A welding video monitoring system, a welding video monitoring method and a welding machine

Abstract

The application discloses a kind of welding video monitoring systems, welding video monitoring method and welding machine, the system includes the welding machine integrated with image acquisition module and edge computing module, and total convergence node module and monitoring terminal module;The image acquisition module is used to collect real-time image of welding, and the edge computing module is used to receive and process image acquisition module video stream data;The total convergence node module is received with edge computing module through wireless transmission edge computing module data and is transmitted to monitoring terminal module;The monitoring terminal module receives total convergence node module data and shows real-time welding image;Method and welding machine equipment based on the above system, by integrating edge computing module and image acquisition module in welding machine, and by wireless signal connection edge computing module and total convergence node module, so that when new extension welding machine is added in the system, wiring is not needed, and equipment deployment is facilitated.

Description

Technical Field

[0001] This invention relates to the field of welding, and more particularly to the field of real-time video monitoring during the welding process of a welding machine. Background Technology

[0002] Welding is a core process in industrial production and infrastructure construction, and its quality directly determines the safety and lifespan of products / projects. With the development of industrial intelligence, real-time monitoring of the welding process (especially monitoring the molten pool condition) has become a key means to improve welding quality, reduce labor costs, and ensure construction safety. Currently, welding video monitoring systems are mainly divided into two categories: centralized workshop monitoring systems and simple portable monitoring systems. Centralized workshop monitoring systems are primarily suitable for workshop scenarios with fixed welding machine layouts and good wiring conditions. They aggregate video streams from multiple cameras to a monitoring terminal via a wired switch for unified monitoring. Simple portable monitoring systems are mainly suitable for temporary monitoring of a single welding machine, transmitting data wirelessly directly to the terminal via a single camera, lacking multi-device collaboration capabilities. In field welding construction scenarios (such as infrastructure construction sites, field pipeline laying, and offshore platform welding), welding machines are typically deployed in a dispersed manner (10-50 meters apart per machine, totaling 5-30 machines), and face the following objective limitations: 1. Lack of fixed wiring conditions; complex field construction environments, high cost and easy damage of network cable installation, and inability to adapt to the mobility requirements of welding machines; 2. Strong electromagnetic interference; welding machines generate a large amount of electromagnetic radiation during operation, interfering with the stability of wireless communication; 3. The need for simultaneous monitoring of multiple welding machines, requiring the system to have good scalability and coordination; 4. Harsh environments, requiring the system to have industrial-grade stability with high and low temperature resistance, dustproof, and vibration resistance. Existing technologies cannot meet the monitoring needs of dispersed welding machines in the field, and there is an urgent need for a welding video monitoring system that is adaptable to field scenarios, requires no wiring, is interference-free, scalable, stable, and reliable. Summary of the Invention

[0003] To address the problems of low stability in real-time monitoring of multiple welding machines in existing technologies, inconvenience in adding new welding machines, and the need for on-site wiring, the technical solution adopted in this invention is as follows: A welding machine video monitoring system includes a table welding machine integrating an image acquisition module and an edge computing module, as well as a main aggregation node module and a monitoring terminal module. The image acquisition module is used to acquire real-time images of the welding area, and the edge computing module is used to receive and process video stream data from the image acquisition module. The main aggregation node module and the edge computing module receive data from the edge computing module wirelessly and transmit it to the monitoring terminal module. The monitoring terminal module receives data from the main aggregation node module and displays real-time welding images.

[0004] The aforementioned monitoring system integrates the image acquisition module and the edge computing module within the welding machine. By optimizing the video stream data at the edge computing layer and transmitting it wirelessly to the main aggregation node, the need for cabling from the welding machine to the main aggregation node is eliminated, facilitating the expansion of new welding machines. Furthermore, the optimized video stream data processing reduces the bandwidth usage of the data transmitted from the edge computing module to the main aggregation node module, thereby improving the stability of data transmission.

[0005] Furthermore, the image acquisition module and the edge computing module use wireless data transmission.

[0006] This can further reduce the wiring requirements in the system.

[0007] Furthermore, the edge computing module has a dual-band selectable communication function of 2.4G / 5.8G.

[0008] When the edge computing module receives data from multiple image acquisition modules, it can allocate independent channels to avoid signal interference.

[0009] The image acquisition module and the edge computing module use wired data transmission.

[0010] Using a wired connection further avoids interference factors in the transmission of video stream data from the image acquisition module to the edge computing module, thus improving stability.

[0011] Furthermore, the edge computing module optimizes the video stream data in terms of bitrate, resolution, and I-frame interval.

[0012] The above optimizations reduce the bandwidth usage of video stream data.

[0013] Furthermore, at least two image acquisition modules are provided.

[0014] Multiple image acquisition modules enable more comprehensive monitoring of the welding area, facilitating the acquisition of more useful information by the terminal.

[0015] Furthermore, the edge computing module combines and aggregates the collected video stream data into a single stream.

[0016] This allows for further optimization of multi-stream video data and reduces bandwidth usage.

[0017] Furthermore, the edge computing module and the main aggregation node module use WiFi or 4G / 5G transmission methods, and the transmission link can be automatically switched.

[0018] This configuration allows for timely adjustments to the transmission method in different scenarios and when signal strength varies, thereby improving signal transmission stability.

[0019] Furthermore, the edge computing module has local storage functionality.

[0020] By storing video stream data, it is possible to trace back after a signal interruption, ensuring comprehensive monitoring of the welding process.

[0021] Furthermore, the main aggregation node module and the video surveillance terminal module are connected via wired or wireless means.

[0022] A welding video monitoring method is implemented through the following steps: S1. The edge computing module integrated in each welding machine receives video stream data from the image acquisition module in its respective welding machine; S2. Each edge computing module processes the acquired video stream data; S3. Each edge computing module wirelessly transmits the processed video stream data to the main aggregation node module; S4. The main aggregation node module processes the data received from multiple edge computing modules and transmits it to the monitoring terminal module; S5. The monitoring terminal module processes the received data and presents the welding status in real time.

[0023] The above method enables stable real-time monitoring of the welding process of multiple welding machines. Furthermore, when a welding machine is equipped with at least two image acquisition modules, the edge computing module in the welding machine performs combining and aggregation processing on the video stream data.

[0024] Optimize video stream data, reduce bandwidth usage, and further improve data transmission stability.

[0025] Furthermore, the edge computing module optimizes the video stream data in terms of bitrate, resolution, and I-frame interval.

[0026] This can further optimize video stream data and reduce bandwidth usage.

[0027] Furthermore, the edge computing module stores the video stream data locally.

[0028] It facilitates the traceability of the welding process and avoids the inability to view past welding records after packet loss.

[0029] Furthermore, the edge computing module establishes independent communication channels with each image acquisition module in its welding machine.

[0030] This optimizes multi-stream video data, reduces bandwidth, and improves data transmission stability.

[0031] Furthermore, the edge computing module and the main aggregation node module use WiFi or 4G / 5G transmission methods, and the transmission link can be automatically switched.

[0032] Optimize transmission methods to adapt to complex environments, ensure signal stability, and avoid data transmission interruptions.

[0033] Furthermore, when a new welding machine is added, the edge computing module in the welding machine establishes a wireless communication connection with the main aggregation node module.

[0034] A welding machine includes a welding machine main unit and a welding torch. The welding torch is wired to the welding machine main unit. The welding machine main unit supplies power to the welding torch and is used to adjust the welding parameters of the welding torch. An edge computing module is integrated in the welding machine main unit, and an image acquisition module is integrated on the welding torch.

[0035] By integrating the edge computing module and image acquisition module into the welding machine, when adding or expanding the welding machine at the construction site, it is only necessary to place this welding machine at the corresponding workstation and connect the edge computing module to the main aggregation node module wirelessly to complete the equipment deployment. There is no need to lay out the monitoring system cabling, thus improving scalability.

[0036] Furthermore, it also includes a wire feeding mechanism and an execution mechanism. The welding torch is mounted on the execution mechanism, and the wire feeding mechanism is connected to the welding machine host, with the wire feeding speed controlled by the welding machine host.

[0037] Furthermore, the image acquisition module and the edge computing module are connected via wired or wireless communication.

[0038] Furthermore, at least two image acquisition modules are integrated on the welding torch.

[0039] In summary, by integrating an edge computing module and an image acquisition module into the welding machine, and connecting the edge computing module to the main aggregation node module via wireless signals, no wiring is required when adding or expanding welding machines into the system, which facilitates equipment deployment. Furthermore, since the edge computing module processes the video stream data at the front end, bandwidth consumption is reduced and the stability of data transmission is improved. Attached Figure Description

[0040] Figure 1 This is a block diagram illustrating the principle of a welding video monitoring system. Figure 2 Flowchart of welding video monitoring method; Figure 3 This is a schematic diagram of a welding machine structure that integrates an image acquisition module and an edge computing module. Detailed Implementation

[0041] To clearly understand the inventive concept of this invention, the technical solution of this invention will be described below in conjunction with the accompanying drawings and specific embodiments.

[0042] This invention is protected from three aspects: system, method and device, based on the same inventive concept. Each aspect will be described in detail here so that those skilled in the art can have a clearer understanding of the inventive concept.

[0043] The first part describes the video monitoring system for the entire welding machine, such as... Figure 1 As shown, the components, in order of signal flow, include an image acquisition module, an edge computing module, a main aggregation node module, and a monitoring terminal module.

[0044] The image acquisition module and the edge computing module are both integrated into the welding machine body. The image acquisition module and the edge computing module are powered by the welding machine, or are independently connected to an external power supply, or are powered by an independent power supply module.

[0045] The aforementioned image acquisition module is primarily designed for real-time image acquisition of the weld seam and molten pool. When setting up the image acquisition module, it should be directly facing the molten pool and / or weld seam. It can be seen that one or more image acquisition modules can be set. When setting up a single image acquisition module, it can be selectively set to the area to be monitored during welding, based on the actual working conditions and the required data, to obtain effective images. Alternatively, two or more image acquisition modules can be set up to monitor the same welding area or different welding areas, obtaining a clearer and more comprehensive real-time welding image. The image acquisition module preferably uses an industrial-grade camera, which has dustproof, vibration-proof, and weld spatter-proof functions, ensuring stable operation outdoors and in complex working environments.

[0046] Since the image acquisition module and the edge computing module are integrated into the welding machine, their relative positions are fixed and close together. Therefore, the image acquisition module and the edge computing module can be connected via wired or wireless means. The welding machine referred to here includes at least the welding machine body and the welding torch. To facilitate the deployment of the edge computing module and the image acquisition module within the welding machine, the edge computing module is integrated into the welding machine body, and the image acquisition module is integrated into the welding torch. Of course, automated welding equipment may also have devices such as robotic arms to control the movement trajectory of the welding torch. In this case, the image acquisition module can also be placed at the end of the automated equipment near the welding torch to collect welding information in real time.

[0047] The image acquisition module can be configured with parameters such as resolution ≥1280×720, frame rate ≥30fps, and H264 / H265 hardware encoding. Specifically, a camera module of model MV-CA023-10GM can be selected. If a wireless connection is used with the edge computing module, it should support wireless communication. These are just examples of available parameters and models; technicians can select other models or parameters of image acquisition modules to acquire welding image information according to actual needs.

[0048] The aforementioned edge computing module is mainly responsible for receiving video stream data from the image acquisition module, processing the video stream data, and then sending it to the main aggregation node module.

[0049] The edge computing module compresses and optimizes the received video stream data before transmitting it back to the main aggregation node module wirelessly (WiFi / 4G / 5G). As mentioned earlier, when the image acquisition module and the edge computing module are connected via a wired connection, data transmission can be performed using a MIPI interface, USB interface, PoE network port, or gigabit network port. Therefore, when using a wired connection, the image acquisition module and the edge computing module should each have at least one matching set of interfaces to ensure reliable connection and data transmission. Furthermore, the number of interfaces should match the number of image acquisition modules. For example, when using two cameras, as is preferable, the edge computing module should also have at least two corresponding interfaces to receive video stream data from each camera.

[0050] To avoid unnecessary wiring, the image acquisition module and edge computing module preferably use wireless connection for video stream data transmission. In wireless transmission, the edge computing module should be an industrial-grade computing module with wireless AP and wireless data forwarding functions. The edge computing module enables an independent wireless AP only to connect to cameras configured on the same welding machine, thus enabling the image acquisition module and edge computing module in a welding machine to build an independent local area network. Specifically, when configuring the wireless AP, the edge computing module enables an independent wireless AP. The SSID can be named according to the welding machine number, an independent password can be set, and a unique channel can be assigned. When connecting to the image acquisition unit, the RTSP receiving port is configured to receive the video stream signal from the image acquisition unit. Each welding machine independently builds a local area network, which can avoid wireless signal crosstalk between different welding machines, realize the isolation of video transmission of each welding machine, and further improve the reliability of the system in scenarios with dense deployment of multiple devices. When the image acquisition module uses two cameras, the edge computing module uses its own software algorithm to merge the two video streams into one and optimize compression parameters, such as adjusting bitrate, resolution, I-frame interval, H264 / H265, etc., thereby reducing the bandwidth consumption when transmitting to the main aggregation node, alleviating bandwidth congestion caused by concurrent uploads from multiple nodes, and reducing the risk of latency and packet loss. Through the merging and aggregation of the two video streams and the optimization of compression parameters, the upload bandwidth consumption of a single edge computing module can be significantly reduced, avoiding network congestion, latency and packet loss caused by multiple welding machines uploading at the same time, and improving the overall stability of the system in large-scale deployment scenarios.

[0051] Due to the complex working conditions at the welding site, such as strong electromagnetic interference, welding spatter, and equipment vibration, wired transmission (MIPI / USB / Ethernet) has significantly better anti-interference capabilities and stability than wireless transmission, ensuring real-time and reliable transmission of video data. Wireless transmission, as the preferred solution for simplifying cabling, needs to improve its anti-interference capabilities through methods such as dual-frequency selection and channel isolation.

[0052] The edge computing module uses a wireless AP with dual-band communication capability of 2.4G / 5.8G. Considering the influence of the on-site environment and interference factors, the appropriate frequency band is adaptively selected for signal connection when the camera is matched with the edge computing module to ensure stable and reliable signal connection.

[0053] Edge computing modules should also preferably have local storage capabilities. When local storage is enabled, storage paths and storage periods can be set, and cyclic overwriting can be supported. Video data can be actively retained when communication is interrupted, facilitating later tracing.

[0054] The edge computing module and the main aggregation node module use wireless transmission, thus avoiding wired connections. No on-site wiring is required when expanding or moving the welding machine, improving the system's scalability. WiFi is prioritized for connection to the main aggregation node module; when the WiFi signal is weak or interrupted, it automatically switches to the 4G / 5G module for transmission. The transmission address is the main aggregation node module's IP address, and the port is fixed. Therefore, the edge processing module should have WiFi / 4G / 5G transmission capabilities.

[0055] In addition, to better monitor system operation, device operating status information, such as CPU utilization, device temperature, wireless signal strength, and camera online status, can be sent synchronously to the main aggregation node module, so that the background can monitor the operating status of system devices in real time.

[0056] Here are some of the available edge computing module models, such as the RK3588 industrial board with ARM architecture and the industrial control computer with x86 architecture.

[0057] The main function of the aforementioned aggregation node module is to receive and aggregate data uploaded by multiple edge computing modules, and after protocol parsing and data processing, forward it to the monitoring terminal module in a unified manner.

[0058] When selecting the main aggregation node module, data transmission between the edge computing module and the main aggregation module should be possible via WiFi or 4G / 5G. The main aggregation module should have WiFi and 4G / 5G access capabilities, support WiFi 6 and above standards, and have an automatic switching mechanism between WiFi and 4G / 5G primary and backup links. It should automatically switch to the 4G / 5G link to ensure uninterrupted communication when the WiFi signal is weak or interrupted. It should also be able to simultaneously connect multiple edge computing modules to ensure that all devices on site can access the network, and reserve a backup expansion channel for future device expansion. Similarly, data transmission between the main aggregation module and the monitoring terminal unit can be achieved via WiFi or 4G / 5G. The main aggregation module should have WiFi and 4G / 5G communication capabilities. Of course, it can also communicate with the monitoring terminal unit via wired transmission; therefore, the main aggregation module can also be configured with a wired Ethernet interface to connect to the monitoring terminal unit. Furthermore, industrial-grade hardware is preferred, meeting the requirements of a wide operating temperature range of -40℃ to 85℃, and featuring dustproof and vibration-proof characteristics. It also integrates routing and gateway functions, enabling network management, IP allocation, and data forwarding to ensure stable operation under complex field conditions. The central aggregation module is positioned in the center of the construction site to receive signals from all edge computing modules, ensuring comprehensive signal coverage.

[0059] The aforementioned monitoring terminal module mainly completes the data reception, parsing, multi-channel video stream decoding, screen display and business processing of the main aggregation module, realizes real-time monitoring of equipment operation status, and provides stable and reliable terminal display and operation and maintenance support for the system.

[0060] It mainly includes an industrial control server, a screen, and an operation and maintenance PC. The hardware is preferably industrial grade. The industrial control server is responsible for data reception, parsing, hardware decoding of multiple video streams (H264 / H265), and business processing. It relies on a multi-core CPU, large memory, and large-capacity hard drive to realize data storage, equipment status management, and command interaction. The screen displays multiple camera images in a split-screen manner. With the operation and maintenance PC running management software, it realizes real-time monitoring of equipment operation status, parameter configuration distribution, and fault diagnosis. The whole system forms a monitoring capability that integrates data processing, video decoding, screen display, and operation and maintenance management.

[0061] As can be seen from the above embodiments, this technical solution integrates the image acquisition module and the edge computing module onto the same welding machine, and independently constructs a wireless local area network within a single welding machine. This creates an isolated dedicated communication link between the image acquisition module and the edge computing module, fundamentally preventing wireless signal interference between multiple welding machines. Simultaneously, the edge computing module performs combining and bandwidth compression optimization on multiple video streams, significantly reducing the amount of data uploaded by a single device. This effectively avoids network congestion when multiple welding machines simultaneously upload data to the central aggregation module, improving the overall system transmission stability and real-time performance. Furthermore, since the image acquisition module and the edge computing module have pre-configured the local area network on the welding machine side, and the edge computing module communicates wirelessly with the central aggregation module, when expanding the system by adding welding machines, only a wireless connection needs to be established between the edge computing module of the new welding machine and the central aggregation module for immediate use. No on-site wiring is required, significantly reducing deployment difficulty and improving expansion convenience and deployment efficiency. The overall solution of this system solves the technical problems of traditional welding monitoring systems, such as complex wiring, difficult expansion, severe multi-machine interference, and easy transmission congestion, through the coordinated cooperation of integration on the welding machine side, local independent LAN, edge bandwidth optimization, and full wireless expansion.

[0062] The following is a detailed explanation of the specific methods for implementing welding monitoring based on the aforementioned welding machine video monitoring system: Based on the above system layout, hardware facilities are deployed at the construction site, and the equipment is configured for networking. For wired signal networking, such as image acquisition modules, edge computing modules, the main aggregation module, and the monitoring terminal module, wired connections can be used for signal transmission. This simply requires connecting the corresponding ports on the devices via data cables. When using wireless signal networking, an independent wireless access point (AP) is activated within each welding machine by the edge computing module, creating an independent local area network (LAN) solely for the image acquisition module of that welding machine. An independent SSID, password, and communication channel are set. Furthermore, when two cameras are installed on a single welding machine, each camera should also establish a separate communication channel with the edge computing module for image transmission, ensuring a dedicated wireless communication connection between the image acquisition module and the edge computing module, thus preventing signal interference between multiple welding machines. Simultaneously, the main aggregation node module is positioned in the center of the construction area, establishing a communication connection between the main aggregation node module and the monitoring terminal module. The target IP addresses and fixed communication ports for both the edge computing module and the main aggregation node module are configured.

[0063] Then, the following method was used to achieve real-time monitoring of multiple welding machines: S1. Real-time image acquisition and local transmission: Each image acquisition module acquires video images of the welding process in real time and transmits the video stream data to the edge computing module with which it has established a network connection.

[0064] Specifically, the image acquisition module facing the weld seam and molten pool area acquires video images of the welding process in real time, and transmits the video data to the corresponding edge computing module via wired or wireless means; when a single welding machine is equipped with two or more image acquisition modules, each image acquisition module simultaneously acquires images of the same or different welding areas and transmits them to the edge computing module respectively.

[0065] S2. Edge computing module data processing: The edge computing module processes the received video stream data.

[0066] Specifically, the edge computing module receives one or more video streams, performs merge and aggregation processing on the multiple video streams using built-in software algorithms, and optimizes and adjusts compression parameters such as bitrate, resolution, I-frame interval, and encoding format to reduce the bandwidth usage of video data transmission. At the same time, the local storage function is enabled, and the video data is cyclically overwritten and stored according to a preset storage path and period. The video data is automatically retained when communication is interrupted, which is convenient for later traceability.

[0067] S3. Data backhaul: The edge computing module wirelessly transmits the processed data to the main aggregation node module.

[0068] Specifically, the edge computing module prioritizes uploading the processed video data to the main aggregation node module via WiFi. When the WiFi signal is weak or interrupted, it automatically switches to 4G / 5G links for transmission to ensure uninterrupted data transmission. At the same time, the edge computing module also uploads operating status information such as CPU utilization, device temperature, wireless signal strength, and camera online status to the main aggregation node module.

[0069] S4. Data forwarding: The main aggregation node module processes the data received from multiple edge computing modules and forwards it to the monitoring terminal module.

[0070] Specifically, the main aggregation node module receives and aggregates video data and device status information uploaded by multiple edge computing modules. After protocol parsing and data processing, it forwards the data to the monitoring terminal module via wired Ethernet, WiFi, or 4G / 5G.

[0071] S5. Data presentation: The monitoring terminal module processes the data received from the main aggregation node module and presents it in real time.

[0072] Specifically, the monitoring terminal module uses an industrial control server to receive video data, parse protocols, and perform H264 / H265 hardware decoding, displaying multiple video feeds in real time on a large splicing screen. The maintenance PC uses management software to monitor the operating status of each device in the system in real time, distribute parameter configurations, and troubleshoot faults, thus completing the visualized monitoring and maintenance management of the entire welding process.

[0073] Based on the above system and method, when it is necessary to expand the number of welding machines in the system, the following method shall be used: When new welding equipment is needed, simply place the welding machine, which integrates the image acquisition module and the edge computing module, at a suitable workstation and power it on. The deployment can be completed by establishing a wireless communication connection between the edge computing module and the main aggregation node module. No on-site wiring is required, enabling convenient system expansion.

[0074] As can be seen from the above system and implementation method, the welding machine integrates an edge computing module and an image acquisition module, thereby optimizing the scalability of the system. The structure of the welding machine in this system will be described below with reference to the accompanying drawings and specific embodiments.

[0075] like Figure 3 The welding machine shown mainly includes a welding torch 2 and a welding machine host 1. The welding machine host 1 provides energy supply and parameter control functions for the welding torch 2. The edge computing module 3 is integrated in the welding machine host 1 and can be powered by the power supply inside the welding machine host 1. The image acquisition module 4 is arranged at the welding torch 2 near the welding operation position to acquire real-time images of the molten pool or weld seam on the working surface.

[0076] Current mainstream automated welding machines also include a wire feeding mechanism 5 and an actuator 6. The wire feeding mechanism 5 is controlled by the welding machine host 1, with its wire feeding speed and other parameters controlled by a separate control system or by the welding machine host 1, with its running trajectory and speed parameters controlled by the host system. The welding torch 2 is mounted on the actuator 6 and is driven by the actuator to perform the welding operation. As mentioned earlier, to facilitate the placement of the image acquisition module 4, the actuator 6 can be considered part of the welding machine. When there is not enough space in the welding torch 2 to integrate the image acquisition module 4, the image acquisition module can be integrated into the execution end of the actuator 6 near the welding torch to collect welding information in real time. The actuator 6 here can adopt a design such as... Figure 3 The robotic arm shown, or other automated equipment, such as welding carriages, gantry robots, etc., is a device that can drive a welding torch to perform welding along a specific path.

[0077] The wire feeding mechanism 5 and the welding torch 2 are connected to the welding machine host 1 via cables. When the image acquisition module 4 and the edge computing module 3 transmit data via wired connection, a data cable should also be connected between the image acquisition module 4 and the edge computing module 3. This data cable can be integrated into the cable harness for joint arrangement. Alternatively, the image acquisition module 4 and the edge computing module 3 can be wirelessly connected via wireless transmission.

Claims

1. A welding video monitoring system, characterized in that: This includes a welding machine that integrates an image acquisition module and an edge computing module, as well as a main aggregation node module and a monitoring terminal module; The image acquisition module is used to acquire real-time welding images, and the edge computing module is used to receive and process video stream data from the image acquisition module. The main aggregation node module and the edge computing module receive data from the edge computing module and transmit it to the monitoring terminal module via wireless transmission. The monitoring terminal module receives data from the main aggregation node module and displays real-time welding images.

2. The welding video monitoring system according to claim 1, characterized in that: The image acquisition module and the edge computing module use wireless data transmission, and the edge computing module has a dual-band selectable communication function of 2.4G / 5.8G.

3. The welding video monitoring system according to claim 1, characterized in that: The image acquisition module and the edge computing module use wired data transmission.

4. The welding video monitoring system according to claim 1, characterized in that: The edge computing module optimizes the bitrate, resolution, and I-frame interval of the video stream data.

5. A welding video monitoring system according to claim 1 or 4, characterized in that: At least two image acquisition modules are provided, and the edge computing module merges and aggregates the acquired video stream data into one stream.

6. A welding video monitoring system according to claim 1, characterized in that: The edge computing module and the main aggregation node module use WiFi or 4G / 5G transmission methods, and the transmission link can be automatically switched. The edge computing module has local storage function, and the main aggregation node module and the video surveillance terminal module are connected by wired or wireless means.

7. A welding video monitoring method for multiple welding machines based on the system of claim 1, characterized in that, This can be achieved through the following steps: S1. The edge computing module integrated in each welding machine receives video stream data from the image acquisition module in its respective welding machine; S2. Each edge computing module processes the acquired video stream data; S3. Each edge computing module wirelessly transmits the processed video stream data to the main aggregation node module; S4. The main aggregation node module processes the data received from multiple edge computing modules and transmits it to the monitoring terminal module; S5. The monitoring terminal module processes the received data and presents the welding status in real time.

8. A welding video monitoring method according to claim 7, characterized in that, When a welding machine is equipped with at least two image acquisition modules, the edge computing module in the welding machine performs combining and aggregation processing on the video stream data.

9. A welding video monitoring method according to claim 7 or 8, characterized in that, The edge computing module optimizes the bitrate, resolution, and I-frame interval of the video stream data, and stores the video stream data locally.

10. A welding video monitoring method according to claim 7, characterized in that, The edge computing module establishes an independent communication channel with each image acquisition module in its welding machine.

11. A welding video monitoring method according to claim 7, characterized in that, The edge computing module and the main aggregation node module use WiFi or 4G / 5G transmission methods, and the transmission link can be automatically switched.

12. A welding video monitoring method according to claim 7, characterized in that, When a new welding machine is added, the edge computing module in the welding machine establishes a wireless communication connection with the main aggregation node module.

13. A welding machine based on the welding monitoring system of claim 1, comprising a welding machine host (1) and a welding torch (2), wherein the welding torch (2) is wiredly connected to the welding machine host (1), and the welding machine host (1) supplies power to the welding torch (2) and is used to adjust the welding parameters of the welding torch (2), characterized in that: An edge computing module (3) is integrated in the welding machine host (1), and an image acquisition module (4) is integrated on the welding gun (1).

14. The welding machine according to claim 13, characterized in that: It also includes a wire feeding mechanism (5) and an execution mechanism (6). The welding gun (1) is mounted on the execution mechanism (6). The wire feeding mechanism (5) is connected to the welding machine host and the wire feeding speed is controlled by the welding machine host (1).

15. The welding machine according to claim 13, characterized in that: The image acquisition module (3) and the edge computing module (3) are connected by wired or wireless communication.

16. The welding machine according to claim 13, characterized in that: At least two image acquisition modules (3) are integrated on the welding torch (2).

Images