Bridge vibration detection method and system based on edge computing

Abstract

Embodiments of the present invention provide a bridge vibration detection method and system based on edge computing. Based on anomaly detection prediction data from the bridge vibration data to be learned, the learning error association information of the bridge vibration anomaly detection model is updated. While adopting a common loss function value used in related technologies, other learning error association information is considered and combined with the loss function value and learning error association information to calculate the learning error of the bridge vibration anomaly detection model. Because the learning error association information is designed to continuously reduce the learning error of the bridge vibration anomaly detection model, it can improve the model convergence speed of the bridge vibration anomaly detection model, thereby enhancing the reliability of bridge vibration detection.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge vibration detection, and in particular to a bridge vibration detection method and system based on edge computing. Background Art

[0002] Road bridge structures often experience vibrations due to dynamic loads such as moving vehicles, crowds, wind, and earthquakes. These vibrations can change with aging. Detecting vibration anomalies in bridge structures is a key analytical step in bridge stability analysis. Improving the reliability of bridge vibration detection is a pressing technical challenge. Summary of the Invention

[0003] To solve the above problems, the present invention provides a bridge vibration detection method based on edge computing, comprising:

[0004] Acquire target bridge vibration data;

[0005] Inputting the target bridge vibration data into a bridge vibration anomaly detection model, and obtaining anomaly detection data of the target bridge vibration data based on the bridge vibration anomaly detection model, wherein the bridge vibration anomaly detection model is implemented by running an edge computing node;

[0006] The training steps of the bridge vibration anomaly detection model include:

[0007] Loading the bridge vibration data to be learned into a bridge vibration anomaly detection model configured according to the AI model structure to obtain multimodal vibration time series characteristics of the bridge vibration data to be learned;

[0008] Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features;

[0009] Inputting the high-dimensional extended mapping feature into the prediction unit of the bridge vibration anomaly detection model to obtain anomaly detection prediction data of the bridge vibration data to be learned by the bridge vibration anomaly detection model;

[0010] updating learning error association information of the bridge vibration anomaly detection model; wherein the learning error association information is used to reduce the learning error of the bridge vibration anomaly detection model;

[0011] Calculating a learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information;

[0012] The bridge vibration anomaly detection model is updated by back propagation based on the learning error, and iterative learning is continued according to the updated bridge vibration anomaly detection model until the model converges, thereby obtaining the bridge vibration anomaly detection model.

[0013] In one possible implementation, the method further includes:

[0014] Determining parameter floating information of the bridge vibration anomaly detection model based on a learning error of the bridge vibration anomaly detection model in a previous round of training;

[0015] Calculate the spatial range of the high-dimensional extended feature space used in this round of training based on the parameter floating information and the spatial range of the high-dimensional extended feature space used in the previous round of training;

[0016] The learning error association information includes the spatial range learning error of the high-dimensional extended feature space, and the updating of the learning error association information of the bridge vibration anomaly detection model includes:

[0017] The spatial range learning error of the high-dimensional extended feature space is calculated based on the spatial range of the high-dimensional extended feature space in this round of training.

[0018] In a possible implementation manner, the learning error association information further includes a vibration threshold value; and the learning error association information for updating the bridge vibration anomaly detection model further includes:

[0019] Based on the matching degree between the bridge vibration data to be learned and each abnormality detection label determined by the abnormality detection prediction data of the bridge vibration data to be learned, the vibration threshold value of the bridge vibration data to be learned and the corresponding abnormality detection label is updated.

[0020] In a possible implementation manner, the learning error associated information includes a vibration threshold value;

[0021] The updating of the learning error association information of the bridge vibration anomaly detection model based on the anomaly detection prediction data of the bridge vibration data to be learned further includes:

[0022] Based on the matching degree between the bridge vibration data to be learned and each abnormality detection label determined by the abnormality detection prediction data of the bridge vibration data to be learned, the vibration threshold value of the bridge vibration data to be learned and the corresponding abnormality detection label is updated.

[0023] In a possible implementation, mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features includes:

[0024] According to the spatial range of the high-dimensional extended feature space of this round of training, the multimodal vibration time series features are mapped to the high-dimensional extended feature space to obtain high-dimensional extended mapping features.

[0025] In one possible implementation, calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes:

[0026] Calculating a first learning error value based on the abnormality detection prediction data, the actual abnormality detection data of the bridge vibration data to be learned, and the vibration threshold value;

[0027] A learning error of the bridge vibration anomaly detection model is obtained based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space.

[0028] In one possible implementation, calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes:

[0029] Calculating a first learning error value based on the abnormality detection prediction data and the actual abnormality detection data of the bridge vibration data to be learned;

[0030] Based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space, a learning error of the bridge vibration anomaly detection model is calculated.

[0031] In one possible implementation, calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes:

[0032] The learning error of the bridge vibration anomaly detection model is calculated based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the vibration threshold value.

[0033] In a possible implementation, the method further includes: before mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features, further including:

[0034] Performing L2 regularization transformation on the multimodal vibration time series features to obtain regularized transformed multimodal vibration time series features;

[0035] Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features includes:

[0036] Mapping the regularized transformed multi-modal vibration time series features into a high-dimensional extended feature space to obtain high-dimensional extended mapping features;

[0037] Before inputting the high-dimensional extended mapping feature into the prediction unit of the bridge vibration anomaly detection model to obtain anomaly detection prediction data of the bridge vibration anomaly detection model for the to-be-learned bridge vibration data, the method further includes:

[0038] Perform L2 regularization transformation on the prediction unit parameters of the bridge vibration anomaly detection model to obtain the prediction unit parameters after regularization transformation.

[0039] The present invention also provides a bridge vibration detection system based on edge computing, including a processor and a readable storage medium, wherein the readable storage medium stores a computer program. When the computer program is executed by the processor, the bridge vibration detection method based on edge computing described in any one of the above methods is implemented.

[0040] In summary, the edge computing-based bridge vibration detection method and system provided by the present invention updates the learning error association information of the bridge vibration anomaly detection model based on the anomaly detection prediction data of the bridge vibration data to be learned. While adopting the general loss function value used in related technologies, it also considers other learning error association information and combines the loss function value and learning error association information to calculate the learning error of the bridge vibration anomaly detection model. Because the learning error association information is used to continuously reduce the learning error of the bridge vibration anomaly detection model, it can improve the model convergence speed of the bridge vibration anomaly detection model, thereby improving the reliability of bridge vibration detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings are only certain embodiments of the present invention and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 It is a flow chart of the bridge vibration detection method based on edge computing provided by the present invention. DETAILED DESCRIPTION

[0043] Figure 1This is a flow chart of the bridge vibration detection method based on edge computing provided by the present invention. This method can be executed by a bridge vibration detection system based on edge computing, which is described in detail below.

[0044] Step S11: Acquire target bridge vibration data.

[0045] Step S12: input the target bridge vibration data into a bridge vibration anomaly detection model, and obtain anomaly detection data of the target bridge vibration data based on the bridge vibration anomaly detection model, wherein the bridge vibration anomaly detection model is implemented by running an edge computing node.

[0046] Among them, for some alternative implementation methods, the training steps of the bridge vibration anomaly detection model may include the following steps, which are specifically introduced as follows.

[0047] Step S101: Load the bridge vibration data to be learned into a bridge vibration anomaly detection model configured according to the AI model structure to obtain multimodal vibration time series characteristics of the bridge vibration data to be learned.

[0048] Step S102 : Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features.

[0049] Step S103 : inputting the high-dimensional extended mapping features into the prediction unit of the bridge vibration anomaly detection model to obtain anomaly detection prediction data of the bridge vibration anomaly detection model for the to-be-learned bridge vibration data.

[0050] Step S104: updating the learning error association information of the bridge vibration anomaly detection model. The learning error association information is used to reduce the learning error of the bridge vibration anomaly detection model.

[0051] Step S105 : calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information.

[0052] Step S106 , performing back propagation based on the learning error to update the bridge vibration anomaly detection model, and continuing iterative learning according to the updated bridge vibration anomaly detection model until the model converges, thereby obtaining the bridge vibration anomaly detection model.

[0053] Based on the above steps, the embodiment of the present application updates the learning error association information of the bridge vibration anomaly detection model based on the anomaly detection prediction data of the bridge vibration data to be learned. While adopting the general loss function value used in related technologies, other learning error association information is considered, and the learning error of the bridge vibration anomaly detection model is calculated by combining the loss function value and the learning error association information. Because the learning error association information is used to continuously reduce the learning error of the bridge vibration anomaly detection model, it can improve the model convergence speed of the bridge vibration anomaly detection model, thereby improving the reliability of bridge vibration detection.

[0054] Among them, for some alternative implementation methods, based on the above embodiments, this embodiment can further determine the parameter floating information of the bridge vibration anomaly detection model based on the learning error of the bridge vibration anomaly detection model in the previous round of training, and calculate the spatial range of the high-dimensional extended feature space of this round of training based on the parameter floating information and the spatial range of the high-dimensional extended feature space adopted in the previous round of training.

[0055] Among them, for some alternative implementation methods, the learning error association information includes the spatial range learning error of the high-dimensional extended feature space, and the updating of the learning error association information of the bridge vibration anomaly detection model includes: calculating the spatial range learning error of the high-dimensional extended feature space based on the spatial range of the high-dimensional extended feature space of this round of training.

[0056] Among them, for some alternative implementation methods, the learning error association information also includes a vibration threshold value; the learning error association information for updating the bridge vibration anomaly detection model also includes: updating the vibration threshold value of the bridge vibration data to be learned and the corresponding anomaly detection label based on the matching degree of the bridge vibration data to be learned and each anomaly detection label determined by the anomaly detection prediction data of the bridge vibration data to be learned.

[0057] Among them, for some alternative implementation methods, the learning error association information includes a vibration threshold value; the learning error association information of the bridge vibration anomaly detection model is updated based on the anomaly detection prediction data of the bridge vibration data to be learned, and also includes: updating the vibration threshold value of the bridge vibration data to be learned and the corresponding anomaly detection label based on the matching degree of the bridge vibration data to be learned and each anomaly detection label determined by the anomaly detection prediction data of the bridge vibration data to be learned.

[0058] Among them, for some alternative implementation methods, mapping the multimodal vibration timing characteristics to a high-dimensional extended feature space to obtain high-dimensional extended mapping characteristics includes: mapping the multimodal vibration timing characteristics to a high-dimensional extended feature space according to the spatial range of the high-dimensional extended feature space of this round of training to obtain high-dimensional extended mapping characteristics.

[0059] Among them, for some alternative implementation methods, the learning error of the bridge vibration anomaly detection model is calculated based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned and the updated learning error association information, including: calculating a first learning error value based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned and the vibration threshold value; obtaining the learning error of the bridge vibration anomaly detection model based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space.

[0060] Among them, for some alternative implementation methods, the learning error of the bridge vibration anomaly detection model is calculated based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned and the updated learning error association information, including: calculating a first learning error value based on the anomaly detection prediction data and the actual anomaly detection data of the bridge vibration data to be learned; and calculating the learning error of the bridge vibration anomaly detection model based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space.

[0061] Among them, for some alternative implementation methods, the learning error of the bridge vibration anomaly detection model is calculated based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned and the updated learning error association information, including: calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned and the vibration threshold value.

[0062] Wherein, for some alternative implementations, before mapping the multimodal vibration time series features to the high-dimensional extended feature space to obtain the high-dimensional extended mapping features, the embodiment of the present application may further perform an L2 regularization transformation on the multimodal vibration time series features to obtain the regularized transformed multimodal vibration time series features. Mapping the multimodal vibration time series features to the high-dimensional extended feature space to obtain the high-dimensional extended mapping features includes: mapping the regularized transformed multimodal vibration time series features to the high-dimensional extended feature space to obtain the high-dimensional extended mapping features;

[0063] Among them, for some alternative implementations, before inputting the high-dimensional extended mapping features into the prediction unit of the bridge vibration anomaly detection model to obtain the anomaly detection prediction data of the bridge vibration anomaly detection model for the bridge vibration data to be learned, it also includes: performing L2 regularization conversion on the prediction unit parameters of the bridge vibration anomaly detection model to obtain the prediction unit parameters after regularization conversion.

[0064] In some alternative embodiments, the aforementioned edge computing-based bridge vibration detection system may include one or more processors and a memory. The memory may store data and instructions for execution or use by the edge computing-based bridge vibration detection system, which may implement the method described in the present invention by executing or using the data and instructions.

[0065] In various embodiments, the edge computing-based bridge vibration detection system can be, but is not limited to, a terminal device such as a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). In various embodiments, the edge computing-based bridge vibration detection system can have more or fewer components and / or a different architecture. For example, in some embodiments, the edge computing-based bridge vibration detection system includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, an application-specific integrated circuit (ASIC), and a speaker.

[0066] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0067] The above is a detailed introduction to the bridge vibration detection method and system based on edge computing provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0068] According to one aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the bridge vibration detection method based on edge computing described in the aforementioned embodiments.

[0069] According to one aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in various optional implementations of the above-described embodiments.

[0070] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0071] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0072] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0073] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0074] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0075] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, ordinary technical members in this field should understand that they can still modify the technical solutions recorded in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

[0076] The embodiments described above are only some of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of protection of the present invention, but merely represents selected embodiments of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims. In addition, based on the embodiments of the present invention, all other embodiments that can be obtained by those skilled in the art without making creative work should fall within the scope of protection of the present invention.

Claims

1. A bridge vibration detection method based on edge computing, characterized in that: include: Acquire target bridge vibration data; Inputting the target bridge vibration data into a bridge vibration anomaly detection model, and obtaining anomaly detection data of the target bridge vibration data based on the bridge vibration anomaly detection model, wherein the bridge vibration anomaly detection model is implemented by running an edge computing node; The training steps of the bridge vibration anomaly detection model include: Loading the bridge vibration data to be learned into a bridge vibration anomaly detection model configured according to the AI model structure to obtain multimodal vibration time series characteristics of the bridge vibration data to be learned; Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features; Inputting the high-dimensional extended mapping feature into the prediction unit of the bridge vibration anomaly detection model to obtain anomaly detection prediction data of the bridge vibration data to be learned by the bridge vibration anomaly detection model; updating learning error association information of the bridge vibration anomaly detection model; wherein the learning error association information is used to reduce the learning error of the bridge vibration anomaly detection model; Calculating a learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information; Back-propagation is performed based on the learning error to update the bridge vibration anomaly detection model, and iterative learning is continued according to the updated bridge vibration anomaly detection model until the model converges, thereby obtaining the bridge vibration anomaly detection model; The method further comprises: Determining parameter floating information of the bridge vibration anomaly detection model based on a learning error of the bridge vibration anomaly detection model in a previous round of training; Calculate the spatial range of the high-dimensional extended feature space used in this round of training based on the parameter floating information and the spatial range of the high-dimensional extended feature space used in the previous round of training; The learning error association information includes the spatial range learning error of the high-dimensional extended feature space, and the updating of the learning error association information of the bridge vibration anomaly detection model includes: Calculate the spatial range learning error of the high-dimensional extended feature space based on the spatial range of the high-dimensional extended feature space of this round of training; The learning error association information further includes a vibration threshold value; and the learning error association information for updating the bridge vibration anomaly detection model further includes: Based on the matching degree between the bridge vibration data to be learned and each abnormality detection label determined by the abnormality detection prediction data of the bridge vibration data to be learned, the vibration threshold value of the bridge vibration data to be learned and the corresponding abnormality detection label is updated.

2. The bridge vibration detection method based on edge computing according to claim 1 is characterized in that: Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features includes: According to the spatial range of the high-dimensional extended feature space of this round of training, the multimodal vibration time series features are mapped to the high-dimensional extended feature space to obtain high-dimensional extended mapping features.

3. The bridge vibration detection method based on edge computing according to claim 1 is characterized in that: Calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes: Calculating a first learning error value based on the abnormality detection prediction data, the actual abnormality detection data of the bridge vibration data to be learned, and the vibration threshold value; A learning error of the bridge vibration anomaly detection model is obtained based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space.

4. The bridge vibration detection method based on edge computing according to claim 1 is characterized in that: The calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes: Calculating a first learning error value based on the abnormality detection prediction data and the actual abnormality detection data of the bridge vibration data to be learned; Based on the first learning error value and the spatial range learning error of the high-dimensional extended feature space, a learning error of the bridge vibration anomaly detection model is calculated.

5. The bridge vibration detection method based on edge computing according to claim 1 is characterized in that: The calculating the learning error of the bridge vibration anomaly detection model based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the updated learning error association information includes: The learning error of the bridge vibration anomaly detection model is calculated based on the anomaly detection prediction data, the actual anomaly detection data of the bridge vibration data to be learned, and the vibration threshold value.

6. The bridge vibration detection method based on edge computing according to claim 1 is characterized in that: The method further includes: before mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features, further including: Performing L2 regularization transformation on the multimodal vibration time series features to obtain regularized transformed multimodal vibration time series features; Mapping the multimodal vibration time series features to a high-dimensional extended feature space to obtain high-dimensional extended mapping features includes: Mapping the regularized transformed multi-modal vibration time series features into a high-dimensional extended feature space to obtain high-dimensional extended mapping features; Before inputting the high-dimensional extended mapping feature into the prediction unit of the bridge vibration anomaly detection model to obtain anomaly detection prediction data of the bridge vibration anomaly detection model for the to-be-learned bridge vibration data, the method further includes: Perform L2 regularization transformation on the prediction unit parameters of the bridge vibration anomaly detection model to obtain the prediction unit parameters after regularization transformation.

7. A bridge vibration detection system based on edge computing, characterized in that: It includes a processor and a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by the processor, the bridge vibration detection method based on edge computing according to any one of claims 1 to 6 is implemented.

Images