A Cloud-Edge Collaborative Intelligent Monitoring and Analysis System and Method for Coal Mine Gas Drainage

Abstract

This invention discloses an intelligent monitoring and analysis system and method for coal mine gas extraction based on cloud-edge collaboration. The system includes a sensing device layer, an edge platform, and a cloud platform. The sensing device layer collects sensor and video data. The edge platform deploys a coal mine central video recognition module and an edge application service module for local real-time monitoring and AI video recognition. The cloud platform deploys a cloud application service module for multi-mine data aggregation and global analysis. The cloud platform and edge platform are connected via a VPN network, and the SymmetricDS data synchronization tool is used to achieve bidirectional data synchronization between the cloud MySQL database and the edge MySQL database. Through this cloud-edge collaborative architecture, the edge platform independently supports local business operations during network outages, ensuring production continuity. Domain-driven design is used to decompose each functional module into microservices, enabling flexible responses to the personalized needs of different coal mine business forms.

Description

Technical Field

[0001] This invention relates to the field of coal mine safety monitoring technology, specifically to a cloud-edge collaborative intelligent monitoring and analysis system and method for coal mine gas extraction. Background Technology

[0002] Currently, coal mine gas drainage is a crucial link in ensuring safe production in coal mines. During the drainage process, it is necessary to manage massive amounts of data, including mine overview, coal seam information, working face information, drainage pipelines, and borehole data, and to monitor in real time sensor data such as gas concentration, flow rate, temperature, and negative pressure in the pipelines. At the same time, video surveillance is used to identify key operational behaviors such as drilling.

[0003] Existing gas extraction and monitoring systems mostly adopt a single cloud or local deployment mode; however, when the network is unstable or disconnected, the local system cannot synchronize with the cloud, resulting in data silos and operational inconsistencies. In addition, pure cloud solutions cannot meet the edge computing requirements of low latency and high real-time performance, such as video AI recognition; the deployment, upgrade, and maintenance of edge applications are complex, and reliability is difficult to guarantee. Therefore, there is a need for a holistic solution that can integrate the powerful computing and storage capabilities of the cloud with the low-latency processing capabilities of the edge, while ensuring data consistency and high system availability. Summary of the Invention

[0004] To address this, the present invention provides a cloud-edge collaborative intelligent monitoring and analysis system and method for coal mine gas extraction, in order to solve the problems of inconsistent cloud-edge data and complex edge computing operation and maintenance in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A cloud-edge collaborative intelligent monitoring and analysis system for coal mine gas extraction includes a cloud platform, an edge platform, and a sensing device layer.

[0007] The sensing device layer is used to collect real-time data during the gas extraction process. It includes various acquisition sensors installed underground in the coal mine and video monitoring equipment. The various acquisition sensors transmit data to the video monitoring equipment, and then the video monitoring equipment distributes the data to the edge platform.

[0008] The edge platform is deployed on local servers in each coal mine and is responsible for local processing, real-time monitoring and video recognition of the gas extraction data of the coal mine. The edge platform is equipped with an edge MySQL database, an edge Redis caching service, a MinIO object storage service, and integrates a coal mine central video recognition module.

[0009] The edge platform also includes an edge application service module; when the VPN network with the cloud platform is interrupted, the edge application service module independently supports all business operations in the local coal mine.

[0010] The cloud platform is deployed on a central cloud server and is responsible for the centralized management, analysis and display of gas extraction data from multiple coal mines, and provides managers with a unified data view and decision support. The cloud platform is equipped with a cloud MySQL database, a cloud Redis caching service, a cloud RabbitMQ message queue service and Huawei Cloud OBS object storage service.

[0011] The cloud platform also has a cloud application service module, which is responsible for processing business logic and providing an operation interface for PC users.

[0012] The cloud platform and the edge platform are connected via a VPN network; the data synchronization module uses a data synchronization tool to achieve bidirectional data synchronization between the cloud MySQL database and the edge MySQL database.

[0013] Furthermore, the coal mine central video recognition module of the edge platform acquires video data from video surveillance equipment, runs a trained and optimized AI deep learning model, performs real-time recognition of drilling behavior in the video footage, outputs the recognized video in real time and saves it as a video file, and transmits the recognition results to the edge application service module.

[0014] Furthermore: The internal software architecture of the cloud application service module and the edge application service module follows the domain-driven design principle and is layered. The specific layer and project dependency relationships of the microservice modules are as follows:

[0015] The Domain.Shared project, including constants, enumerations and other shared objects, does not depend on other sub-projects, while other sub-projects depend on the Domain.Shared project directly or indirectly.

[0016] The Domain project is a sub-project of the system's domain, including aggregate roots, entities, value objects, domain services, and repository interfaces, and depends on the Domain.Shared project.

[0017] The Application.Contracts project, which includes application service interfaces and application-layer data transfer objects, depends on the Domain.Shared project.

[0018] The Application project includes the implementation of application service interfaces in the Application.Contracts project, and depends on the Application.Contracts project and the Domain project;

[0019] The HttpApi project provides common API interfaces for different microservices and serves as the base class for all projects that provide external interface services. It depends on the Application.Contracts project.

[0020] The WebApi project provides an external API interface, responsible for receiving and responding to user requests, and depends on the HttpApi project and the Application.Contracts project.

[0021] Furthermore: the static files saved by the system are stored on the edge server and in the cloud OBS, including PNG images, doc documents, xls documents and text; the cloud MySQL database and the edge MySQL database store structured data; the cloud Redis caching service and the edge Redis caching service store key-value format data.

[0022] Furthermore: the edge application service module includes a coal mining face management module, which is used to manage the basic information and mining progress of the coal mining face, update the information of the fault and collapse column in front in real time, and automatically calculate the information of the adjacent layer based on the Ki value;

[0023] The specific method for automatically calculating adjacent layers is as follows:

[0024] (1) Query the list of all coal seams and exclude the coal seam to which the current working face belongs;

[0025] (2) Traverse the Ki values ​​of the coal seam to be judged, find the first standard Ki value that is greater than the Ki value of the coal seam to be judged in the Ki value standard table, and obtain the distance value of the upper adjacent layer, the distance value of the gently dipping layer and the distance value of the steeply dipping layer corresponding to the standard Ki value;

[0026] (3) Obtain the position of the coal seam to be determined relative to the mining layer. If it is the upper adjacent layer and the distance from the mining layer is less than or equal to the upper adjacent layer value corresponding to the standard Ki value, it is determined to be the adjacent layer. If it is the lower adjacent layer, continue to determine the dip angle type. If it is gently inclined and the distance from the mining layer is less than or equal to the gently inclined value corresponding to the standard Ki value, or if it is steeply inclined and the distance from the mining layer is less than or equal to the steeply inclined value corresponding to the standard Ki value, it is determined to be the adjacent layer.

[0027] (4) The coal mining face management module automatically updates the list of adjacent layers and displays it on the interface.

[0028] Furthermore, the edge application service module includes a drainage pipeline configuration module, which is used to add and bind drainage pipelines in the coal mine working face, fill in basic pipeline information and bind corresponding monitoring sensors, and support the drawing of drainage system diagrams. Users can drag and drop pipeline nodes onto the canvas and establish connections between pipelines by connecting lines. The system automatically records the position coordinates of each pipeline node and reconstructs the pipeline system diagram based on these coordinates in the subsequent real-time monitoring screen of the mine.

[0029] After the sensor is attached to the pipeline, the system calculates the sampling flow rate every n seconds in the background, where n≥1; the calculation formula is: ;

[0030] in, This refers to the pure flow rate extracted under standard conditions. The operating flow rate is measured in real time by the flow sensor. The methane concentration is measured in real time by a gas concentration sensor; This is the comprehensive correction factor used to correct the operating flow rate to the standard flow rate.

[0031] Furthermore: the edge application service module includes a sampling pipeline monitoring module, which includes a real-time monitoring sub-function and a pipeline listing sub-function;

[0032] The real-time monitoring sub-function displays the real-time monitoring data of all sensors in the form of cards or tables. The real-time monitoring data includes sensor name, sensor location, real-time monitoring value, maximum value of the day, minimum value of the day, average value compared to yesterday, fluctuation changes and status. When the user clicks on any monitoring value, the system pops up a curve window to display the change curve of the sensor over the past m minutes, where m≥1.

[0033] The pipeline list sub-function groups pipelines and displays the real-time values ​​of all sensors attached to each pipeline. After a user selects a pipeline, the system lists the real-time data of the methane sensor, temperature sensor, flow sensor, and negative pressure sensor attached to that pipeline, as well as the real-time calculated value of the pipeline's extraction flow rate.

[0034] Furthermore: The SymmetricDS data synchronization engine deployed and configured in the data synchronization module monitors data change logs in real time;

[0035] When data in the edge MySQL database is modified or new records are added by the user, the SymmetricDS data synchronization engine captures the change and synchronizes it to the corresponding table in the cloud MySQL database via the VPN network;

[0036] Conversely, when data in the cloud-based MySQL database is updated, the change is also synchronized to the edge MySQL database.

[0037] Users can freely choose to access the cloud platform via a PC browser or the local edge platform based on their location and network conditions. After logging in, the data presented by the two systems is consistent, and users can perform data entry, query, and approval operations on either end.

[0038] To achieve the above objectives, the present invention also provides a cloud-edge collaborative intelligent monitoring and analysis method for coal mine gas extraction, comprising the following steps:

[0039] Step S1 involves real-time data collection via methane sensors, temperature sensors, flow sensors, and negative pressure sensors deployed underground in the coal mine, and transmission to the edge platform via the industrial ring network; simultaneously, video surveillance equipment deployed in key areas of the drilling site pushes high-definition video streams to the edge platform.

[0040] Step S2: The acquired video stream data is decoded and preprocessed in the edge platform, and a trained and optimized AI deep learning model is run to identify drilling behavior in the video in real time; the structured data and labeled short video files generated by the identification are temporarily stored on the edge; the real-time sensor data and AI identification results are displayed and queried by the local user interface through the API interface of the edge application service module.

[0041] Step S3: Deploy and configure the data synchronization engine on the MySQL database instances at the edge and in the cloud. The data synchronization engine monitors the data change log in real time. When data in the edge database is modified or added by a user, the change is captured and reliably synchronized to the corresponding table in the cloud database via VPN network security. Conversely, changes to data in the cloud will also be synchronized to the edge.

[0042] Step S4: Perform global analysis and management in the cloud system. The cloud application service module aggregates full business data and monitoring data from one or more coal mines, performs cross-mine, long-term data aggregation, statistical analysis and trend prediction; and automatically executes business logic, including the calculation of adjacent layers of Ki value of working face, according to preset algorithms, to provide decision support for gas extraction management in the whole mining area.

[0043] In step S5, users can freely choose to access the cloud system via a PC browser or the local edge platform, depending on their location and network conditions. After logging in, users can perform data entry, query, and approval operations on either end.

[0044] Furthermore, in step S2, the real-time identification of drilling behavior in the video footage specifically includes: identifying the start and end of the drilling operation and the number of drill rods.

[0045] This invention has the following advantages: By constructing a three-layer architecture of cloud platform, edge platform, and sensing device layer, it effectively solves the problems of data silos and insufficient collaboration in the existing single deployment mode; by using SymmetricDS to achieve bidirectional real-time synchronization of cloud and edge databases, it ensures that data is not lost when the network is interrupted or restored, and users can obtain a consistent data view and operating experience regardless of which end they access, meeting the stringent requirements of coal mine production for high data availability; this invention adopts Domain-Driven Design (DDD) for microservice decomposition, making the responsibilities of each functional module clear and independent, which can flexibly adapt to the personalized needs of different coal mine business forms, significantly improving the maintainability and scalability of the system.

[0046] Other features and advantages of the present invention will be set forth in the following description. Attached Figure Description

[0047] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0048] Figure 1 This is a system block diagram of a cloud-edge collaborative intelligent monitoring and analysis system for coal mine gas extraction provided in an embodiment of this application.

[0049] Figure 2 This is a schematic diagram of the architecture layering and project dependency relationships in the cloud-edge collaborative intelligent monitoring and analysis system for coal mine gas extraction of the present invention.

[0050] Figure 3 This is a schematic diagram of the system development view hierarchy and calling relationship of the present invention.

[0051] Figure 4 This is a schematic diagram of the system application architecture of the present invention.

[0052] Figure 5 This is a schematic diagram of the data structure in the system of the present invention.

[0053] Figure 6 A flowchart of the intelligent monitoring and analysis method for coal mine gas extraction based on cloud-edge collaboration provided in the embodiments of this application.

[0054] Figure 7 This is a flowchart illustrating the automatic calculation of adjacent layers in the coal mining face management module of this application embodiment. Detailed Implementation

[0055] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Those skilled in the art can make some non-essential improvements and adjustments to the present invention based on the above-described content.

[0056] Please see Figures 1-5 A cloud-edge collaborative intelligent monitoring and analysis system for coal mine gas extraction, comprising a cloud platform, an edge platform, and a sensing device layer;

[0057] The sensing device layer is used to collect real-time data during the gas extraction process. The sensing device layer includes methane sensors, temperature sensors, flow sensors, negative pressure sensors, and AIVAT hard disk recorders installed underground in the coal mine. The sensors transmit data to the AIVAT hard disk recorder via industrial Ethernet or RS485 bus, and the AIVAT hard disk recorder then distributes the data to the edge platform.

[0058] The cloud platform is responsible for the centralized management, analysis and display of gas extraction data from multiple coal mines, and provides managers with a unified data view and decision support;

[0059] The cloud platform is deployed on a central cloud server, specifically using the Huawei Cloud platform. The cloud platform includes a cloud-based MySQL database, a cloud-based Redis caching service, a cloud-based RabbitMQ message queue service, and a Huawei Cloud OBS object storage service.

[0060] In addition, the cloud platform also has a cloud application service module, which is the core of business processing. It is developed based on the Java platform and built using the front-end and back-end separation and domain-driven design (DDD) pattern. It is responsible for handling complex business logic, such as multi-mine data aggregation, historical trend analysis, report generation, etc., and provides an operation interface for PC users. In one embodiment, this module meets the performance constraint that the response time of all API interfaces does not exceed 2 seconds.

[0061] The edge platform is responsible for local processing, real-time monitoring, and video recognition of the gas extraction data of this coal mine; the coal mine central video recognition module of the edge platform obtains video data from the AIVAT hard disk recorder, outputs the recognized video in real time and saves it as a video file, and transmits the recognition results to the edge application system.

[0062] The edge platform is deployed on local servers in each coal mine; the edge platform is equipped with an edge MySQL database, an edge Redis caching service, a MinIO object storage service, and integrates a coal mine central video recognition module;

[0063] In addition, the edge platform also has an edge application service module, which is fully equivalent to the cloud application service module in function. Under normal circumstances, the edge application service module serves as an extension of the cloud service. When the VPN network with the cloud is interrupted, the edge application service module can independently support all local business operations of the coal mine, ensuring that production is not affected.

[0064] The cloud platform and the edge platform are connected via a VPN network to ensure the security and integrity of data transmission; the data synchronization module uses the SymmetricDS data synchronization tool to achieve bidirectional data synchronization between the cloud MySQL database and the edge MySQL database; users can perform all operations through the cloud system or the edge platform, with no operational differences between the two ends.

[0065] See Figure 2 In this embodiment, to meet the constraints of potentially inconsistent business forms across different coal mines and to improve the maintainability and scalability of the system, the internal software architecture of the cloud application service module and the edge application service module strictly follows the DDD (Decentralized Development) principle for layering; the layer and project dependency relationships of their microservice modules are as follows:

[0066] The Domain.Shared project contains constants, enumerations, and other shared objects. For example, the sensor type enumeration (TransducerType) and the alarm status enumeration (AlarmStatus) are both located in this project. The Domain.Shared project does not depend on other sub-projects, while other sub-projects depend on the Domain.Shared project directly or indirectly.

[0067] The Domain project is a sub-project of the system's domain, containing aggregate roots, entities, value objects, domain services, and repository interfaces. For example, the Pipeline Aggregate, WorkingFace Entity, CoalSeam Value Object, PipelineManagerService, and PipelineRepository are all placed in this project. The Domain project depends on the Domain.Shared project.

[0068] The Application.Contracts project contains application service interfaces and application-layer data transfer objects (DTOs); for example, the pipeline application service interface (IPipelineAppService) and the pipeline creation data transfer object (PipelineCreationDto) are both placed in this project; the Application.Contracts project depends on the Domain.Shared project.

[0069] The Application project contains the implementations of application service interfaces from the Application.Contracts project. For example, the Pipeline application service implementation class (PipelineAppService) is placed in this project and is responsible for calling domain objects to execute specific application logic. The Application project depends on the Application.Contracts project and the Domain project.

[0070] The HttpApi project provides common API interfaces for different microservices and base classes for all projects that provide external interface services. For example, the BaseController base class is placed in this project, providing unified exception handling, logging, and permission verification functions for all API controllers. The HttpApi project depends on the Application.Contracts project.

[0071] The Web API project provides external API interfaces and is responsible for receiving and responding to user requests. After receiving an HTTP request, the API controller converts the request parameters into the corresponding DTO object, calls the application service in the Application layer to execute specific business logic, and returns the execution result to the client.

[0072] The WebAPI project depends on the HttpAPI project and the Application.Contracts project.

[0073] This architecture completely decouples domain logic from application logic and the interface layer, enabling the system to flexibly respond to changes in business forms across different coal mines. When it is necessary to adjust a form field in a module, only the entities and DTOs within that domain need to be modified, thus strictly limiting the scope of impact.

[0074] like Figure 3 The development view shown further reveals the organizational structure of the above architecture at the actual compilation and deployment level; the system is divided into layers such as User Interface, Application Contracts, Application, and Domain, each with a clear physical path and calling relationship; this provides the development team with a unified coding standard and module positioning.

[0075] See Figure 4 From the perspective of application architecture, the vertical relationship of the system from the sensing device layer to the application layer is described. The bottom sensing device layer supports AIVAT video equipment and various gas sensors. Their data are aggregated to the platform layer (i.e., the cloud platform and edge platform of this invention). After data processing and business logic processing, the application layer finally supports the various functions of the "gas extraction system dedicated software".

[0076] See Figure 5 The data structure diagram shown uses MySQL as the relational database. Since the system generates relatively small amounts of data daily, there's no need to build a database cluster or perform special database optimizations. Because the system uses a cloud-edge architecture, data needs to be synchronized bidirectionally between the cloud and the edge. Static files are stored on the edge server and in the cloud's OBS. Furthermore, data storage formats include static files such as PNG images, DOC documents, XLS documents, and text. MySQL stores structured data, and Redis stores key-value data.

[0077] See Figure 6 A cloud-edge collaborative intelligent monitoring and analysis method for coal mine gas extraction includes the following steps:

[0078] S1, Data Acquisition: Various sensors deployed underground in coal mines (such as methane, temperature, flow, and negative pressure sensors) collect data in real time and transmit it to the edge platform through the industrial ring network; at the same time, AIVAT video monitoring equipment deployed in key areas such as drilling sites pushes high-definition video streams to the edge platform.

[0079] S2, Edge Computing: Executed in the "Coal Mine Hub" module of the edge platform;

[0080] First, the acquired video stream data is decoded and preprocessed;

[0081] Then, the trained and optimized AI deep learning model is run to identify drilling behavior in the video footage in real time, such as the start and end of drilling operations and the number of drill rods. In this embodiment, under certain lighting and working conditions, the drilling behavior recognition rate is required to reach more than 95%. The structured data and labeled short video files generated by the recognition are temporarily stored on the edge.

[0082] Meanwhile, real-time sensor data and AI recognition results are displayed and queried by the local user interface through the API interface of the edge application service module;

[0083] In the sampling pipeline monitoring module, users can view the real-time values, daily maximum and minimum values, and fluctuation status of all sensors attached to each pipeline, and can view the real-time change curves for the past 5 minutes by clicking.

[0084] S3, Data Synchronization: This step is crucial for achieving cloud-edge collaboration and data consistency. In this embodiment, a SymmetricDS data synchronization engine is deployed and configured on MySQL database instances at both the edge and cloud. This SymmetricDS data synchronization engine monitors data change logs (such as binlog) in real time. When sensor configuration information in the edge database is modified by the user, or a new borehole log record is added, the SymmetricDS data synchronization engine captures the change and reliably and securely synchronizes it to the corresponding table in the cloud database via VPN.

[0085] Conversely, when the cloud administrator updates the Ki value parameter of a certain work surface, the changed Ki value parameter will also be synchronized to the edge; this two-way synchronization mechanism ensures that any operation performed on one end will eventually be reflected on the other end, thereby achieving the goal of no difference between the two ends.

[0086] S4, Cloud Analysis and Management Step: Global analysis and management are performed in the cloud system. The cloud application service module aggregates the full business data and monitoring data of one or more coal mines, performs cross-mine, long-term data aggregation, statistical analysis and trend prediction, and automatically executes complex business logic, including the calculation of adjacent layers of Ki value of working face, according to preset algorithms, to provide decision support for gas extraction management in the whole mining area.

[0087] The automatic calculation of adjacent layers is achieved through the coal mining face management module.

[0088] S5, Hybrid Operation Steps: Users (such as mine technicians and group managers) can freely choose to access the cloud system via a PC browser or the local edge platform, depending on their location and network conditions. After logging in, due to the S3 data synchronization step, the data presented by the two systems is consistent. Users can perform data entry, query, and approval operations on either end. For example, a local technician can access the edge platform underground via an explosion-proof mobile phone or an underground PC to enter the day's tunneling footage data; group managers can access the cloud system from their office to immediately see the summarized tunneling progress for the entire group. This flexible and consistent operation mode greatly improves the efficiency of cross-regional collaborative management in coal mines.

[0089] See Figure 7 In this embodiment, the coal mining face management module is used to manage the basic information and mining progress of the coal mining face, update the information of the fault and collapse column ahead in real time, and automatically calculate the information of the adjacent layer based on the Ki value.

[0090] The interface of the coal mining face management module includes a data display area, a data editing area, and a mining progress report area. The data display area shows all coal mining faces in a list format, with each row corresponding to one face, displaying basic information about the face, its current advance position, and information on the fault and collapse pillar ahead. The data editing area provides forms for users to add or modify face information. The mining progress report area displays the daily mining progress data of the face in tables and charts.

[0091] After a user enters this module, the system automatically queries all coal mining face data for the current coal mine and displays it in the data display area; the user clicks the add button, fills in the face name, coal seam, design length, dip angle, layout and other information in the pop-up dialog box and then saves it; the user clicks the edit button to modify the information of an existing face.

[0092] The specific method for determining adjacent layers in the coal mining face management module is as follows:

[0093] (1) Query the list of all coal seams (excluding the coal seam to which the current working face belongs);

[0094] (2) Traverse the Ki values ​​of the coal seam to be judged, find the first standard Ki value that is greater than the Ki value of the coal seam to be judged in the Ki value standard table, and obtain the distance value of the upper adjacent layer, the distance value of the gently dipping layer and the distance value of the steeply dipping layer corresponding to the standard Ki value;

[0095] (3) Obtain the position of the coal seam to be judged relative to the mining layer. If it is the upper adjacent layer and the distance from the mining layer is less than or equal to the upper adjacent layer value corresponding to the standard Ki value, it is judged as the adjacent layer. If it is the lower adjacent layer, continue to judge the dip angle type. If it is gently inclined and the distance from the mining layer is less than or equal to the gently inclined value corresponding to the standard Ki value, or if it is steeply inclined and the distance from the mining layer is less than or equal to the steeply inclined value corresponding to the standard Ki value, it is judged as the adjacent layer.

[0096] (4) The system automatically updates the list of adjacent layers and displays it on the interface, providing a direct basis for the design of gas extraction radius.

[0097] In this embodiment, some data from the Ki value standard table is shown in Table 1 below:

[0098] Table 1

[0099]

[0100] The values ​​in the table above are illustrative data; the actual Ki values ​​can be adjusted according to the specific conditions of the coal mine.

[0101] The coal mining face management module also supports real-time updates of fault and collapse column information ahead; users can add fault and collapse column information in the face data editing interface, including fields such as collapse column number, depth, major axis length, minor axis length, and whether it is filled with water; the system will associate and store this information with the face and display it in real time in the face card in the data display area.

[0102] After the working face data is edited, users can enter the "Recovery Footage Report" page to fill in the daily recovery footage data; the system records the daily footage data by date and automatically calculates the daily average footage, ten-day average footage and monthly average footage.

[0103] The data structures of this module include a working face table, a fault table, a subsided column table, and an adjacent coal seam table.

[0104] The main fields of the working face table are: uuid (primary key), org_code (organization code), org_name (organization name), coal_seam_id (coal seam ID), workingface_name (working face name), design_length (design length), dip (dip), deploy_method (deployment method), complete_status (complete status), daily_drilling_footage (daily average drilling footage), period_drilling_footage (period average drilling footage), month_drilling_footage (monthly average drilling footage), create_time (creation time), and del_flag (deletion flag).

[0105] The fault table and collapse column table are associated with the working face table via workingface_id;

[0106] The adjacent layer table is associated with the working face table through workingface_id, and records information such as the name of the adjacent coal seam, its relative location, distance from the mining layer, average thickness, coal seam dip angle, and residual gas content.

[0107] Users input or update the absolute gas emission rate of the coal mining face on the interface, and the system calculates the Ki value accordingly.

[0108] The system queries the database for a list of all coal seams except the one belonging to the current working face.

[0109] In addition to coal mining face management, the system also provides extraction pipeline configuration function; in this embodiment, the extraction pipeline configuration module is used to add and bind coal mine working face extraction pipelines, fill in the basic pipeline information and bind the corresponding monitoring sensors, and support the drawing of extraction system diagrams.

[0110] The interface of the extraction pipeline configuration module includes a pipeline configuration list, a data editing area, and an extraction system diagram drawing area. The pipeline configuration list displays all configured pipelines and their hierarchical relationships (main pipeline, branch pipeline, trunk pipeline) in a tree structure or list format. The data editing area provides forms for users to add or modify pipeline information. The extraction system diagram drawing area provides a visual canvas on which users can draw the topology diagram of the pipeline.

[0111] After entering this module, users can view all the extraction pipelines of the current coal mine in the pipeline configuration list; click the add button, fill in the basic information of the pipeline in the pop-up dialog box, including pipeline number, pipeline name, pipeline type (main pipeline / branch pipeline / trunk pipeline), material, length, pipe diameter and other fields, and select the working face or roadway to which the pipeline belongs.

[0112] After the basic pipeline information is filled in, the user needs to bind monitoring sensors to the pipeline. The system provides a sensor selection interface, listing all available methane sensors, temperature sensors, flow sensors, and negative pressure sensors in the current coal mine. The user binds the corresponding methane sensor, temperature sensor, flow sensor, and negative pressure sensor to the pipeline respectively. After binding, the system records the association between the pipeline and the sensor in the pipeline sensor configuration table (pipeline_sensor_config).

[0113] In this embodiment, after the sensor is attached to the pipeline, the system calculates the extraction flow rate in the background every 30 seconds; the formula for calculating the extraction flow rate is: ;

[0114] in, This refers to the pure flow rate extracted under standard conditions. The operating flow rate is measured in real time by the flow sensor. The methane concentration is measured in real time by a gas concentration sensor; This is the comprehensive correction factor used to correct the operating flow rate to the standard state flow rate;

[0115] Comprehensive correction coefficient Determined by the following formula:

[0116] ;

[0117] in, Local atmospheric pressure; This refers to the gas pressure gauge reading inside the pipeline. Standard atmospheric pressure; The absolute temperature under standard conditions; The absolute temperature of the gas inside the pipeline;

[0118] This extraction flow rate value is used to calculate the gas emission rate at the working face where the pipeline is located.

[0119] After adding the pipeline data, users can enter the "Drawing the Extraction System Diagram" page to draw the extraction pipeline diagram. On the drawing page, users can drag and drop pipeline nodes onto the canvas and establish connections between pipelines by connecting lines. The system automatically records the position coordinates (x-axis, y-axis, z-axis) of each pipeline node and reconstructs the pipeline system diagram based on these coordinates in the subsequent real-time mine monitoring screen.

[0120] The main fields of the pipeline sensor configuration table are: uid (primary key), pipeline_id (pipeline ID), transducer_code (sensor number), transducer_type (sensor type: methane / temperature / flow / negative pressure), transducer_point (sensor location), transducer_name (sensor type name), analog_unit (unit), analog_lower (lower limit of analog range), analog_high (upper limit of analog range), analog_alarm_high (upper limit of alarm), analog_alarm_lower (lower limit of alarm), analog_power_off_lower (lower limit of power failure), analog_power_off_high (upper limit of power failure), analog_power_lower (lower limit of power restoration), analog_power_high (upper limit of power restoration), flow_rate (flow rate type), pressure_type (pressure type), local_pressure (local atmospheric pressure), create_time (creation time), and del_flag (deletion flag).

[0121] In this embodiment, the extraction pipeline monitoring module includes two sub-functions: real-time monitoring and pipeline listing.

[0122] The real-time monitoring sub-function displays the real-time monitoring data of all sensors in the form of cards or tables. Each sensor card or row displays the following information: sensor name, sensor location, real-time monitoring value, maximum value of the day, minimum value of the day, average value compared to yesterday, fluctuation changes, and status (normal / alarm / power failure). When the user clicks on any monitoring value, the system pops up a graph window to display the change curve of that sensor over the past 5 minutes.

[0123] The pipeline list sub-function groups pipelines and displays the real-time values ​​of all sensors attached to each pipeline. After the user selects a pipeline, the system lists the real-time data of the methane sensor, temperature sensor, flow sensor, and negative pressure sensor attached to that pipeline. The pipeline list also displays the real-time calculated value of the extraction flow rate for that pipeline.

[0124] The data for the extraction pipeline monitoring module comes from real-time and historical sensor data from the data governance platform. The system calls the services of the data governance platform through an HTTP interface to obtain the real-time monitoring values ​​and historical monitoring data of the sensors.

[0125] This invention enables bidirectional synchronization between cloud and edge databases, ensuring that data is not lost during network interruption or recovery. Users can obtain a consistent data view and operating experience regardless of whether they access the cloud or edge platform, thus meeting the stringent requirements for high data availability in coal mine production environments.

[0126] Furthermore, this invention also uses Domain-Driven Design (DDD) to decompose the application system into microservices, making the responsibilities of each functional module (such as coal seam information, coal mining face management, etc.) clear and independent. This enables the system to quickly respond to personalized needs such as inconsistent business forms in different coal mines, and facilitates the continuous iteration and expansion of system functions.

[0127] 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, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cloud-edge collaborative intelligent monitoring and analysis system for coal mine gas extraction, characterized in that, This includes cloud platforms, edge platforms, and sensing device layers; The sensing device layer is used to collect real-time data during the gas extraction process. It includes various acquisition sensors installed underground in the coal mine and video monitoring equipment. The various acquisition sensors transmit data to the video monitoring equipment, and then the video monitoring equipment distributes the data to the edge platform. The edge platform is deployed on local servers in each coal mine and is responsible for local processing, real-time monitoring and video recognition of the gas extraction data of the coal mine. The edge platform is equipped with an edge MySQL database, an edge Redis caching service, a MinIO object storage service, and integrates a coal mine central video recognition module. The edge platform also includes an edge application service module; when the VPN network with the cloud platform is interrupted, the edge application service module independently supports all business operations in the local coal mine. The cloud platform is deployed on a central cloud server and is responsible for the centralized management, analysis and display of gas extraction data from multiple coal mines, and provides managers with a unified data view and decision support. The cloud platform is equipped with a cloud MySQL database, a cloud Redis caching service, a cloud RabbitMQ message queue service and Huawei Cloud OBS object storage service. The cloud platform also has a cloud application service module, which is responsible for processing business logic and providing an operation interface for PC users. The cloud platform and the edge platform are connected via a VPN network; the data synchronization module uses a data synchronization tool to achieve bidirectional data synchronization between the cloud MySQL database and the edge MySQL database.

2. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The coal mine central video recognition module of the edge platform acquires video data from video monitoring equipment, runs a trained and optimized AI deep learning model, identifies drilling behavior in the video in real time, outputs the identified video in real time and saves it as a video file, and transmits the recognition result to the edge application service module.

3. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The internal software architecture of the cloud application service module and the edge application service module follows the domain-driven design principle and is layered. The specific layer and project dependency relationships of the microservice modules are as follows: The Domain.Shared project, including constants, enumerations and other shared objects, does not depend on other sub-projects, while other sub-projects depend on the Domain.Shared project directly or indirectly. The Domain project is a sub-project of the system's domain, including aggregate roots, entities, value objects, domain services, and repository interfaces, and depends on the Domain.Shared project. The Application.Contracts project, which includes application service interfaces and application-layer data transfer objects, depends on the Domain.Shared project. The Application project includes the implementation of application service interfaces in the Application.Contracts project, and depends on the Application.Contracts project and the Domain project; The HttpApi project provides common API interfaces for different microservices and serves as the base class for all projects that provide external interface services. It depends on the Application.Contracts project. The WebApi project provides an external API interface, responsible for receiving and responding to user requests, and depends on the HttpApi project and the Application.Contracts project.

4. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The static files saved by the system are stored on the edge server and in the cloud OBS. The static files include PNG images, doc documents, xls documents and text; the cloud MySQL database and the edge MySQL database store structured data. Cloud-based Redis caching services and edge-based Redis caching services store key-value format data.

5. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The edge application service module includes a coal mining face management module, which is used to manage the basic information and mining progress of the coal mining face, update the information of the fault and collapse column in front in real time, and automatically calculate the information of the adjacent layer based on the Ki value. The specific method for automatically calculating adjacent layers is as follows: (1) Query the list of all coal seams and exclude the coal seam to which the current working face belongs; (2) Traverse the Ki values ​​of the coal seam to be judged, find the first standard Ki value that is greater than the Ki value of the coal seam to be judged in the Ki value standard table, and obtain the distance value of the upper adjacent layer, the distance value of the gently dipping layer and the distance value of the steeply dipping layer corresponding to the standard Ki value; (3) Obtain the position of the coal seam to be determined relative to the mining layer. If it is the upper adjacent layer and the distance from the mining layer is less than or equal to the upper adjacent layer value corresponding to the standard Ki value, it is determined to be the adjacent layer. If it is the lower adjacent layer, continue to determine the dip angle type. If it is gently inclined and the distance from the mining layer is less than or equal to the gently inclined value corresponding to the standard Ki value, or if it is steeply inclined and the distance from the mining layer is less than or equal to the steeply inclined value corresponding to the standard Ki value, it is determined to be the adjacent layer. (4) The coal mining face management module automatically updates the list of adjacent layers and displays it on the interface.

6. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The edge application service module includes a drainage pipeline configuration module, which is used to add and bind drainage pipelines in coal mine working faces, fill in basic pipeline information and bind corresponding monitoring sensors, and support the drawing of drainage system diagrams. Users can drag and drop pipeline nodes onto the canvas and establish connections between pipelines by connecting lines. The system automatically records the position coordinates of each pipeline node and reconstructs the pipeline system diagram based on these coordinates in the subsequent real-time monitoring screen of the mine. After the sensor is attached to the pipeline, the system calculates the sampling flow rate every n seconds in the background, where n≥1; the calculation formula is: ； in, This refers to the pure flow rate extracted under standard conditions. The operating flow rate is measured in real time by the flow sensor. The methane concentration is measured in real time by a gas concentration sensor; This is the comprehensive correction factor used to correct the operating flow rate to the standard flow rate.

7. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 6, characterized in that, The edge application service module includes a sampling pipeline monitoring module, which includes a real-time monitoring sub-function and a pipeline listing sub-function. The real-time monitoring sub-function displays the real-time monitoring data of all sensors in the form of cards or tables. The real-time monitoring data includes sensor name, sensor location, real-time monitoring value, maximum value of the day, minimum value of the day, average value compared to yesterday, fluctuation changes and status. When the user clicks on any monitoring value, the system pops up a curve window to display the change curve of the sensor over the past m minutes, where m≥1. The pipeline list sub-function groups pipelines and displays the real-time values ​​of all sensors attached to each pipeline. After a user selects a pipeline, the system lists the real-time data of the methane sensor, temperature sensor, flow sensor, and negative pressure sensor attached to that pipeline, as well as the real-time calculated value of the pipeline's extraction flow rate.

8. The intelligent monitoring and analysis system for coal mine gas extraction based on cloud-edge collaboration as described in claim 1, characterized in that, The SymmetricDS data synchronization engine deployed and configured in the data synchronization module monitors data change logs in real time. When data in the edge MySQL database is modified or new records are added by the user, the SymmetricDS data synchronization engine captures the change and synchronizes it to the corresponding table in the cloud MySQL database via the VPN network. Conversely, when data in the cloud-based MySQL database is updated, the change is also synchronized to the edge-based MySQL database. Users can freely choose to access the cloud platform via a PC browser or the local edge platform based on their location and network conditions. After logging in, the data presented by the two systems is consistent, and users can perform data entry, query, and approval operations on either end.

9. A cloud-edge collaborative intelligent monitoring and analysis method for coal mine gas extraction, characterized in that, Includes the following steps: Step S1 involves real-time data collection via methane sensors, temperature sensors, flow sensors, and negative pressure sensors deployed underground in the coal mine, and transmission to the edge platform via the industrial ring network; simultaneously, video surveillance equipment deployed in key areas of the drilling site pushes high-definition video streams to the edge platform. Step S2: The acquired video stream data is decoded and preprocessed in the edge platform, and a trained and optimized AI deep learning model is run to identify drilling behavior in the video in real time; the structured data and labeled short video files generated by the identification are temporarily stored on the edge; the real-time sensor data and AI identification results are displayed and queried by the local user interface through the API interface of the edge application service module. Step S3: Deploy and configure the data synchronization engine on the MySQL database instances at the edge and in the cloud. The data synchronization engine monitors the data change log in real time. When data in the edge database is modified or added by a user, the change is captured and reliably synchronized to the corresponding table in the cloud database via VPN network security. Conversely, changes to data in the cloud will also be synchronized to the edge. Step S4: Perform global analysis and management in the cloud system. The cloud application service module aggregates full business data and monitoring data from one or more coal mines, performs cross-mine, long-term data aggregation, statistical analysis and trend prediction; and automatically executes business logic, including the calculation of adjacent layers of Ki value of working face, according to preset algorithms, to provide decision support for gas extraction management in the whole mining area. In step S5, users can freely choose to access the cloud system via a PC browser or the local edge platform, depending on their location and network conditions. After logging in, users can perform data entry, query, and approval operations on either end.

10. The intelligent monitoring and analysis method for coal mine gas extraction based on cloud-edge collaboration according to claim 9, characterized in that, In step S2, the real-time identification of drilling behavior in the video footage specifically includes: identifying the start and end of the drilling operation and the number of drill rods.

Images