A method, device and electronic equipment for monitoring the service performance of multi-machine coordinated track

Abstract

The present invention provides a multi-machine collaborative track service performance monitoring method, device and electronic equipment, wherein the multi-machine collaborative track service performance monitoring method includes: obtaining the track detection results and track health information of the current track; the track detection results are collected by at least two pre-deployed robot clusters; based on the track detection results and the track health information, identifying the fault type of the current track, and predicting future diseases and damages of the current track; calling a pre-set performance judgment method, based on the fault type of the current track and future diseases and damages, determining the service performance of the current track, and obtaining a service performance monitoring result. Through the present invention, the evaluation and monitoring results of the service performance of the current track are more representative and more accurate, solving the problem of incomplete monitoring and evaluation in the existing related technologies.

Description

Technical Field

[0001] The present invention relates to the field of track monitoring technology, and in particular to a method, device and electronic equipment for monitoring the service performance of a multi-machine coordinated track. Background Art

[0002] With the development of economy and technology, rail systems have been widely used in people's production and life, bringing endless convenience to people's production and life. At the same time, the rail system has a large transportation volume and lower pollution than other public transportation. It can effectively improve the congestion in cities during peak hours, and play an important role in accelerating urban economic construction and improving the happiness index of citizens.

[0003] Traditional track inspection methods mainly rely on manual inspections or single inspection equipment, which is not only inefficient but also difficult to cover all potential safety hazards. With the rapid development of railway transportation and the increasing emphasis on safety, existing track inspection technologies face the following challenges: Manual inspections are time-consuming and difficult to meet the rapid inspection needs of large-scale railway networks; Insufficient inspection coverage: Single equipment can often only detect a certain type of problem and cannot conduct a comprehensive assessment; Limited data processing capabilities: Traditional data processing methods are difficult to effectively integrate complex data obtained by multiple inspection methods; Lack of effective data analysis algorithms makes it difficult to accurately predict future failure risks; Inaccurate maintenance decisions: Without the guidance of comprehensive evaluation methods, maintenance decisions are not scientific and reasonable.

[0004] There is currently no effective solution to the problem of incomplete monitoring and evaluation in existing related technologies. Summary of the invention

[0005] The present invention provides a method, device and electronic equipment for monitoring the service performance of a multi-machine coordinated track, so as to solve the defect of incomplete monitoring and evaluation in the prior art.

[0006] In a first aspect, the present invention provides a method for monitoring track service performance of multiple machines in coordination, comprising:

[0007] Obtaining track detection results and track health information of the current track; the track detection results are collected by at least two pre-deployed robot clusters;

[0008] Based on the track inspection result and the track health information, identifying the fault type of the current track and predicting future diseases and damages of the current track;

[0009] A pre-set performance determination method is called to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain a service performance monitoring result.

[0010] According to a multi-machine collaborative track service performance monitoring method provided by the present invention, the robot cluster includes a comprehensive detection robot and a special detection robot;

[0011] The comprehensive inspection robot is used to perform geometric shape inspection, surface and internal damage inspection and weld damage inspection on the current track;

[0012] The special inspection robot is used to perform fastener inspection and turnout structure limit inspection on the current track.

[0013] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, obtaining track health information of the current track includes:

[0014] Integrate a plurality of preset monitoring means to monitor the current track and obtain track health information of the current track;

[0015] The detection means include track circuit monitoring, traction return current monitoring, optical fiber monitoring, stress monitoring and rail electrical signal carrier monitoring.

[0016] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, based on the track detection result and the track health information, the fault type of the current track is identified, including:

[0017] Determining a mean and a standard deviation of the track detection results;

[0018] Based on the mean and standard deviation of the track detection results, it is judged whether there is a track fault on the current track, and the fault type of the track fault is determined.

[0019] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, based on the track detection result and the track health information, the fault type of the current track is identified, including:

[0020] The pre-trained deep learning model is called, and the track detection result of the current track is used as input to monitor the current track, determine whether there is a track fault on the current track, and determine the fault type of the track fault.

[0021] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, based on the track detection results and the track health information, predicting the future disease and damage of the current track, comprises:

[0022] A trend prediction model is called to predict future diseases and damages of the current track based on the track health information of the current track.

[0023] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, the trend prediction model is an autoregressive integrated moving average model; the track health information of the current track includes stationary time series information and non-stationary time series information;

[0024] Based on the track health information of the current track, predicting the future disease and damage of the current track, including:

[0025] The non-stationary time series information is subjected to differential processing to convert the non-stationary time series information into stationary time series information.

[0026] According to a multi-machine coordinated track service performance monitoring method provided by the present invention, the performance determination method includes at least one of threshold setting, scoring system determination, and early warning mechanism determination.

[0027] In a second aspect, the present invention further provides a multi-machine coordinated track service performance monitoring device, comprising:

[0028] An acquisition module, used to acquire the track detection results and track health information of the current track; the track detection results are collected by at least two pre-deployed robot clusters;

[0029] an identification module, configured to identify the fault type of the current track based on the track detection result and the track health information, and predict future diseases and damages of the current track;

[0030] The determination module is used to call a preset performance determination method to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain a service performance monitoring result.

[0031] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for monitoring the service performance of a multi-machine coordinated track as described in the first aspect above is implemented.

[0032] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the multi-machine collaborative track service performance monitoring method as described in the first aspect above.

[0033] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the multi-machine collaborative track service performance monitoring method as described in the first aspect above.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The multi-machine collaborative track service performance monitoring method provided by the present invention detects the current track by integrating a variety of different types of robot clusters and collects data, so that the final evaluation and monitoring results of the service performance of the current track are more representative and more accurate, solving the problem of incomplete monitoring and evaluation in existing related technologies. In addition, the present invention also uses an anomaly detection algorithm to identify whether there is a fault in the current track and the specific fault type, and a trend prediction algorithm to predict future track diseases and damage through time series analysis; finally, a service performance determination method is established, and the safety status of the track is comprehensively evaluated by setting thresholds, establishing a scoring system, setting up an early warning mechanism, and proposing maintenance suggestions, etc., to ensure that problems in the track can be discovered and solved in a timely manner, and to ensure the safety and reliability of railway transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a flow chart of the multi-machine cooperative track service performance monitoring method provided by the present invention;

[0038] Figure 2 It is a structural block diagram of the multi-machine cooperative track service performance monitoring device provided by the present invention;

[0039] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The present invention provides a method for monitoring the service performance of a multi-machine coordinated track. Figure 1 is a flow chart of the multi-machine coordinated track service performance monitoring method provided by the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0042] Step S101, obtaining the track detection result and track health information of the current track; the track detection result is collected by at least two robot clusters deployed in advance.

[0043] Step S102: Based on the track inspection results and track health information, identify the fault type of the current track and predict future diseases and damages of the current track.

[0044] Step S103, calling a preset performance determination method, based on the fault type and future diseases and damages of the current track, determining the service performance of the current track, and obtaining a service performance monitoring result.

[0045] In this method, for the track that needs to be monitored for service performance, the current track is first inspected by a pre-deployed robot cluster, and the track inspection results are collected to obtain the track health information of the current track. Among them, there are at least two types of robot clusters, which can monitor and inspect the current track more comprehensively, and help to evaluate the service performance of the current track more comprehensively. Then, based on the track inspection results and track health information, it is identified whether there is a fault in the current track. If there is a fault, the fault type of the current track is identified, and the diseases and damage that may occur in the current track in the future are predicted. Finally, the pre-set performance judgment method is called again, and the service performance of the current track is determined according to the above-mentioned test results. Since the present invention integrates a variety of different types of robot clusters to detect the current track and collect data, the final evaluation and monitoring results of the service performance of the current track are more representative and more accurate, which solves the problem of incomplete monitoring and evaluation in the existing related technologies.

[0046] In some embodiments, the robot cluster includes a comprehensive inspection robot and a special inspection robot. The comprehensive inspection robot is used to perform geometric shape inspection, surface and internal damage inspection, and weld damage inspection on the current track; the special inspection robot is used to perform fastener inspection and turnout structure clearance inspection on the current track.

[0047] In this embodiment, the comprehensive inspection robot integrates the functions of track geometry inspection, rail surface and internal damage inspection, weld damage inspection, etc., and uses a variety of technical means such as dynamic continuous measurement, magnetic flux leakage eddy current composite detection, and ultrasonic waves. The special inspection robot is responsible for fastener inspection and turnout structure limit measurement, and uses visual imaging technology and other applicable technologies for inspection.

[0048] Specifically, the comprehensive inspection robot includes the following subsystems: track geometry detection system, rail surface and internal damage detection system, and weld damage detection system. Among them, the track geometry detection system includes laser radar and optical sensors. Laser radar technology is used to perform continuous track geometry detection, such as gauge, level, height, etc. Optical sensors are used to detect the straightness and flatness of the current track. The rail surface and internal damage detection system is used to perform leakage magnetic eddy current composite detection and ultrasonic detection. Leakage magnetic eddy current composite detection combines leakage magnetic detection and eddy current detection technology to detect surface and near-surface defects, such as cracks, wear, etc. Ultrasonic detection uses ultrasonic probes to detect cracks, holes and other defects inside the rail. The weld damage detection system is used to perform infrared thermal imaging and ultrasonic testing. Infrared thermal imaging uses infrared thermal imaging technology to detect temperature changes in the weld area to identify possible damage. Ultrasonic detection uses ultrasonic detection technology for the weld area to identify defects inside the weld.

[0049] The steering inspection robot includes the following subsystems: a fastener inspection system and a turnout structure clearance inspection system. The fastener inspection system uses visual imaging technology for inspection, specifically using a high-definition camera to take photos of the fasteners and record the status of the fasteners. In addition, the fastener inspection system combines image processing algorithms for monitoring, specifically using image processing technology to analyze photos and identify missing, loose or damaged fasteners. The turnout structure clearance inspection system uses a three-dimensional laser scanner to obtain the precise dimensional information of the turnout structure, and combines structural analysis software to process the scanned data to evaluate whether the turnout structure meets safety standards.

[0050] In some of the embodiments, obtaining the track health information of the current track includes: integrating several preset monitoring means to monitor the current track to obtain the track health information of the current track; the detection means include track circuit monitoring, traction return current monitoring, optical fiber monitoring, stress monitoring and rail electrical signal carrier monitoring.

[0051] Specifically, track circuit monitoring mainly includes current monitoring and voltage monitoring. Current monitoring is used to monitor the current changes in the current track circuit to determine whether there is a circuit break or short circuit. Voltage monitoring is used to monitor the voltage changes in the current track circuit to ensure that the voltage is within the normal range. Traction return current monitoring mainly includes current sensor monitoring and temperature monitoring. Current sensor monitoring is used to monitor the current intensity in the traction return path to ensure smooth return. Temperature monitoring is used to monitor the temperature changes in the return path to prevent safety hazards caused by overheating. Fiber optic monitoring mainly includes temperature monitoring and strain monitoring. Temperature monitoring uses fiber optic sensing technology to monitor temperature changes around the track and identify potential thermal expansion and contraction problems. Strain monitoring is used to monitor the strain changes of the track and evaluate the structural integrity of the track. Stress monitoring mainly includes pressure sensor monitoring and strain gauge monitoring. Pressure sensor monitoring refers to installing pressure sensors at key positions of rails and sleepers to monitor stress distribution; strain gauge monitoring refers to using strain gauges to monitor the strain changes of the current track when a train passes. Rail electrical signal carrier monitoring mainly includes signal sensor monitoring and data acquisition system monitoring. Signal sensor monitoring refers to capturing changes in electrical signals through signal sensors installed on rails for diagnosing the health of tracks; data acquisition system monitoring refers to collecting signal data and sending it to the central processing unit for analysis through wireless transmission technology.

[0052] In some of the embodiments, an anomaly detection algorithm may be used to identify the fault type of the current track based on the track detection results and track health information. The anomaly detection algorithm includes a statistically based method, which is as follows: determining the mean and standard deviation of the track detection results; judging whether there is a track fault on the current track based on the mean and standard deviation of the track detection results, and determining the fault type of the track fault. This method can be mainly used to identify abnormal values ​​in data such as geometric shape detection, rail surface and internal damage detection, and weld damage detection of the current track.

[0053] For example, the mean and standard deviation of the track detection results are calculated, and data points that are outside the normal range (for example, the mean plus or minus 2 times the standard deviation) are considered abnormal. , the calculation formulas for the mean and standard deviation are as follows:

[0054]

[0055]

[0056] in, X A dataset representing track detection results, n Indicates the number of data in the dataset. μ represents the mean, i Indicates the data point number, xi Indicates i Data points of track detection results, represents the standard deviation. If a data point x i satisfy , then the data point x i is considered as an abnormal data point, where k is a pre-set threshold, usually 2 or 3.

[0057] In addition, the anomaly detection algorithm also includes a method based on deep learning, including: calling a pre-trained deep learning model, taking the track detection result of the current track as input, monitoring the current track, determining whether there is a track fault on the current track, and determining the fault type of the track fault. In this embodiment, the deep learning model incorporates an attention mechanism, and the method can be used to identify outliers in the measurement data of fastener detection and turnout structure limits.

[0058] For example, a convolutional neural network (CNN) is used in combination with an attention mechanism to extract features and classify them, identify whether there is a fault on the current track, and determine the type of fault. The specific formula is as follows:

[0059]

[0060] in, is the classification function, W 1 and W 2 represents the weight matrix, b 1 and b 2 is the bias term, Attention is the attention mechanism function, x Represents input data.

[0061] When using a deep learning model for detection, first collect the track detection results of fastener detection and turnout structure limits, then clean the collected data, remove noise, and standardize and normalize the data. Then input the trained convolutional neural network to obtain the fault identification results of the current track. For the training of the convolutional neural network, the detection data of the track with known fault types can be collected in advance as a training set and a verification set, and the convolutional neural network is trained with the training set, and then the convolutional neural network is verified, evaluated, and fine-tuned with the verification set.

[0062] In some of the embodiments, based on the track inspection results and the track health information, predicting the future disease and damage of the current track includes: calling a trend prediction model to predict the future disease and damage of the current track based on the track health information of the current track.

[0063] In this embodiment, the trend prediction model is the Autoregressive Integrated Moving Average Model (ARIMA), which is a method widely used for stationary time series prediction. In the present invention, it can be used to predict abnormal trends in data such as track circuit monitoring, traction return current monitoring, optical fiber monitoring, stress monitoring, and rail electric signal carrier monitoring. Specific prediction contents include: Track circuit monitoring: predicting abnormal trends in circuit status, such as the possibility of open circuit or short circuit; Traction return current monitoring: predicting abnormal trends in traction current circuits, such as changes in current intensity; Optical fiber monitoring: predicting abnormal trends in temperature and strain, and identifying potential thermal expansion and contraction problems; Stress monitoring: predicting the changing trend of stress distribution and evaluating possible fatigue damage in the future; Rail electric signal carrier monitoring: predicting the changing trend of electric signals and identifying possible signal attenuation or interruption.

[0064] The track health information of the current track includes stationary time series information and non-stationary time series information. Based on the track health information of the current track, the future disease and damage of the current track are predicted, including: performing differential processing on the non-stationary time series information, and converting the non-stationary time series information into stationary time series information.

[0065] The specific formula is as follows:

[0066]

[0067] in, is the moving average polynomial, is an autoregressive polynomial, Indicates at a point in time t The observed value of B represents the lag operator, represents a white noise sequence, d Indicates the number of times the difference is processed.

[0068] In some of the embodiments, the performance determination method includes at least one of threshold setting, scoring system determination, and early warning mechanism determination. The above determination method can be used to determine the safety status of the current track, ensure that problems existing in the current track can be discovered and solved in a timely manner, and ensure the safety and reliability of railway transportation.

[0069] Specifically, threshold setting is to set the normal range and warning threshold of each indicator based on historical data and expert experience. For each monitoring indicator, a normal range and warning threshold are set. The threshold is adjusted regularly based on newly collected data to adapt to changes in track status.

[0070] The scoring system is to give a score to each test item, and the comprehensive score reflects the overall health of the track. The scoring system includes two forms: single-item scoring and comprehensive scoring. Single-item scoring is to give a score to each test or monitoring data according to its degree of deviation from the normal range. Comprehensive scoring is to calculate the comprehensive score based on the scores of each individual item and the importance weight. Assume that there is Monitoring indicators, each of which is scored as follows: , the corresponding importance weight is , then the comprehensive score S The calculation formula is as follows:

[0071]

[0072] Among them, S represents the comprehensive score, Indicates i The importance weight of each monitoring indicator is Indicates i A single score for each monitoring indicator.

[0073] The early warning mechanism refers to triggering an early warning when the detected data exceeds the set threshold, prompting relevant personnel to take timely measures. Before the judgment, different early warning levels need to be divided according to the severity of the anomaly. When the monitoring data exceeds the threshold, an early warning notification is automatically sent to relevant personnel.

[0074] In addition, based on the above judgment results, specific maintenance suggestions or repair plans can be put forward, including preventive maintenance and corrective maintenance. Preventive maintenance refers to the formulation of preventive maintenance plans based on comprehensive scores and early warning information. Corrective maintenance refers to the proposal of specific repair measures for problems that have already occurred. Through the performance judgment method, possible faults or diseases of the track can be predicted, as follows:

[0075] Track geometry anomalies: When track geometry detection data deviates from the normal range, it may indicate anomalies in track gauge, level, height, etc., and requires adjustment or replacement.

[0076] Rail surface and internal damage: Through the rail surface damage detection data and internal damage detection data, it is possible to predict damage such as rail surface cracks, holes or internal cracks, which require grinding or replacement.

[0077] Weld damage: Through the weld damage detection data, the damage of the weld area can be predicted, and welding repair or replacement is required.

[0078] Missing or damaged fasteners: Fastener inspection data can be used to predict missing or damaged fasteners and the need for reinstallation or replacement.

[0079] Abnormal turnout structure limits: Through the turnout structure limit detection data, abnormal conditions of the turnout structure can be predicted, and adjustments or replacements are required.

[0080] In summary, the patent of this invention deploys two types of robot clusters and integrates multiple monitoring methods to collect comprehensive information about the health of the track to ensure the diversity and accuracy of the monitoring data. Then, an anomaly detection algorithm is used to identify whether there is a fault on the current track and the specific fault type, and a trend prediction algorithm is used to predict future track diseases and damage through time series analysis; finally, a service performance determination method is established to comprehensively judge the safety status of the track by setting thresholds, establishing a scoring system, setting up an early warning mechanism, and making maintenance recommendations, ensuring that problems in the track can be discovered and resolved in a timely manner, and ensuring the safety and reliability of railway transportation.

[0081] The present invention also provides a multi-machine cooperative track service performance monitoring device. The multi-machine cooperative track service performance monitoring device provided by the present invention is described below. The multi-machine cooperative track service performance monitoring device described below and the multi-machine cooperative track service performance monitoring method described above can be referenced to each other. Figure 2 is a structural block diagram of a multi-machine coordinated track service performance monitoring device provided by the present invention, such as Figure 2 As shown, the device comprises:

[0082] An acquisition module 201 is used to acquire the track detection result and track health information of the current track; the track detection result is collected by at least two robot clusters deployed in advance;

[0083] An identification module 202 is used to identify the fault type of the current track based on the track inspection results and track health information, and predict future diseases and damages of the current track;

[0084] The determination module 203 is used to call a preset performance determination method to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain the service performance monitoring result.

[0085] When the device is in use, for the track that needs to be monitored for service performance, the current track is first detected by a pre-deployed robot cluster. The acquisition module 201 collects the track detection results and obtains the track health information of the current track. Among them, there are at least two types of robot clusters, which can monitor and detect the current track more comprehensively, which helps to evaluate the service performance of the current track more comprehensively. Then, the identification module 202 identifies whether there is a fault in the current track based on the track detection results and track health information. If there is a fault, the fault type of the current track is identified, and the diseases and damage that may occur in the current track in the future are predicted. Finally, the judgment module 203 calls the pre-set performance judgment method again, and determines the service performance of the current track based on the above-mentioned detection results. Because the present invention integrates a variety of different types of robot clusters to detect the current track and collect data, the final evaluation and monitoring results of the service performance of the current track are more representative and more accurate, which solves the problem of incomplete monitoring and evaluation in the existing related technologies.

[0086] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 301, a communications interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communications interface 302 and the memory 303 communicate with each other through the communication bus 304. The processor 301 may call the logic instructions in the memory 303 to execute the multi-machine coordinated track service performance monitoring method, which includes:

[0087] Obtain the track inspection results and track health information of the current track; the track inspection results are collected by at least two pre-deployed robot clusters;

[0088] Based on the track inspection results and track health information, identify the fault type of the current track and predict the future disease and damage of the current track;

[0089] The preset performance determination method is called to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain the service performance monitoring results.

[0090] In addition, the logic instructions in the above-mentioned memory 303 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0091] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the multi-machine coordinated track service performance monitoring method provided by the above methods, the method includes:

[0092] Obtain the track inspection results and track health information of the current track; the track inspection results are collected by at least two pre-deployed robot clusters;

[0093] Based on the track inspection results and track health information, identify the fault type of the current track and predict the future disease and damage of the current track;

[0094] The preset performance determination method is called to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain the service performance monitoring results.

[0095] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for monitoring the service performance of a multi-machine coordinated track provided by the above methods is implemented, and the method comprises:

[0096] Obtain the track inspection results and track health information of the current track; the track inspection results are collected by at least two pre-deployed robot clusters;

[0097] Based on the track inspection results and track health information, identify the fault type of the current track and predict the future disease and damage of the current track;

[0098] The preset performance determination method is called to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain the service performance monitoring results.

[0099] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-machine coordinated track service performance monitoring method, characterized in that: include: Obtaining track detection results and track health information of the current track; the track detection results are collected by at least two pre-deployed robot clusters; Based on the track inspection result and the track health information, identifying the fault type of the current track and predicting future diseases and damages of the current track; Calling a preset performance determination method to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtaining a service performance monitoring result; The robot cluster includes comprehensive detection robots and special detection robots; The comprehensive inspection robot is used to perform geometric shape inspection, surface and internal damage inspection and weld damage inspection on the current track; The special inspection robot is used to perform fastener inspection and turnout structure clearance inspection on the current track; Comprehensive inspection robot Includes the following subsystems: Track geometry detection system, rail surface and internal damage detection system, weld damage detection system; The special inspection robot includes the following subsystems: fastener inspection system and turnout structure clearance inspection system; Get the track health information of the current track, including: Integrate preset monitoring means to monitor the current track and obtain track health information of the current track; The monitoring means include track circuit monitoring, traction return current monitoring, optical fiber monitoring, stress monitoring and rail electrical signal carrier monitoring; When identifying abnormal values ​​in the current track geometry detection, rail surface and internal damage detection, and weld damage detection data, a statistical method is used, as follows: Determining a mean and a standard deviation of the track detection results; Based on the mean and standard deviation of the track detection results, judging whether there is a track fault on the current track, and determining the fault type of the track fault; A deep learning-based approach was used to identify outliers in the measurement data for fastener inspection and turnout structure clearances, as follows: The pre-trained deep learning model is called, and the track detection result of the current track is used as input to monitor the current track, determine whether there is a track fault on the current track, and determine the fault type of the track fault. The deep learning model integrates the attention mechanism. The specific method is as follows: A convolutional neural network combined with an attention mechanism is used for feature extraction and classification to identify whether there is a fault on the current track and determine the type of fault. The specific formula is as follows: in, is the classification function, W1 and W2 represent weight matrices, b1 and b2 are bias terms, Attention is the attention mechanism function, and x represents input data; When using the deep learning model for detection, first collect the track detection results of fastener detection and turnout structure limits, then clean the collected data, remove noise, and standardize and normalize the data; then input the trained convolutional neural network to obtain the fault identification results of the current track; Based on the track inspection result and the track health information, predicting future diseases and damages of the current track, including: Invoking a trend prediction model to predict future diseases and damages of the current track based on the track health information of the current track; The trend prediction model is an autoregressive integrated moving average model; the orbit health information of the current orbit includes stationary time series information and non-stationary time series information; Based on the track health information of the current track, predicting the future disease and damage of the current track, including: The non-stationary time series information is subjected to differential processing to convert the non-stationary time series information into stationary time series information. The specific formula is as follows: in, is the moving average polynomial, is an autoregressive polynomial, represents the observed value at time point t, B represents the lag operator, represents the white noise sequence, d represents the number of differential processing; The performance determination method includes at least one of threshold setting, scoring system determination, and early warning mechanism determination.

2. A multi-machine coordinated track service performance monitoring device, used to implement the multi-machine coordinated track service performance monitoring method according to claim 1, characterized in that: include: The acquisition module is used to obtain the track detection results and track health information of the current track; The track detection results are collected by at least two robot clusters deployed in advance; an identification module, configured to identify the fault type of the current track based on the track detection result and the track health information, and predict future diseases and damages of the current track; The determination module is used to call a preset performance determination method to determine the service performance of the current track based on the fault type and future diseases and damages of the current track, and obtain a service performance monitoring result.

3. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the multi-machine collaborative track service performance monitoring method as described in claim 1 is implemented.

Images