Infrastructure project site digital management method and system

Abstract

The invention provides an infrastructure project site digital management method and system, and relates to the technical field of infrastructure site Internet of Things. The method comprises the following steps: acquiring environmental parameters through a sensor to obtain capital construction sensing data, packaging the capital construction sensing data through an edge data transmission protocol, and sending the capital construction sensing data to an edge proxy terminal through a communication interface of the sensor; the edge proxy terminal performs edge calculation on the capital construction sensing data through an edge detection algorithm to obtain a management and control processing result, converts the management and control processing result into a management and control signal, and sends the management and control signal to the sensor through the communication interface for management and control processing; and performing classified management on the management and control processing result according to the capital construction data type by utilizing the pre-constructed project management cloud platform, and displaying a classification result. And the security of the construction process of the power grid infrastructure project is monitored by utilizing the Internet of Things sensing equipment and an edge detection algorithm, so that early-warning management and control are realized, and early-warning management and control are realized on the field security of the power grid project under construction.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of infrastructure site Internet of Things, and specifically to a digital management method and system for infrastructure project sites. Background Art

[0002] During the construction of power grid infrastructure projects, it is necessary to strictly supervise the staff and the working environment to ensure the safety of the construction process. However, since the infrastructure project site is affected by the surrounding environment, there is no fixed network and the mobile network coverage is also poor. The construction personnel at the power grid infrastructure project site do not have the ability to maintain and operate the information system. Therefore, it is very difficult to use only cloud servers to process various supervision data at the construction site, which makes it difficult to implement safety supervision.

[0003] With the emergence of edge computing, some cloud server computing tasks can be placed in the edge nodes between data sources and cloud servers, reducing dependence on cloud servers and the demand for network bandwidth. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for digital management of infrastructure engineering sites in view of the deficiencies in the prior art.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] A method for digital management of infrastructure construction site, comprising the following steps:

[0007] The sensor group collects different environmental parameters through multiple different sensors set up at the infrastructure project site to obtain multiple infrastructure sensor data, and respectively encapsulates the multiple infrastructure sensor data through the edge data transmission protocol, and sends each infrastructure sensor data to the edge proxy terminal through the communication interface corresponding to each sensor;

[0008] The edge proxy terminal respectively unpacks the encapsulated multiple infrastructure sensor data, performs edge computing on the unpacked multiple infrastructure sensor data through the edge detection algorithm, obtains multiple control processing results, converts the multiple control processing results into corresponding control signals according to the preset early warning rules, and sends each of the control signals to the corresponding sensor through the corresponding communication interface for control processing;

[0009] The edge proxy terminal uploads the control and processing results of each sensor to the pre-built engineering management cloud platform, uses the pre-built engineering management cloud platform to classify and manage each control and processing result according to the infrastructure data type, and displays the classification results.

[0010] Another technical solution of the present invention to solve the above technical problems is as follows:

[0011] A digital management system for infrastructure engineering site, comprising:

[0012] Deploy a sensor group to the perception layer, wherein the sensor group collects different environmental parameters through a plurality of different sensors set up at the infrastructure project site, obtains a plurality of infrastructure sensor data, and respectively encapsulates the plurality of infrastructure sensor data through an edge data transmission protocol, and sends each infrastructure sensor data to an edge proxy terminal through a communication interface corresponding to each sensor;

[0013] Deploy edge proxy terminals to the edge layer, which are used for the edge proxy terminals to respectively unpack multiple packaged infrastructure sensor data, perform edge computing on the multiple unpacked infrastructure sensor data through edge detection algorithms, obtain multiple control processing results, convert the multiple control processing results into corresponding control signals according to preset early warning rules, and send each of the control signals to the corresponding sensor through the corresponding communication interface for control processing;

[0014] Deploy a pre-built engineering management cloud platform to the application layer, which is used by the edge proxy terminal to upload the management and control processing results of each sensor to the pre-built engineering management cloud platform, and use the pre-built engineering management cloud platform to classify and manage each management and control processing result according to the infrastructure data type, and display the classification results.

[0015] The beneficial effects of the present invention are: using the computing resources of edge nodes to share the computing pressure of the cloud platform, while the edge nodes can save network bandwidth, and through edge computing to analyze various sensor data at the infrastructure construction site, various sensor data can be structured, reducing the dependence of computing resources on the cloud platform and the demand for bandwidth. The safety of the construction process of power grid infrastructure projects is monitored using IoT sensing devices and edge detection algorithms to achieve early warning control management, and the on-site construction data is managed through the engineering management cloud platform, realizing effective control of the safety of the on-site power grid projects under construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a method for digital management of infrastructure construction sites provided by an embodiment of the present invention;

[0017] Figure 2 An architectural diagram of a method for digital management of infrastructure construction sites provided by an embodiment of the present invention;

[0018] Figure 3 A module block diagram of a digital management system for infrastructure projects provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0020] When using a cloud server to process sensor data, the image needs to be transmitted to the cloud server through the network, and the calculation result is returned to the device after the calculation. However, this process will inevitably have a certain delay, especially in the field scene of power engineering construction. Embedded computing means that the sensor data of the device does not need to be connected to the Internet to perform large-scale calculations through the cloud server, but the sensor data of the device is calculated locally, and real-time environmental perception, human-computer interaction, and decision-making control are performed without connecting to the Internet. With the emergence of edge computing, part of the computing tasks of the cloud server can be placed in the edge nodes between the data source and the cloud server, and the computing resources and network resources of the edge nodes can be used to share the computing pressure of the cloud server. At present, the cloud-edge collaborative system architecture is mainly oriented to the application scenarios of the Internet and mobile Internet, and for the industrial Internet of Things scenarios, including power infrastructure application scenarios, it is currently an urgent need to build an industrial Internet of Things cloud-edge collaborative management platform.

[0021] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a method for digital management of infrastructure construction sites, comprising the following steps:

[0022] The sensor group collects different environmental parameters through multiple different sensors set up at the infrastructure project site to obtain multiple infrastructure sensor data, and respectively encapsulates the multiple infrastructure sensor data through the edge data transmission protocol, and sends each infrastructure sensor data to the edge proxy terminal through the communication interface corresponding to each sensor;

[0023] The edge proxy terminal respectively unpacks the encapsulated multiple infrastructure sensor data, performs edge computing on the unpacked multiple infrastructure sensor data through the edge detection algorithm, obtains multiple control processing results, converts the multiple control processing results into corresponding control signals according to the preset early warning rules, and sends each of the control signals to the corresponding sensor through the corresponding communication interface for control processing;

[0024] The edge proxy terminal uploads the control and processing results of each sensor to the pre-built engineering management cloud platform, uses the pre-built engineering management cloud platform to classify and manage each control and processing result according to the infrastructure data type, and displays the classification results.

[0025] Specifically, the sensing terminal (i.e., sensor group) includes: environmental monitoring equipment, mobile control balls, bracelet equipment, inclination sensors, tension sensors, helmet positioning sensors, slope sensors, gates, personnel positioning equipment, foundation settlement monitoring equipment, high-rise formwork monitoring equipment, human health detection equipment, video surveillance equipment, and harmful gas detection equipment, etc.

[0026] It should be understood that the engineering management cloud platform can display the classification results of personnel management, safety management, progress management, quality management, and alarm management through large screens, mobile APPs, and WEB terminals. By analyzing infrastructure sensor data through edge computing nodes, the data can be structured or filtered to reduce the system's dependence on cloud systems and bandwidth requirements. By using IoT perception and analysis, remote monitoring capabilities are enhanced to achieve effective control of the safety of power grid projects under construction.

[0027] In the embodiment of the present invention, the computing resources of edge nodes (i.e. edge computing nodes) are used to share the computing pressure of the cloud platform. At the same time, the edge nodes can save network bandwidth. Through edge computing, the analysis of various sensor data at the infrastructure construction site can structure various sensor data, reduce the dependence of computing resources on the cloud platform and the demand for bandwidth. The safety of the construction process of power grid infrastructure projects is monitored using IoT sensing devices and edge detection algorithms to achieve early warning control management, and the on-site construction data is managed through the engineering management cloud platform to enhance remote monitoring capabilities, thereby achieving effective control of the safety of the on-site power grid projects under construction.

[0028] Preferably, the encapsulating the plurality of infrastructure sensing data respectively through the edge data transmission protocol comprises:

[0029] Constructing an edge data message format in the order of a message header, a frame type, a message type, a message length, a device unique identifier, a timestamp, a manufacturer unique identifier, a message content, and a check bit, and encapsulating each of the infrastructure sensor data in accordance with the edge data message format;

[0030] Among them, encapsulating the infrastructure sensor data includes: calculating the message header, frame type, message type, message length, device unique identifier, timestamp, manufacturer unique identifier and message content through the CRC16 check algorithm to obtain the check code of the check bit.

[0031] Specifically, the message header is used to identify the online monitoring datagram, which is represented by the hexadecimal value 8BB8 or the decimal value 35768. The frame type is used to distinguish and identify the data frame according to the function. It is represented by the lower four bits of a single byte in storage, and the upper four bits are reserved. The message type is used to distinguish and identify the data frame according to different monitoring types. It is represented by the lower four bits of a single byte in storage, and the upper four bits are reserved. The message length is the length of the message content data in the frame structure, in bytes. The device unique identifier is an identifier that distinguishes the device. The timestamp is the time information when the device sends data. The manufacturer unique identifier is an identifier that distinguishes the manufacturer of the device. The message content is a data byte length that is not fixed. The check bit is the most commonly used error check code in the field of data communication, and its characteristic is that the length of the information field and the check field can be selected arbitrarily. The check code is generated by a specific algorithm and appended to the end of the original data. When the data is transmitted or stored, the receiving end will use the same algorithm to recalculate the check code and compare it with the received check code. If the check codes match, the transmitted data is considered to be error-free. If they do not match, a data error is detected and needs to be retransmitted.

[0032] The frame data arrangement format is integer (occupying 2 bytes) and long integer (occupying 4 bytes), both of which are stored in the low byte first mode, that is, the bytes are arranged from low Bl to high Bn, and the byte positions are arranged from high b7 to b0. Frame types include: datagram (sent from the sensor to the edge detection algorithm), data response message (sent from the edge detection algorithm to the sensor), control message (sent from the edge detection algorithm to the sensor), response message of control message / video control message (sent from the sensor to the edge detection algorithm), heartbeat message (sent from the sensor to the edge detection algorithm), image and video data type (sent from the sensor to the edge detection algorithm), and image and video control type (sent from the sensor to the edge detection algorithm).

[0033] Message types include: environmental monitoring equipment, safety helmets, video analysis equipment, tower crane monitoring equipment, tension sensors, inclination sensors, settlement sensors, deep foundation pit monitoring data reports, high formwork data reports, wristband devices, slope sensors, gas monitoring sensors and other equipment.

[0034] It also includes: using ACK mechanism to do retransmission response. When the receiver receives a complete packet of data frames, it sends an ACK message as a response; if the sender does not receive the ACK response message within 30S, it will resend it twice (a total of 3 times). Response mode data message format: message header, frame type, message type, message length, data transmission status (10xFF success or 20x00 failure) and check bit. Table 1 is the command mode data message format of the settlement sensor data message, and Table 2 is the response mode data message format.

[0035] Table 1:

[0036]

[0037] Table 2:

[0038]

[0039] In the embodiment of the present invention, sensor data is encapsulated in a specific message format to achieve data transmission from a device terminal to a computing node.

[0040] Preferably, the step of sending each infrastructure sensing data to the edge proxy terminal through a communication interface corresponding to each sensor includes:

[0041] If the sensor is communicating over long distances, the packaged different infrastructure sensing data are sent to the Lora interface of the edge proxy terminal through the Lora interface;

[0042] If the sensor is a short-range communication and / or battery-powered device, the packaged different infrastructure sensor data is sent to the Zigbee interface of the edge proxy terminal through the Zigbee interface;

[0043] If the sensor is a long-distance communication and / or industrial device, the packaged different infrastructure sensor data is sent to the Modbus interface of the edge proxy terminal through the Modbus interface;

[0044] If the sensor is a short-distance communication, the packaged different infrastructure sensor data are sent to the Wifi interface of the edge proxy terminal through the Wifi interface.

[0045] Specifically, environmental monitoring sensors, mobile control balls and helmet positioning sensors communicate through the Wifi interface, health monitoring bracelets, tilt sensors, tension sensors and slope sensors communicate through the Lora interface, gates communicate through the Modbus interface, lighting control and temperature controllers communicate through the Zigbee interface, personnel positioning equipment communicates through the Beidou / GPS interface, and environmental monitoring sensors, cameras and mobile control balls can also communicate through network cables. That is, it supports transmission methods such as WiFi, Lora, 4G / 5G, power wireless private network, Beidou / GPS, Modbus and Zigbee.

[0046] In the embodiment of the present invention, corresponding sensor data are transmitted in different communication modes through different communication interfaces, so that multiple data supervision of the construction site is realized and the safety of infrastructure is improved.

[0047] Preferably, the edge detection algorithm is used to perform edge computing on the multiple infrastructure sensor data after unblocking to obtain multiple control processing results, including:

[0048] If the unsealed infrastructure sensor data is an image, the image is subjected to feature recognition through an image detection algorithm to obtain an image control processing result;

[0049] If the unsealed sensor data is a numerical value, the numerical value is compared with a preset threshold to obtain a numerical control processing result.

[0050] It should be understood that infrastructure sensor data includes: gate sensor data for face recognition punching, personnel health sensor data such as heart rate, blood pressure and body temperature obtained from health monitoring bracelets, personnel clothing sensor data for detecting illegal clothing behaviors such as wearing shorts, short sleeves and not wearing safety helmets, fire sensor data for detecting smoke and fire phenomena, and environmental sensor data for detecting environmental data such as PM2.5, wind speed, temperature and humidity. Among them, image data can be extracted frame by frame from the video. If PM2.5 exceeds the standard, fog cannon dust removal operation is automatically or manually started; for important construction scenes, such as dust shed operations, PM2.5, wind speed, temperature and humidity at the on-site construction site are queried and statistically managed, or during the construction of main equipment, intelligent micro-environment monitoring means are used to collect and monitor the status of harmful gases such as oxygen content and carbon monoxide in real time, and to realize real-time alarm on site when the threshold is exceeded. Or when working in a closed pipe gallery, aerobic detection is first performed and real-time video monitoring and storage are performed to ensure the safety of entering the construction.

[0051] In the embodiment of the present invention, various sensor data of the infrastructure construction site are detected to ensure the safety of the workers and the construction safety of the construction site.

[0052] Preferably, the image detection algorithm is used to perform feature recognition on the image to obtain an image control processing result, including:

[0053] A face recognition model is constructed based on a convolutional neural network (CNN), and the face recognition model is trained using a face image training set. The trained face recognition model is used to perform feature recognition on face images collected by gates at infrastructure project sites to obtain face key feature vectors. A comparison algorithm is used to match the face key feature vectors with face key feature vectors in a pre-constructed personnel face database to obtain a matching result.

[0054] Specifically, the existing gate machines are upgraded and reconstructed to add face recognition functions and provide non-cooperative recognition functions for face recognition gate opening and attendance clocking in. That is, when the matching result is successful, the attendance data is recorded, including the check-in time and attendance status image, etc. The personnel check-in results are transmitted to the cloud server (i.e., the engineering management cloud platform) through Beidou short messages.

[0055] It is also possible to construct a vehicle recognition model, train the vehicle recognition model using a vehicle training set, and use the trained vehicle recognition model to identify vehicles entering and exiting the gate and record the information of vehicles entering and exiting.

[0056] It should be understood that when a worker stands in front of the time clock, the time clock will automatically start the built-in camera or infrared sensor to capture the worker's facial image; the facial recognition model will perform face detection and positioning on the collected facial image to find the face position and area in the image; then it will extract the key feature points of the face, such as the shape and position information of the eyes, nose, mouth and other parts, and the extracted facial features will be compared with the pre-stored worker face database, which contains the facial feature information of all workers on the construction site and the corresponding identity information; the comparison algorithm will be used to determine whether the current facial features match the workers in the database.

[0057] In an embodiment of the present invention, at a construction site with a poor network environment, edge computing nodes are used to enable offline check-in for personnel without a network environment, thereby achieving attendance control of staff and control of entry and exit of personnel at the construction site.

[0058] Preferably, after the step of obtaining the matching result, the method further comprises:

[0059] If the matching result is successful, a facial action prompt instruction is generated, and multiple facial action images of the current person performing the actions according to the action prompt instruction are continuously collected at a preset frequency. The multiple facial action images are detected and judged through the liveness detection algorithm to identify whether the current person is alive. If not, the matching result is modified to failure.

[0060] In the embodiment of the present invention, a liveness detection algorithm is used to determine whether the person currently punching in is a living person rather than an image, thereby avoiding situations where someone punches in for someone else or impersonates a staff member.

[0061] Preferably, the step of performing feature recognition on the image by using an image detection algorithm to obtain an image control processing result further includes:

[0062] A clothing detection model is constructed based on the YOLOv5 algorithm architecture, and the clothing detection model is trained using a clothing image training set. The trained clothing detection model is used to detect clothing images of personnel collected by cameras at infrastructure project sites to obtain clothing detection results.

[0063] Specifically, video analysis is used to detect whether the people appearing in the video are dressed for work as required, and to analyze and supervise illegal dressing behaviors such as wearing shorts, short sleeves, and not wearing safety helmets.

[0064] It should be understood that the YOLOv5 model is mainly composed of three parts: Backbone, Neck and Head. Among them, Backbone is mainly responsible for feature extraction of input images; Neck is responsible for multi-scale feature fusion of feature maps and passing these features to the prediction layer (ie Head); Head performs the final regression prediction.

[0065] In the embodiment of the present invention, the illegal dressing behaviors of the staff are monitored to prevent the staff from not dressing according to the regulations and causing safety accidents, which will affect the safety of life.

[0066] Preferably, the step of performing feature recognition on the image by using an image detection algorithm to obtain an image control processing result further includes:

[0067] A fireworks recognition model is constructed based on the YOLOv3 algorithm architecture, and the fireworks recognition model is trained using a fireworks image training set. The trained fireworks recognition model is used to detect infrastructure environment images captured by cameras at infrastructure project sites to obtain fireworks detection results.

[0068] Specifically, by adding high-point panoramic cameras to build a high-low linkage video monitoring system, the entire substation construction site can be controlled in real time and in full scene. Through video analysis, smoke and fire appearing in the video footage can be identified, which effectively assists the fire management of the infrastructure site and strengthens the safety management and control of the site.

[0069] In the embodiment of the present invention, the fire safety at the construction site is controlled to avoid the occurrence of fire accidents suspected of fire.

[0070] Preferably, the converting of the plurality of control processing results into corresponding control signals according to preset warning rules and sending them to corresponding sensors for control processing includes:

[0071] When the result of the numerical control processing does not meet the preset threshold, a corresponding control signal is generated according to the data type and sent to the corresponding sensor for alarm to prompt the staff to handle it. For example, when the health data such as heart rate, blood pressure and body temperature obtained by the health monitoring bracelet exceeds or falls below the preset threshold, a sound alarm or light alarm is issued; when the environmental data such as PM2.5, wind speed, temperature and humidity exceeds or falls below the preset threshold, a sound alarm or light alarm is issued.

[0072] When the image control processing result is that the target feature is identified, a corresponding control signal is generated according to the feature type and sent to the corresponding sensor for alarm to prompt the staff to handle it. For example, when a person punches in at the gate, if the face recognition match is passed, the punch-in is successful and the gate is opened according to the gate opening signal; when the clothing of the construction site personnel is detected, the characteristics of the personnel with illegal dress are detected, and a violation alarm signal is generated to control the corresponding sensor to issue an alarm reminder; when the fire scene of the construction site is detected, the characteristics of fireworks are detected, and a fireworks alarm signal is generated to control the corresponding sensor to issue an alarm reminder.

[0073] In the embodiment of the present invention, warnings are issued according to the warning rules corresponding to different monitoring data at the infrastructure project site to achieve personnel management and production safety monitoring.

[0074] Preferably, uploading the control and processing results of each sensor to the pre-built engineering management cloud platform includes:

[0075] The control and processing results of each sensor are encapsulated through an encryption protocol and uploaded to the pre-built engineering management cloud platform.

[0076] Specifically, an asymmetric encryption algorithm (including RSA, ECC, etc.) is used to encrypt and calculate the multiple infrastructure sensor data and the corresponding management and control processing results to encapsulate and transmit the data to ensure the security of the data during transmission. The data is encrypted and decrypted between the two nodes of the communication link (i.e., the sensor and the edge node) to ensure the security of the data during transmission. During the link encryption process, the sender converts the plaintext data into ciphertext data through an encryption algorithm and then transmits it on the link. After receiving the ciphertext data, the receiver uses the same key to decrypt it and restore it to the original plaintext data. Since the data is always encrypted during transmission, even if the link is attacked, the attacker cannot obtain useful information.

[0077] In the embodiment of the present invention, an asymmetric encryption algorithm is used to implement functions such as confidential distribution of keys and digital signatures, thereby ensuring security during data transmission.

[0078] like Figure 3 As shown, an infrastructure project site digital management system provided by an embodiment of the present invention includes:

[0079] Deploy a sensor group to the perception layer, wherein the sensor group collects different environmental parameters through a plurality of different sensors set up at the infrastructure project site, obtains a plurality of infrastructure sensor data, and respectively encapsulates the plurality of infrastructure sensor data through an edge data transmission protocol, and sends each infrastructure sensor data to an edge proxy terminal through a communication interface corresponding to each sensor;

[0080] Deploy edge proxy terminals to the edge layer, which are used for the edge proxy terminals to respectively unpack multiple packaged infrastructure sensor data, perform edge computing on the multiple unpacked infrastructure sensor data through edge detection algorithms, obtain multiple control processing results, convert the multiple control processing results into corresponding control signals according to preset early warning rules, and send each of the control signals to the corresponding sensor through the corresponding communication interface for control processing;

[0081] Deploy a pre-built engineering management cloud platform to the application layer, which is used by the edge proxy terminal to upload the management and control processing results of each sensor to the pre-built engineering management cloud platform, and use the pre-built engineering management cloud platform to classify and manage each management and control processing result according to the infrastructure data type, and display the classification results.

[0082] It should be understood that the edge proxy terminal is a low-power chip that can greatly reduce energy consumption. The power supply problem of the chip can be solved through on-site solar energy and the chip's energy storage module to achieve real-time edge computing.

[0083] The above-mentioned infrastructure construction site digital management system can refer to the implementation content and beneficial effects of the infrastructure construction site digital management method specifically described above, which will not be repeated here.

[0084] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0085] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for digital management of infrastructure construction sites, characterized in that: The steps include: The sensor group collects different environmental parameters through multiple different sensors set up at the infrastructure project site to obtain multiple infrastructure sensor data, and respectively encapsulates the multiple infrastructure sensor data through the edge data transmission protocol, and sends each infrastructure sensor data to the edge proxy terminal through the communication interface corresponding to each sensor; The edge proxy terminal respectively unpacks the encapsulated multiple infrastructure sensor data, performs edge computing on the unpacked multiple infrastructure sensor data through the edge detection algorithm, obtains multiple control processing results, converts the multiple control processing results into corresponding control signals according to the preset early warning rules, and sends each of the control signals to the corresponding sensor through the corresponding communication interface for control processing; The edge proxy terminal uploads the control and processing results of each sensor to the pre-built engineering management cloud platform, uses the pre-built engineering management cloud platform to classify and manage each control and processing result according to the infrastructure data type, and displays the classification results.

2. The method for digital management of infrastructure construction site according to claim 1, characterized in that: The encapsulating the plurality of infrastructure sensing data respectively through the edge data transmission protocol comprises: Constructing an edge data message format in the order of a message header, a frame type, a message type, a message length, a device unique identifier, a timestamp, a manufacturer unique identifier, a message content, and a check bit, and encapsulating each of the infrastructure sensor data in accordance with the edge data message format; Among them, encapsulating the infrastructure sensor data includes: calculating the message header, frame type, message type, message length, device unique identifier, timestamp, manufacturer unique identifier and message content through the CRC16 check algorithm to obtain the check code of the check bit.

3. The method for digital management of infrastructure construction site according to claim 1, characterized in that: The step of sending each infrastructure sensing data to the edge proxy terminal through the communication interface corresponding to each sensor includes: If the sensor is communicating over long distances, the packaged different infrastructure sensing data are sent to the Lora interface of the edge proxy terminal through the Lora interface; If the sensor is a short-range communication and / or battery-powered device, the packaged different infrastructure sensor data is sent to the Zigbee interface of the edge proxy terminal through the Zigbee interface; If the sensor is a long-distance communication and / or industrial device, the packaged different infrastructure sensor data is sent to the Modbus interface of the edge proxy terminal through the Modbus interface; If the sensor is a short-distance communication, the packaged different infrastructure sensor data are sent to the Wifi interface of the edge proxy terminal through the Wifi interface.

4. The method for digital management of infrastructure construction site according to claim 1, characterized in that: The edge detection algorithm is used to perform edge computing on the multiple infrastructure sensor data after unblocking to obtain multiple control processing results, including: If the unsealed infrastructure sensor data is an image, the image is subjected to feature recognition through an image detection algorithm to obtain an image control processing result; If the unsealed sensor data is a numerical value, the numerical value is compared with a preset threshold to obtain a numerical control processing result.

5. The method for digital management of infrastructure construction site according to claim 4, characterized in that: The image detection algorithm is used to perform feature recognition on the image to obtain an image control processing result, including: A face recognition model is constructed based on a convolutional neural network (CNN), and the face recognition model is trained using a face image training set. The trained face recognition model is used to perform feature recognition on face images collected by gates at infrastructure project sites to obtain face key feature vectors. A comparison algorithm is used to match the face key feature vectors with face key feature vectors in a pre-constructed personnel face database to obtain a matching result.

6. The method for digital management of infrastructure construction site according to claim 5, characterized in that: After the step of obtaining the matching result, the method further includes: If the matching result is successful, a facial action prompt instruction is generated, and multiple facial action images of the current person performing the actions according to the action prompt instruction are continuously collected at a preset frequency. The multiple facial action images are detected and judged through the liveness detection algorithm to identify whether the current person is alive. If not, the matching result is modified to failure.

7. The method for digital management of infrastructure construction site according to claim 4, characterized in that: The performing feature recognition on the image by using an image detection algorithm to obtain an image control processing result also includes: A clothing detection model is constructed based on the YOLOv5 algorithm architecture, and the clothing detection model is trained using a clothing image training set. The trained clothing detection model is used to detect clothing images of personnel collected by cameras at infrastructure project sites to obtain clothing detection results.

8. The method for digital management of infrastructure construction site according to claim 4, characterized in that: The performing feature recognition on the image by using an image detection algorithm to obtain an image control processing result also includes: A fireworks recognition model is constructed based on the YOLOv3 algorithm architecture, and the fireworks recognition model is trained using a fireworks image training set. The trained fireworks recognition model is used to detect infrastructure environment images captured by cameras at infrastructure project sites to obtain fireworks detection results.

9. The method for digital management of infrastructure construction site according to claim 1, characterized in that: The process of uploading the control and processing results of each sensor to the pre-built engineering management cloud platform includes: The control and processing results of each sensor are encapsulated through an encryption protocol and uploaded to the pre-built engineering management cloud platform.

10. A digital management system for infrastructure construction projects, characterized in that: include: Deploy a sensor group to the perception layer, wherein the sensor group collects different environmental parameters through a plurality of different sensors set up at the infrastructure project site, obtains a plurality of infrastructure sensor data, and respectively encapsulates the plurality of infrastructure sensor data through an edge data transmission protocol, and sends each infrastructure sensor data to an edge proxy terminal through a communication interface corresponding to each sensor; Deploy edge proxy terminals to the edge layer, which are used for the edge proxy terminals to respectively unpack multiple packaged infrastructure sensor data, perform edge computing on the multiple unpacked infrastructure sensor data through edge detection algorithms, obtain multiple control processing results, convert the multiple control processing results into corresponding control signals according to preset early warning rules, and send each of the control signals to the corresponding sensor through the corresponding communication interface for control processing; Deploy a pre-built engineering management cloud platform to the application layer, which is used by the edge proxy terminal to upload the management and control processing results of each sensor to the pre-built engineering management cloud platform, and use the pre-built engineering management cloud platform to classify and manage each management and control processing result according to the infrastructure data type, and display the classification results.

Images