knowledge graph; it uses the predicted structural state change data to simulate the anomaly monitoring point locations of the physical model, and combines the knowledge graph to infer the anomaly response data of the anomaly monitoring point.
[0010] The early warning module is used to classify and warn about the risk level of the structural health status based on the abnormal response data of the abnormal monitoring points.
[0011] Furthermore, the data acquisition and preprocessing module includes: a data acquisition unit and a data preprocessing unit;
[0012] The data acquisition unit is used to collect static monitoring data and / or dynamic monitoring data from various monitoring points within the tunnel; the static monitoring data is one or more of the following: tunnel structural design parameters, historical maintenance records, and geological exploration data; the dynamic monitoring data is one or more of the following: deformation data, strain data, environmental data, and traffic flow data.
[0013] The data preprocessing unit is used to perform data cleaning, data transformation, and data normalization on various monitoring data.
[0014] Furthermore, the anomaly prediction module includes: a data prediction unit and an anomaly identification unit;
[0015] The data prediction unit is used to analyze the distribution of data at each monitoring point using histogram or box plot methods, identify data skewness and outlier characteristics, and use statistical methods to compare real-time monitoring data with historical data to determine the changing trend of structural state data at each monitoring point; or to use a deep learning model to analyze the monitoring data of each monitoring point in the current period and predict the changes in the structural state of each monitoring point in the future.
[0016] The anomaly identification unit is used to identify abnormal monitoring points based on a preset risk threshold.
[0017] Furthermore, the method for the anomaly identification unit to identify anomaly monitoring points includes:
[0018] Based on the structural status data of each monitoring point in the tunnel over a future period, the risk coefficients for various structural status data are determined.
[0019] Based on the physical attributes of each monitoring point, a risk threshold is set for each monitoring point.
[0020] For each monitoring point, the risk coefficients of its various structural state data are weighted and summed to obtain the comprehensive risk coefficient of the monitoring point. If the comprehensive risk coefficient exceeds the risk threshold of the monitoring point, the monitoring point is regarded as an abnormal monitoring point.
[0021] Furthermore, the simulation module includes a numerical simulation unit and a knowledge reasoning unit;
[0022] The numerical simulation unit is used to construct a physical model of the tunnel based on its geometric features and material properties, and to simulate each anomaly monitoring point of the physical model using the finite element analysis method, thereby simulating the mechanical behavior of each anomaly monitoring point under the predicted structural state data.
[0023] The knowledge reasoning unit constructs a relationship network between various components of the tunnel in the form of a knowledge graph, and infers the mechanical behavior of each simulated abnormal monitoring point to obtain the abnormal propagation behavior of each abnormal monitoring point in the tunnel, and predicts whether the abnormal monitoring point is at risk of apparent defects or deformation.
[0024] Furthermore, the process of using knowledge graphs to infer the abnormal propagation behavior of each anomaly monitoring point in the tunnel includes:
[0025] The interaction relationships between various components in the tunnel under different mechanical behaviors, the interaction relationships between various components and risk factors, the maintenance measures corresponding to each risk factor, and the basic information of each component are represented by nodes and edges.
[0026] Each monitoring point is coded;
[0027] Using the code corresponding to the abnormal monitoring point as a retrieval field, knowledge is extracted from the knowledge graph to obtain the various components affected by the current mechanical behavior of the abnormal monitoring point, determine the abnormal propagation path, scope of influence and intensity, and provide potential risks and corresponding handling measures.
[0028] Furthermore, each node of the knowledge graph is stored in an encoded form, and the specific encoding methods include:
[0029] The first-level code is set as the code for the target tunnel. Based on the first-level code, the second-level code is set as the code for each monitoring point in the tunnel.
[0030] Based on the second-level coding, a third-level coding is added as the coding for the subordinate components of each monitoring point;
[0031] Based on the third-level coding, location code, item code, and attribute code are added as basic information for each component;
[0032] Based on the third-level coding, risk codes and measure codes are added as risk categories for each component and corresponding maintenance measures.
[0033] Furthermore, the early warning module includes: a risk level classification unit and an early warning unit;
[0034] The risk level classification unit is used to classify the risk level of each abnormal monitoring point according to the comprehensive risk coefficient calculated by the abnormal point identification unit.
[0035] The early warning unit is used to highlight the abnormal response data in a specific form at the corresponding position of the physical model based on the risk level of each abnormal monitoring point and the abnormal response data inferred by the simulation module.
[0036] Furthermore, the warning unit assigns a color scheme to each risk category, and displays the risk level of a specific risk category in descending order of darkness.
[0037] As can be seen from the above technical solution, compared with the prior art, the present invention has the following beneficial effects:
[0038] 1. This invention first performs an overall analysis of various structural data of each monitoring point in the tunnel during the current period to identify abnormal monitoring points that may pose risks in the future; then, it uses physical models and knowledge graphs to perform fixed-point simulation and anomaly reasoning on the abnormal monitoring points to achieve specific anomaly identification for each abnormal monitoring point; the whole process first performs a coarse screening of abnormal points, and then further subdivides the specific abnormal manifestations of the abnormal points to improve the data processing capability and accuracy.
[0039] 2. This invention uses predicted structural state data to simulate the location of abnormal monitoring points in the tunnel physical model, simulates the behavior of abnormal monitoring points in the physical environment, and combines the rich knowledge network and semantic information in the knowledge graph to determine the propagation path, scope and impact of abnormal situations. This allows for accurate early warning of the risk trend of abnormal monitoring points, facilitating the timely development of more targeted monitoring and maintenance strategies and effectively reducing the potential risks to the tunnel structure. Attached Figure Description
[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0041] Figure 1 This is an overall architecture diagram of the intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information provided by the present invention;
[0042] Figure 2 This is a schematic diagram of the intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information provided by the present invention. Detailed Implementation
[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0044] like Figures 1-2As shown in the figure, this invention discloses an intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information, including: a data acquisition and preprocessing module, an anomaly prediction module, a simulation module, and an early warning module;
[0045] The data acquisition and preprocessing module is used to collect various structural status data from different monitoring points inside the tunnel in real time and perform preprocessing.
[0046] The anomaly prediction module is used to predict the structural state changes of each monitoring point in the future based on the monitoring data of each monitoring point in the current time period, and to identify abnormal monitoring points.
[0047] The simulation module is used to construct a physical model of the tunnel and associate the physical model with a knowledge graph; it uses the predicted structural state change data to simulate the anomaly monitoring points in the physical model, and combines the knowledge graph to infer the anomaly response data of the anomaly monitoring points;
[0048] The early warning module is used to classify the risk level of the structural health status and issue early warnings based on the abnormal response data of abnormal monitoring points.
[0049] The following provides further explanation of each module.
[0050] The data acquisition and preprocessing module includes: a data acquisition unit and a data preprocessing unit;
[0051] The data acquisition unit is used to collect static and / or dynamic monitoring data from various monitoring points within the tunnel. Static monitoring data includes one or more of the following: tunnel structural design parameters, historical maintenance records, and geological survey data. Tunnel structural design parameters include the tunnel's geometric dimensions, material properties, and design loads. Historical maintenance records include information on past repairs, reinforcements, and component replacements, which helps in understanding the tunnel's aging trends and maintenance cycles. Geological survey data includes the geological structure, soil and rock properties, and groundwater level of the tunnel's location; this data is crucial for assessing the tunnel's stability and safety.
[0052] In addition to the above, static monitoring data can also include acceptance testing data, such as static acceptance testing data conducted after tunnel construction, including the geometric accuracy, surface condition, and material strength of the tunnel's main structure. Routine inspection data refers to static information such as tunnel surface condition, cracks, and water leakage obtained through manual inspections or simple monitoring equipment. Specialized inspection data refers to detailed inspections conducted on specific problems or locations, such as lining thickness, backfill voids, and foundation variations. Static measurement data includes inspection data of the track using tools or equipment such as track gauges, chord lines, and light track inspection instruments, including internal geometric parameters of track shape and position such as gauge, superelevation, level, alignment, elevation, versine, and torsion, as well as the condition of components such as rails, connecting parts, sleepers, anti-creep devices, ballast, and turnouts.
[0053] Dynamic monitoring data includes one or more of the following: deformation data, strain data, environmental data, and traffic flow data. Deformation data, obtained through sensors installed inside the tunnel, monitors the tunnel's deformation in real time, such as arch subsidence and sidewall displacement. This data reflects the real-time changes in the tunnel structure. Strain data measures the stress and strain distribution within the tunnel structure to assess its safety and stability; this includes measurements of concrete strain and steel reinforcement stress. Environmental data includes parameters such as temperature, humidity, and air quality within the tunnel; this data is crucial for assessing the comfort and safety of the working environment inside the tunnel. Traffic flow data, obtained through traffic flow monitoring equipment installed at the tunnel entrances and exits, records vehicle traffic in real time. This data helps analyze tunnel traffic conditions and predict future traffic demand.
[0054] The data preprocessing unit is used to perform data cleaning, data transformation, and data normalization on various monitoring data. Through data cleaning operations, missing values can be deleted, missing values can be filled, and outliers can be deleted. Through data transformation operations, various data can be converted into the same format, improving the efficiency of subsequent data processing. Through normalization operations, the data is unified to a certain numerical range.
[0055] The anomaly prediction module includes: a data prediction unit and an anomaly identification unit;
[0056] The data prediction unit is used to analyze the distribution of data at each monitoring point using histograms or box plots, identify data skewness and outliers, and compare real-time monitoring data with historical data using statistical methods to determine the changing trends of structural state data at each monitoring point. Alternatively, it can use a deep learning model to analyze the monitoring data of each monitoring point in the current time period and predict the structural state changes of each monitoring point in the future. The deep learning model can use a Long Short-Term Memory (LSTM) network, which can contain multiple input nodes and multiple output nodes depending on the prediction task. The deep learning model is pre-trained using past tunnel monitoring cases and fault data to learn the development trends of various structural data under various fault states, and then predicts the structural data changes of each monitoring point in the future based on the various structural data of the current time period.
[0057] The anomaly identification unit is used to identify anomaly monitoring points based on a preset risk threshold. Specifically, the method by which the anomaly identification unit identifies anomaly monitoring points includes:
[0058] Based on the structural status data of each monitoring point in the tunnel over a future period, the risk coefficients for various structural status data are determined.
[0059] Based on the physical properties of each monitoring point, risk thresholds are set for each monitoring point. Since the design parameters, materials, and geological environment of each monitoring point may be different, the risk index of various anomalies will also be different. Therefore, in order to make a more scientific assessment, different risk thresholds are set for each monitoring point.
[0060] For each monitoring point, the risk coefficients of its various structural state data are weighted and summed to obtain the comprehensive risk coefficient of the monitoring point. If the comprehensive risk coefficient exceeds the risk threshold of the monitoring point, the monitoring point is regarded as an abnormal monitoring point.
[0061] At this point, the anomaly prediction module has made a preliminary prediction of the structural status data of different monitoring areas of the entire tunnel and made a preliminary assessment of whether they are abnormal. In order to more accurately assess the development trend and potential impact of anomalies, it is necessary to use physical models and knowledge graphs for collaborative analysis.
[0062] Specifically, the simulation module includes a numerical simulation unit and a knowledge reasoning unit;
[0063] The numerical simulation unit is used to construct a physical model of the tunnel based on its geometric features and material properties, and to simulate each anomaly monitoring point of the physical model using the finite element analysis method, simulating the mechanical behavior of each anomaly monitoring point under the predicted structural state data, such as stress distribution.
[0064] The knowledge reasoning unit constructs a relationship network between various components of the tunnel in the form of a knowledge graph, and infers the mechanical behavior of each simulated abnormal monitoring point to obtain the abnormal propagation behavior of each abnormal monitoring point in the tunnel, and predicts whether the abnormal monitoring point is at risk of apparent defects or deformation. The apparent defects include water leakage, spalling, cracks, etc., and the deformation risks include settlement, lateral convergence deformation, etc.
[0065] The process of using knowledge graphs to infer the abnormal propagation behavior of each anomaly monitoring point in the tunnel includes:
[0066] The interaction relationships between various components in the tunnel under different mechanical behaviors, the interaction relationships between various components and risk factors, the maintenance measures corresponding to each risk factor, and the basic information of each component are represented by nodes and edges.
[0067] Each monitoring point is coded;
[0068] Using the code corresponding to the abnormal monitoring point as a retrieval field, knowledge is extracted from the knowledge graph to obtain the various components affected by the current mechanical behavior of the abnormal monitoring point, determine the abnormal propagation path, scope of influence and intensity, and provide potential risks and corresponding handling measures.
[0069] Specifically, to improve the standardization and uniformity of knowledge graphs, each node of the knowledge graph is stored in an encoded form. The specific encoding methods include:
[0070] The first-level code is set as the code for the target tunnel. Based on the first-level code, the second-level code is set as the code for each monitoring point in the tunnel.
[0071] Based on the second-level coding, a third-level coding is added as the coding for the subordinate components of each monitoring point;
[0072] Based on the third-level coding, location code, project code, and attribute code are added as basic information for each component; the location code contains the location information of the tunnel component; the project code contains the tunnel project code information, the location of the tunnel construction, and the tunnel type information; the attribute code contains the component material information, component size information, and material model information.
[0073] Based on the third-level coding, risk codes and measure codes are added as risk categories and corresponding maintenance measures for each component. The risk code contains information on various potential risks, and the measure code contains various maintenance measures corresponding to various risks.
[0074] For example, for a specific tunnel, its name's initials (in pinyin) are used as the first-level code. The monitoring points are then coded using numbers from 001 to 111 as the second-level code. The letters A, Z, and Z are used to code the subordinate components of each monitoring point. Location codes, project codes, attribute codes, risk codes, and measure codes are also coded in specific formats, with each code corresponding to a specific type of information. This coding of various information from the monitoring points facilitates the knowledge extraction process and improves the efficiency of anomaly detection.
[0075] This invention simulates and infers anomalies at monitoring points, and then visually displays the anomalies through an early warning module. The early warning module includes a risk level classification unit and an early warning unit.
[0076] The risk level classification unit is used to classify the risk level of each anomaly monitoring point based on the comprehensive risk coefficient calculated by the anomaly identification unit.
[0077] The early warning unit is used to highlight specific information in the corresponding location of the physical model based on the risk level of each anomaly monitoring point and the anomaly response data inferred by the simulation module.
[0078] The early warning unit assigns a color to each risk category, displaying the risk level of a specific risk category in descending order of intensity. For example, if a crack risk is expected to occur at a certain anomaly monitoring point in the future, its affected area can be defined in the physical model. The crack risk is assigned to a red, orange, or yellow color scheme. When the risk level is high, the defined area is displayed in red; when the risk level is medium, the defined area is displayed in orange; and when the risk level is low, the defined area is displayed in yellow.
[0079] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.
[0080] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A smart processing and alarm system for railway tunnel operation and maintenance monitoring information, characterized in that, include: The system includes a data acquisition and preprocessing module, an anomaly prediction module, a simulation module, and an early warning module. The data acquisition and preprocessing module is used to acquire various structural status data from different monitoring points in the tunnel in real time and perform preprocessing. The anomaly prediction module is used to predict the structural state changes of each monitoring point in the future based on the monitoring data of each monitoring point in the current time period, and to identify abnormal monitoring points. The simulation module is used to construct a physical model of the tunnel and associate the physical model with a knowledge graph; it uses the predicted structural state change data to simulate the anomaly monitoring point locations of the physical model, and combines the knowledge graph to infer the anomaly response data of the anomaly monitoring point; The early warning module is used to classify and warn about the risk level of the structural health status based on the abnormal response data of the abnormal monitoring points.
2. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 1, characterized in that, The data acquisition and preprocessing module includes: a data acquisition unit and a data preprocessing unit; The data acquisition unit is used to collect static monitoring data and / or dynamic monitoring data from various monitoring points within the tunnel; the static monitoring data is one or more of the following: tunnel structural design parameters, historical maintenance records, and geological survey data; the dynamic monitoring data is one or more of the following: deformation data, strain data, environmental data, and traffic flow data. The data preprocessing unit is used to perform data cleaning, data transformation, and data normalization on various monitoring data.
3. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 1, characterized in that, The anomaly prediction module includes: a data prediction unit and an anomaly identification unit; The data prediction unit is used to analyze the distribution of data at each monitoring point using histogram or box plot methods, identify data skewness and outlier characteristics, and use statistical methods to compare real-time monitoring data with historical data to determine the changing trend of structural state data at each monitoring point; or to use a deep learning model to analyze the monitoring data of each monitoring point in the current period and predict the changes in the structural state of each monitoring point in the future. The anomaly identification unit is used to identify abnormal monitoring points based on a preset risk threshold.
4. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 3, characterized in that, The method for the anomaly identification unit to identify anomaly monitoring points includes: Based on the structural status data of each monitoring point in the tunnel over a future period, the risk coefficients for various structural status data are determined. Based on the physical attributes of each monitoring point, a risk threshold is set for each monitoring point. For each monitoring point, the risk coefficients of its various structural state data are weighted and summed to obtain the comprehensive risk coefficient of the monitoring point. If the comprehensive risk coefficient exceeds the risk threshold of the monitoring point, the monitoring point is regarded as an abnormal monitoring point.
5. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 1, characterized in that, The simulation module includes a numerical simulation unit and a knowledge reasoning unit; The numerical simulation unit is used to construct a physical model of the tunnel based on its geometric features and material properties, and to simulate each anomaly monitoring point of the physical model using the finite element analysis method, thereby simulating the mechanical behavior of each anomaly monitoring point under the predicted structural state data. The knowledge reasoning unit constructs a relationship network between various components of the tunnel in the form of a knowledge graph, and infers the mechanical behavior of each simulated abnormal monitoring point to obtain the abnormal propagation behavior of each abnormal monitoring point in the tunnel, and predicts whether the abnormal monitoring point is at risk of apparent defects or deformation.
6. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 5, characterized in that, The process of using knowledge graphs to infer the abnormal propagation behavior of each anomaly monitoring point in the tunnel includes: The interaction relationships between various components in the tunnel under different mechanical behaviors, the interaction relationships between various components and risk factors, the maintenance measures corresponding to each risk factor, and the basic information of each component are represented by nodes and edges. Each monitoring point is coded; Using the code corresponding to the abnormal monitoring point as a retrieval field, knowledge is extracted from the knowledge graph to obtain the various components affected by the current mechanical behavior of the abnormal monitoring point, determine the abnormal propagation path, scope of influence and intensity, and provide potential risks and corresponding handling measures.
7. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 6, characterized in that, Each node of the knowledge graph is stored in an encoded form, and the specific encoding methods include: The first-level code is set as the code for the target tunnel. Based on the first-level code, the second-level code is set as the code for each monitoring point in the tunnel. Based on the second-level coding, a third-level coding is added as the coding for the subordinate components of each monitoring point; Based on the third-level coding, location code, item code, and attribute code are added as basic information for each component; Based on the third-level coding, risk codes and measure codes are added as risk categories for each component and corresponding maintenance measures.
8. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 4, characterized in that, The early warning module includes: a risk level classification unit and an early warning unit; The risk level classification unit is used to classify the risk level of each abnormal monitoring point according to the comprehensive risk coefficient calculated by the abnormal point identification unit. The early warning unit is used to highlight the abnormal response data in a specific form at the corresponding position of the physical model based on the risk level of each abnormal monitoring point and the abnormal response data inferred by the simulation module.
9. The intelligent processing and alarm system for railway tunnel operation and maintenance monitoring information according to claim 8, characterized in that, The warning unit assigns a color scheme to each risk category and displays the risk level of a specific risk category in descending order of darkness.