Rail transit equipment operation and maintenance management system and method based on cloud computing

Abstract

The invention provides a rail transit equipment operation and maintenance management system and method based on cloud computing, relates to the field of intelligent traffic management, and adopts a data analysis technology based on artificial intelligence. Image data, shot by the inspection robot, of the vehicle bottom of the rail running vehicle are analyzed to obtain key component characteristics of the vehicle bottom of the rail running vehicle, and then the key component characteristics of the vehicle bottom of the rail running vehicle are input into the generator to generate a detailed vehicle bottom inspection report. Therefore, the automation and intelligence level of rail transit equipment inspection is improved, and the problems of high labor intensity and low efficiency in manual inspection are effectively solved.

Description

Technical Field

[0001] The present application relates to the field of intelligent traffic management, and more specifically, to a cloud computing-based rail transit equipment operation and maintenance management system and method. Background Art

[0002] Failures in key parts of rail transit vehicles can lead to serious safety accidents. Therefore, it is necessary to inspect key maintenance points on the undercarriage of rail transit equipment to promptly discover potential safety problems, prevent accidents, and ensure the safety of passengers and operations. Currently, many locations still use manual inspections, which are labor-intensive and costly. In addition, inspections are usually conducted late at night when the human body's biological clock is in a resting state. The work efficiency and judgment of inspectors may be affected, resulting in poor inspection quality.

[0003] Therefore, a cloud computing-based rail transit equipment operation and maintenance management solution is needed. Summary of the invention

[0004] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a rail transit equipment operation and maintenance management system and method based on cloud computing, which adopts artificial intelligence-based data analysis technology, analyzes the image data of the bottom of the rail running vehicle taken by the inspection robot to obtain the characteristics of the key components of the bottom of the rail running vehicle, and then inputs the characteristics of the key components of the bottom of the running vehicle into the generator to generate a detailed bottom vehicle inspection report. In this way, the automation and intelligence level of rail transit equipment inspection is improved, and the problems of high labor intensity and low efficiency in manual inspection are effectively solved.

[0005] According to one aspect of the present application, a cloud computing-based rail transit equipment operation and maintenance management system is provided, which includes:

[0006] The vehicle bottom image data acquisition module is used to obtain the image data of the bottom of the rail vehicle taken by the inspection robot;

[0007] A vehicle bottom image data encoding module, used for performing feature encoding on the image data of the bottom of the rail vehicle to obtain a state feature diagram of key components of the vehicle bottom;

[0008] A vehicle bottom image data feature enhancement module, used for performing feature enhancement on the vehicle bottom key component state feature map to obtain a vehicle bottom key component state enhanced feature map;

[0009] The inspection report generation module is used to obtain a detailed inspection report of the underbody based on the enhanced characteristic map of the key underbody component status.

[0010] According to another aspect of the present application, a cloud computing-based rail transit equipment operation and maintenance management method is provided, which includes:

[0011] Acquire image data of the bottom of the rail vehicle taken by the inspection robot;

[0012] Performing feature coding on the image data of the underbody of the rail vehicle to obtain a state feature diagram of key components of the underbody;

[0013] Performing feature enhancement on the vehicle bottom key component state characteristic diagram to obtain a vehicle bottom key component state enhanced characteristic diagram;

[0014] Based on the enhanced characteristic diagram of the status of key components of the vehicle bottom, a detailed inspection report of the vehicle bottom is obtained.

[0015] Compared with the prior art, the cloud computing-based rail transit equipment operation and maintenance management system and method provided by the present application adopts artificial intelligence-based data analysis technology to analyze the image data of the bottom of the rail running vehicle taken by the inspection robot to obtain the characteristics of the key components of the bottom of the rail running vehicle, and then input the characteristics of the key components of the bottom of the running vehicle into the generator to generate a detailed bottom vehicle inspection report. In this way, the automation and intelligence level of rail transit equipment inspection is improved, and the problems of high labor intensity and low efficiency in manual inspection are effectively solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 This is a system block diagram of a rail transit equipment operation and maintenance management system based on cloud computing according to an embodiment of the present application.

[0018] Figure 2 This is an architecture diagram of a rail transit equipment operation and maintenance management system based on cloud computing according to an embodiment of the present application.

[0019] Figure 3 It is a block diagram of the underbody image data encoding module in the cloud computing-based rail transit equipment operation and maintenance management system according to an embodiment of the present application.

[0020] Figure 4 It is a block diagram of an inspection report generation module in a cloud computing-based rail transit equipment operation and maintenance management system according to an embodiment of the present application.

[0021] Figure 5It is a flowchart of a rail transit equipment operation and maintenance management method based on cloud computing according to an embodiment of the present application.

[0022] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0024] During long-term high-speed operation of rail transit vehicles, wear, looseness or damage to key parts of the vehicle bottom is likely to cause serious safety accidents, posing a major threat to the safety of passengers and the operation order of the railway. Therefore, detailed inspections of key maintenance points on the bottom of rail transit equipment are the key link to prevent failures, eliminate hidden dangers and ensure operational safety.

[0025] However, many rail transit stations still rely on traditional manual inspection methods. This method is not only labor-intensive and requires a lot of human resources, but also costly. What is more noteworthy is that since inspections are often scheduled late at night or early in the morning to reduce the impact on daily operations, this is contrary to the natural law of the human body's biological clock. Late at night, the human body's reaction speed and concentration generally decrease, and the inspectors work in a fatigued state, which inevitably affects their work efficiency and judgment, which undoubtedly increases the risk of missed inspections and misjudgments and reduces the overall quality of inspections. Therefore, an optimized cloud computing-based rail transit equipment operation and maintenance management solution is expected.

[0026] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for the operation and maintenance of rail transit equipment.

[0027] Figure 1 This is a system block diagram of a rail transit equipment operation and maintenance management system based on cloud computing according to an embodiment of the present application. Figure 2 FIG. 1 is an architecture diagram of a rail transit equipment operation and maintenance management system based on cloud computing according to an embodiment of the present application. Figure 1 and Figure 2As shown, in the cloud computing-based rail transit equipment operation and maintenance management system 100, it includes: a vehicle bottom image data acquisition module 110, which is used to obtain the image data of the bottom of the rail vehicle taken by the inspection robot; a vehicle bottom image data encoding module 120, which is used to perform feature encoding on the image data of the bottom of the rail vehicle to obtain a vehicle bottom key component state feature map; a vehicle bottom image data feature enhancement module 130, which is used to perform feature enhancement on the vehicle bottom key component state feature map to obtain a vehicle bottom key component state enhanced feature map; an inspection report generation module 140, which is used to obtain a vehicle bottom detailed inspection report based on the vehicle bottom key component state enhanced feature map.

[0028] Specifically, in the technical solution of the present application, image data of the bottom of the rail vehicle taken by the inspection robot is obtained. It should be understood that the image data of the bottom of the rail vehicle directly shows the morphology and structure of each component of the bottom (such as bogies, wheels, brake systems, suspension systems, pipelines, cables, etc.), and the integrity and structural state of these components are the basis for evaluating whether they are operating normally. Moreover, the surface of the component in the image can show conditions such as wear, cracks, corrosion, oil stains, and foreign matter adhesion. Changes in these surface conditions often indicate a decrease in component performance or the existence of potential faults. It is worth noting that the color and texture information of the component are also important references. For example, some components change color due to overheating, or the texture changes due to wear, which may be a precursor to a fault. In general, the image data of the bottom of the rail vehicle is the basis and basis for generating a detailed inspection report of the bottom of the vehicle. By analyzing the information such as the component morphology and surface condition in the image, different types of abnormalities can be identified, and the severity of the abnormality can be preliminarily evaluated based on the information such as the degree of wear, crack size, and corrosion range of the component in the image.

[0029] Cloud computing is an Internet-based computing method and service model that allows users to obtain the required computing resources (including servers, storage, databases, software, etc.) and services through the network in an on-demand and easily scalable manner. Cloud computing has the advantages of high flexibility, strong scalability, and good cost-effectiveness. It can help users quickly build and deploy applications and improve business response speed and innovation capabilities. Cloud computing can provide strong support and guarantee for the implementation of the technical solution of this application. For example, the image data of the bottom of the rail vehicle taken by the inspection robot can be stored and processed through the cloud computing platform.

[0030] In an embodiment of the present application, a possible implementation method for obtaining image data of the bottom of a rail-running vehicle taken by a patrol robot may be: obtaining a patrol task on a cloud server, including patrol time, patrol route, patrol points, etc.; the patrol robot automatically starts according to the set task, moves to a designated position along the track, and after arriving at the designated position, the patrol robot starts the camera to photograph the bottom of the rail-running vehicle; the patrol robot transmits the photographed image data to the cloud server via a wireless network; the cloud server receives the image data transmitted by the patrol robot, and analyzes it using cloud computing capabilities.

[0031] Figure 3 FIG. 1 is a block diagram of a vehicle bottom image data encoding module in a cloud computing-based rail transit equipment operation and maintenance management system according to an embodiment of the present application. Figure 3 As shown, the vehicle bottom image data encoding module includes: a vehicle bottom image data preprocessing unit 121, which is used to perform image preprocessing on the image data of the bottom of the rail-running vehicle to obtain a preprocessed vehicle bottom image of the rail-running vehicle; and a vehicle bottom key component state feature generation and extraction unit 122, which is used to pass the preprocessed vehicle bottom image of the rail-running vehicle through a vehicle bottom image feature encoder to obtain the vehicle bottom key component state feature map.

[0032] Considering that the original image is easily affected by various factors such as lighting conditions, equipment jitter, stains, etc., these unfavorable factors will cause the quality of the collected underbody image of the rail-running vehicle to decrease. Therefore, in the technical solution of the present application, the image data of the underbody of the rail-running vehicle is preprocessed to obtain the underbody image of the rail-running vehicle after preprocessing. Through image preprocessing, the impact of these unfavorable factors can be reduced, and the clarity and recognizability of the underbody image of the rail vehicle can be improved. In comparison, high-quality images can more accurately reflect the real state of the underbody of the vehicle, thereby reducing misjudgments and missed judgments.

[0033] In an embodiment of the present application, an implementation method for performing image preprocessing on the image data of the bottom of the rail vehicle to obtain a preprocessed bottom image of the rail vehicle can be: a. grayscale preprocessing the image: converting the color bottom image of the rail vehicle into a grayscale image. It should be understood that the color image contains rich color information, but in many cases, the color information is not the focus of the analysis, and it will increase the complexity of the processing. Grayscale can remove the color information and retain the brightness information of the image. b. Perform noise reduction processing on the image: noise will reduce the image quality and affect the subsequent feature extraction and fault detection. The image is smoothed by filtering algorithms (such as Gaussian filtering, median filtering, etc.) to reduce noise interference. c. Perform image enhancement preprocessing on the image: including contrast enhancement, sharpening and other operations. Contrast enhancement can increase the brightness difference between different objects in the image, making the image clearer; sharpening can enhance the edge information of the image and make the outline of the object clearer.

[0034] Next, the pre-processed underbody image of the track-running vehicle is passed through a underbody image feature encoder to obtain a state feature map of the key components of the underbody. It should be understood that the original underbody image data of the track-running vehicle usually contains a large amount of redundant information, and these redundancies can be removed through feature extraction to retain key information useful for subsequent underbody abnormality analysis. Based on the extracted features, it can be used to identify the status of each key component of the underbody, such as whether there are abnormal conditions such as cracks, wear, looseness, etc. Specifically, the structure of the underbody image feature encoder in the present application is a convolutional neural network model using a deep and shallow feature fusion module. The shallow network of the convolutional neural network model can extract detailed information, which is crucial for identifying subtle anomalies (such as small cracks), while the deep network can extract more abstract and advanced features, which have stronger distinguishing capabilities for anomaly detection in complex scenarios. Fusion of deep and shallow features can combine the advantages of both and improve the accuracy and robustness of detection.

[0035] In an embodiment of the present application, a possible implementation method of passing the preprocessed underbody image of the rail-running vehicle through an underbody image feature encoder to obtain a state feature map of key components of the underbody may be: extracting a shallow feature map of key components of the underbody from the i-th layer of the underbody image feature encoder, the i-th layer being the first to sixth layers of the underbody image feature encoder; extracting a deep feature map of key components of the underbody from the j-th layer of the underbody image feature encoder, the ratio between the j-th layer and the i-th layer being greater than or equal to 5; and using the shallow-deep feature fusion module to fuse the shallow feature map of key components of the underbody and the deep feature map of key components of the underbody to obtain the state feature map of key components of the underbody.

[0036] Although the state feature map of the key components under the vehicle already contains rich information, there may still be certain limitations in directly using the feature map to generate a detailed inspection report under the vehicle. Considering that different areas and components in the undercar image have different importance for the evaluation of the overall state, directly using the state feature map may not fully highlight the features of these important areas, thereby affecting the accuracy of the detection result. Therefore, the state feature map of the key components under the vehicle is passed through the undercar image feature enhancer based on the spatial attention mechanism to obtain the enhanced state feature map of the key components under the vehicle. The undercar image feature enhancer based on the spatial attention mechanism in this application is a convolutional neural network model using the spatial attention mechanism. Those of ordinary skill in the art should know that the spatial attention mechanism can automatically learn and identify important areas in the undercar image, and enhance the features of these areas, which helps to reduce the impact of noise and interference on the detection results and improve the detection accuracy. In detail, by enhancing the features of important areas, the spatial attention mechanism can improve the expression ability of the overall features, so that the status of each component under the vehicle can be better described in the generated detailed inspection report under the vehicle. And because the spatial attention mechanism can automatically focus on important areas, it can reduce the calculation and processing of irrelevant areas, thereby improving the detection efficiency.

[0037] In an embodiment of the present application, a possible implementation method of passing the state feature map of the key components of the vehicle bottom through a vehicle bottom image feature enhancer based on a spatial attention mechanism to obtain an enhanced state feature map of the key components of the vehicle bottom may be: using each layer of the vehicle bottom image feature enhancer based on a spatial attention mechanism to perform the following on the input data in the forward pass of the layer: performing convolution processing based on a convolution kernel on the input data to obtain a convolution feature map; passing the convolution feature map through a spatial attention unit to obtain a spatial attention map; calculating the point-by-point multiplication of the convolution feature map and the spatial attention map to obtain a spatial attention feature map; inputting the spatial attention feature map into a nonlinear activation unit to obtain an activation feature map; wherein the input of the first layer of the vehicle bottom image feature enhancer based on the spatial attention mechanism is the state feature map of the key components of the vehicle bottom, and the output of the last layer of the vehicle bottom image feature enhancer based on the spatial attention mechanism is the enhanced state feature map of the key components of the vehicle bottom.

[0038] Figure 4 FIG. 1 is a block diagram of an inspection report generation module in a cloud computing-based rail transit equipment operation and maintenance management system according to an embodiment of the present application. Figure 4As shown, the inspection report generation module 140 includes: a vehicle bottom key component state enhancement feature map pooling unit 141, which is used to pool the various feature matrices of the vehicle bottom key component state enhancement feature map along the channel dimension to obtain a vehicle bottom key component state enhancement feature vector; a vehicle bottom key component state optimization unit 142, which is used to perform a reverse error analysis on the vehicle bottom key component state enhancement feature vector based on a fine-grained error metric to obtain an optimized vehicle bottom key component state enhancement feature vector; an optimized vehicle bottom key component state enhancement feature parsing unit 143, which is used to pass the optimized vehicle bottom key component state enhancement feature vector through a report generator to obtain a generation result, and the generation result is a detailed vehicle bottom inspection report.

[0039] Considering that the undercarriage key component state enhancement feature map contains a large amount of detailed information of the undercarriage key components, but not all information is crucial for state judgment, in order to extract the most representative features from the feature map and remove redundant information, the undercarriage key component state enhancement feature map is pooled along the channel dimension of each feature matrix to obtain the undercarriage key component state enhancement feature vector. The feature vector extracted by the pooling operation is more refined and effective, and can more accurately reflect the state information of the undercarriage key components, thereby improving the accuracy of rail equipment operation and maintenance management.

[0040] In particular, in the technical solution of the present application, it is considered that the image preprocessing step may not be able to completely remove all noise or anomalies, resulting in the subsequent feature extraction being affected. The underbody image feature encoder may not be able to fully capture the status information of all key components, especially when the component status changes are subtle or the encoder design is not precise enough. The underbody image feature enhancer based on the spatial attention mechanism may not be able to evenly focus on all key areas, resulting in some important features being ignored. The global pooling operation may lose some local feature information, especially when certain areas of the feature map contain details that are particularly important for diagnosis, which in turn causes the enhanced feature vector of the state of the underbody key components to have a class harmony offset relative to the predetermined class label of the generator. If the design of the report generator does not take into account all possible vehicle states, or its decision logic does not match the complexity of the actual vehicle state, it may cause a class harmony offset. In order to reduce the class harmony offset, in one embodiment of the present application, the enhanced feature vector of the state of the underbody key components is subjected to an inverse error analysis based on a fine-grained error metric.

[0041] Specifically, in an embodiment of the present application, a reverse error analysis based on a fine-grained error metric is performed on the state enhancement feature vector of the key components of the vehicle bottom to obtain an optimized state enhancement feature vector of the key components of the vehicle bottom, including: performing a probabilistic activation based on a Sigmoid function on the state enhancement feature vector of the key components of the vehicle bottom to obtain a probabilistic state enhancement feature vector of the key components of the vehicle bottom; inputting the state enhancement feature vector of the key components of the vehicle bottom into the report generator to obtain an initial class label probability value; calculating the absolute value of the difference between the feature value of each position in the probabilistic state enhancement feature vector of the key components of the vehicle bottom and the initial class label probability value to obtain a backward machine learning fine-grained error metric vector; and using the latter The machine learning fine-grained error metric vector is used as a key vector, the state enhancement feature vector of the key components of the vehicle bottom is used as a query vector and a value vector, and the key vector, the query vector and the value vector are input into a converter structure to obtain a backward learning type attention optimized state enhancement feature vector of the key components of the vehicle bottom; the ratio between the sum of the backward learning type machine learning fine-grained error metric vector and the length of the backward learning type machine learning fine-grained error metric vector is calculated, and the value of a natural exponential function with the ratio as a power is calculated as an optimization scaling factor; the product of the optimization scaling factor and the backward learning type attention optimized state enhancement feature vector of the key components of the vehicle bottom is calculated to obtain the optimized state enhancement feature vector of the key components of the vehicle bottom.

[0042] In an embodiment of the present application, another expression of performing inverse error analysis based on fine-grained error metrics on the state enhancement feature vector of the key components of the vehicle bottom to obtain the optimized state enhancement feature vector of the key components of the vehicle bottom can be: processing the state enhancement feature vector of the key components of the vehicle bottom with the following formula to obtain the optimized state enhancement feature vector of the key components of the vehicle bottom; wherein the formula is:

[0043]

[0044] Among them, V c represents the probabilistic enhanced feature vector of the key components of the vehicle bottom, P represents the probability value of the initial class label, Indicates subtraction by position, V e represents the fine-grained error metric vector for backward machine learning, v i represents the eigenvalue of the ith position of the backward machine learning fine-grained error metric vector, L represents the length of the backward machine learning fine-grained error metric vector, T represents the transpose of the vector, represents vector multiplication, Softmax represents the normalized exponential function, ⊙ represents point multiplication by position, V' c Represents the optimized enhanced feature vector of the key components of the vehicle bottom.

[0045] Based on this, in the technical solution of the present application, the state enhancement feature vector of the key components of the vehicle bottom is subjected to reverse error analysis based on fine-grained error measurement, which first implements probabilistic activation based on the Sigmoid function on the state enhancement feature vector of the key components of the vehicle bottom to obtain a probabilistic state enhancement feature vector of the key components of the vehicle bottom, and inputs the state enhancement feature vector of the key components of the vehicle bottom into the report generator to obtain the initial class label probability value. Then, the error between the probabilistic state enhancement feature vector of the key components of the vehicle bottom and the initial class label probability value is calculated through reverse error analysis to generate a fine-grained error measurement vector, which quantifies the viewpoint measurement contribution of each feature position to the generated error.

[0046] Then, the backward machine learning fine-grained error metric vector is used as a key vector, and the state enhancement feature vector of the key components of the vehicle bottom is used as a query vector and a value vector, and the key vector, the query vector and the value vector are input into a transformer structure to perform error metric-based attention-based backward machine learning on the state enhancement feature vector of the key components of the vehicle bottom through a Transformer-like architecture. In this way, the state enhancement feature vector of the key components of the vehicle bottom is subjected to a backward interpretability analysis so that the obtained backward learning-like attention-optimized state enhancement feature vector of the key components of the vehicle bottom has a backward learning sensitive preference relative to the predetermined class label of the report generator. And further, the ratio of the sum of the error metric vector and the vector length is calculated, and the optimization scaling factor obtained by calculating the natural exponential function value with this ratio as an exponent is used to perform backward scaling optimization on the state enhancement feature vector of the key components of the vehicle bottom by the backward learning-like attention-optimized state, so as to improve the sensitivity preference and robustness of the state enhancement feature vector of the key components of the vehicle bottom relative to the predetermined class label of the report generator, thereby improving the accuracy of the generated result.

[0047] Finally, the optimized enhanced feature vector of the state of the key components of the vehicle bottom is passed through the report generator to obtain the generated result, and the generated result is the detailed inspection report of the vehicle bottom. It should be understood that although the optimized enhanced feature vector of the state of the key components of the vehicle bottom contains rich state information, this information exists in the form of data in a high-dimensional space and is not easy for track operation and maintenance personnel to understand. Therefore, it is necessary to pass the optimized enhanced feature vector of the state of the key components of the vehicle bottom through the report generator. The report generator is a machine learning model that is used to learn the mode of generating a detailed inspection report of the vehicle bottom according to the input data based on the training data. The detailed inspection report of the vehicle bottom usually contains the test results of each key component of the vehicle bottom, including normal state, abnormal state (such as cracks, wear, looseness, etc.) and detailed information such as its location, size, severity, etc. The test results and analysis and evaluation in the report provide important decision-making basis for operation and maintenance managers, helping them to judge the equipment status, formulate maintenance plans, etc. Through the automated detection and report generation process, manual intervention and errors are reduced, and the efficiency and accuracy of operation and maintenance are improved.

[0048] In summary, the cloud computing-based rail transit equipment operation and maintenance management system 100 according to the embodiment of the present application is explained, which adopts artificial intelligence-based data analysis technology, analyzes the image data of the bottom of the rail running vehicle taken by the inspection robot to obtain the key component features of the bottom of the rail running vehicle, and then inputs the key component features of the bottom of the running vehicle into the generator to generate a detailed bottom vehicle inspection report. In this way, the automation and intelligence level of rail transit equipment inspection is improved, and the problems of high labor intensity and low efficiency in manual inspection are effectively solved.

[0049] As described above, the cloud-based rail transit equipment operation and maintenance management system 100 according to the embodiment of the present application can be implemented in various terminal devices, such as a server for cloud-based rail transit equipment operation and maintenance management. In one example, the cloud-based rail transit equipment operation and maintenance management system 100 according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the cloud-based rail transit equipment operation and maintenance management system 100 can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the cloud-based rail transit equipment operation and maintenance management system 100 can also be one of the many hardware modules of the terminal device.

[0050] Alternatively, in another example, the cloud computing-based rail transit equipment operation and maintenance management system 100 and the terminal device may also be separate devices, and the cloud computing-based rail transit equipment operation and maintenance management system 100 may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0051] Figure 5 FIG. 1 is a flow chart of a rail transit equipment operation and maintenance management method based on cloud computing according to an embodiment of the present application. Figure 5 As shown, in the rail transit equipment operation and maintenance management method based on cloud computing, it includes: S110, acquiring image data of the bottom of the rail vehicle taken by the inspection robot; S120, feature encoding the image data of the bottom of the rail vehicle to obtain a state characteristic map of key components of the bottom of the vehicle; S130, feature enhancing the state characteristic map of key components of the bottom of the vehicle to obtain an enhanced state characteristic map of key components of the bottom of the vehicle; S140, obtaining a detailed inspection report of the bottom of the vehicle based on the enhanced state characteristic map of key components of the bottom of the vehicle.

[0052] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned cloud computing-based rail transit equipment operation and maintenance management method have been referred to above. Figures 1 to 4 The description of the cloud computing-based rail transit equipment operation and maintenance management system has been introduced in detail, and therefore, its repeated description will be omitted.

[0053] In summary, the cloud computing-based rail transit equipment operation and maintenance management method according to the embodiment of the present application is explained, which adopts artificial intelligence-based data analysis technology, analyzes the image data of the bottom of the rail running vehicle taken by the inspection robot to obtain the key component features of the bottom of the rail running vehicle, and then inputs the key component features of the bottom of the running vehicle into the generator to generate a detailed bottom vehicle inspection report. In this way, the automation and intelligence level of rail transit equipment inspection is improved, and the problems of high labor intensity and low efficiency in manual inspection are effectively solved.

[0054] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application. Figure 6 1 is a block diagram of an electronic device according to an embodiment of the present application. For example, the electronic device 10 can be provided as a server or a terminal device. Figure 6 The electronic device 10 includes a processing component 11, which further includes one or more processors, and a memory resource represented by a memory 12 for storing instructions executable by the processing component 11, such as an application. The application stored in the memory 12 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 11 is configured to execute instructions to perform the above method.

[0055] The electronic device 10 may further include a power supply component 13 configured to perform power management of the electronic device 10, a wired or wireless network interface 14 configured to connect the electronic device 10 to a network, and an input / output (I / O) interface 15. The electronic device 10 may operate based on an operating system stored in the memory 12, such as Windows Server™, Mac OSX™, Unix™, Linux™, FreeBSD™ or the like.

[0056] In an embodiment of the present application, a computer-readable storage medium is also provided, such as a memory 12 including computer program instructions. The computer program instructions can be executed by a processing component 11 of an electronic device 10 to complete the above method.

[0057] The present application may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer program instructions for causing a processor to implement various aspects of the present application.

[0058] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0059] The computer program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer program instructions from the network and forwards the computer program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0060] The computer program instructions for performing the operation of the present application can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer program instructions can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA), the electronic circuit can execute the computer program instructions, thereby realizing various aspects of the present application.

[0061] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer program instructions.

[0062] These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0063] Computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0064] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details of the above application are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present application to the use of the above specific details to be implemented.

Claims

1. A rail transit equipment operation and maintenance management system based on cloud computing, characterized in that: include: The vehicle bottom image data acquisition module is used to obtain the image data of the bottom of the rail vehicle taken by the inspection robot; A vehicle bottom image data encoding module, used for performing feature encoding on the image data of the bottom of the rail vehicle to obtain a state feature diagram of key components of the vehicle bottom; A vehicle bottom image data feature enhancement module, used for performing feature enhancement on the vehicle bottom key component state feature map to obtain a vehicle bottom key component state enhanced feature map; The inspection report generation module is used to obtain a detailed inspection report of the underbody based on the enhanced characteristic map of the key underbody component status.

2. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 1 is characterized in that: The vehicle bottom image data encoding module comprises: A vehicle bottom image data preprocessing unit, used for performing image preprocessing on the image data of the bottom of the rail-running vehicle to obtain a preprocessed vehicle bottom image of the rail-running vehicle; The key component state feature generation and extraction unit is used to pass the pre-processed track running vehicle underbody image through the underbody image feature encoder to obtain the key component state feature image of the underbody.

3. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 2 is characterized in that: The vehicle bottom image data feature enhancement module is used to: pass the vehicle bottom key component state feature map through the vehicle bottom image feature enhancer based on the spatial attention mechanism to obtain the vehicle bottom key component state enhanced feature map.

4. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 3 is characterized in that: The vehicle bottom image feature encoder is a convolutional neural network model using a deep and shallow feature fusion module.

5. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 4 is characterized in that: The vehicle bottom image feature enhancer based on the spatial attention mechanism is a convolutional neural network model using the spatial attention mechanism.

6. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 5, characterized in that: The inspection report generating module comprises: A vehicle bottom key component state enhancement feature map pooling unit, used for pooling the vehicle bottom key component state enhancement feature map along the channel dimension to obtain a vehicle bottom key component state enhancement feature vector; A vehicle bottom key component state optimization unit, used for performing a reverse error analysis based on a fine-grained error metric on the vehicle bottom key component state enhancement feature vector to obtain an optimized vehicle bottom key component state enhancement feature vector; The optimized vehicle bottom key component state enhancement feature analysis unit is used to pass the optimized vehicle bottom key component state enhancement feature vector through a report generator to obtain a generation result, and the generation result is a detailed vehicle bottom inspection report.

7. The cloud computing-based rail transit equipment operation and maintenance management system according to claim 6, characterized in that: The underbody key component state optimization unit is used to: Probabilistically activating the state enhancement feature vector of the key components of the vehicle bottom based on the Sigmoid function to obtain a probabilistic state enhancement feature vector of the key components of the vehicle bottom; Inputting the enhanced feature vector of the key components of the vehicle bottom into the report generator to obtain an initial class label probability value; Calculating the absolute value of the difference between the feature value of each position in the probabilistic enhanced feature vector of the state of the key components of the underbody and the initial class label probability value to obtain a backward machine learning fine-grained error metric vector; Using the backward machine learning fine-grained error metric vector as a key vector and the vehicle bottom key component state enhancement feature vector as a query vector and a value vector, the key vector, the query vector and the value vector are input into a converter structure to obtain a backward learning class attention optimized vehicle bottom key component state enhancement feature vector; Calculating a ratio between a sum of the backward machine learning fine-grained error metric vector and a length of the backward machine learning fine-grained error metric vector, and calculating a natural exponential function value raised to a power of the ratio as an optimization scaling factor; The product of the optimized scaling factor and the backward learning type attention optimized vehicle bottom key component state enhancement feature vector is calculated to obtain the optimized vehicle bottom key component state enhancement feature vector.

8. A rail transit equipment operation and maintenance management method based on cloud computing, characterized in that: include: Acquire image data of the bottom of the rail vehicle taken by the inspection robot; Performing feature coding on the image data of the underbody of the rail vehicle to obtain a state feature diagram of key components of the underbody; Performing feature enhancement on the vehicle bottom key component state characteristic diagram to obtain a vehicle bottom key component state enhanced characteristic diagram; Based on the enhanced characteristic diagram of the status of key components of the vehicle bottom, a detailed inspection report of the vehicle bottom is obtained.

9. The cloud computing-based rail transit equipment operation and maintenance management method according to claim 8, characterized in that: Characteristic encoding is performed on the image data of the underbody of the rail vehicle to obtain a characteristic diagram of the state of key components of the underbody, including: Performing image preprocessing on the image data of the underside of the rail-running vehicle to obtain a preprocessed underside image of the rail-running vehicle; The pre-processed underbody image of the rail-running vehicle is passed through an underbody image feature encoder to obtain a state feature image of key underbody components.

10. The cloud computing-based rail transit equipment operation and maintenance management method according to claim 9, characterized in that: Based on the enhanced characteristic diagram of the key components of the underbody, a detailed underbody inspection report is obtained, including: Pooling the feature matrices of the vehicle bottom key component state enhancement feature map along the channel dimension to obtain a vehicle bottom key component state enhancement feature vector; Performing an inverse error analysis based on a fine-grained error metric on the state enhancement feature vector of the key component of the vehicle bottom to obtain an optimized state enhancement feature vector of the key component of the vehicle bottom; The optimized enhanced feature vector of the key components of the underbody is passed through a report generator to obtain a generated result, which is a detailed underbody inspection report.

Images