Intelligent visual control recognition system and method for coal mine acquisition working face and electronic equipment

Abstract

The invention provides an intelligent visual control recognition system and method for a coal mine collection working face and electronic equipment, and the system comprises coal mining equipment, and also comprises a sensing module which is arranged on the coal mining equipment and is used for collecting video data of the internal working face of a coal mine; the edge analysis module is in signal connection with the sensing module and is used for receiving the video data, analyzing the video data by adopting a preset algorithm model and judging whether an abnormal working condition occurs or not according to the video data; and the control module is in signal connection with the edge analysis module and is used for adaptively adjusting operation parameters of the coal mining equipment according to an analysis result of the edge analysis module when an abnormal working condition occurs so as to perform linkage control on the coal mining equipment. Abnormal conditions in the coal mining process can be found and processed in time, the coal mine operation safety and the production efficiency are improved, and the manual monitoring cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent visual control and recognition technology for coal mine mining faces, specifically to an intelligent visual control and recognition system and method, and electronic equipment for coal mine mining faces. Background Technology

[0002] As a major energy source in my country, coal safety is of paramount importance. The coal mining face is the core area of ​​coal production and also a dangerous area prone to accidents such as roof falls, gas leaks, and mechanical and transportation malfunctions. Traditionally, monitoring of the working face mainly relies on the following two methods: On the one hand, there is manual inspection and monitoring, relying on personnel such as safety inspectors and electricians to conduct regular and fixed-point inspections. This method has obvious limitations. First, manual inspection cannot achieve 24-hour uninterrupted monitoring, resulting in blind spots and time gaps. Second, the underground environment is harsh, with dim lighting, high dust levels, and strong noise, making inspectors prone to fatigue. Subjective judgment can easily lead to missed inspections and misjudgments, placing personnel in a high-risk environment and failing to fundamentally guarantee personal safety. On the other hand, there is automated monitoring based on traditional sensors. Currently, some mines have deployed monitoring systems composed of various sensors (such as pressure, displacement, and infrared sensors) and PLC controllers. Although these systems have achieved the collection of equipment status data to a certain extent, their perception dimensions are limited. Sensors can usually only monitor a single physical quantity and cannot effectively identify complex working conditions that require visual judgment, such as whether the coal mining machine drum is cutting rock, whether the scraper conveyor chain is broken or tilted, or whether personnel have entered dangerous areas. They lack scene understanding capabilities. For example, when the sensor detects abnormal motor current, the equipment failure has often already occurred, making it impossible to provide early warning. Therefore, there is a lack of a comprehensive, real-time monitoring and intelligent control solution for coal mine mining faces in the existing technology.

[0003] Therefore, the existing technology still needs further development. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide an intelligent visual control and recognition system and method, as well as electronic equipment, for coal mine mining face monitoring. This invention addresses the technical problems of low efficiency, high labor intensity, susceptibility to subjective factors leading to missed detections and misjudgments in traditional coal mine mining face monitoring, and the limitations of simple sensors that can only monitor single physical quantities and cannot perform comprehensive and accurate analysis of complex operating scenarios. The goal is to improve the safety and efficiency of coal mine production and reduce the workload and labor intensity of personnel.

[0005] To achieve the above-mentioned technical objectives, according to a first aspect of the present invention, the present invention provides an intelligent visual control and recognition system for coal mine mining faces, including coal mining equipment, and further comprising: A sensing module, installed on the coal mining equipment, is used to collect video data from the working face inside the coal mine. An edge analysis module, which is signal-connected to the sensing module, is used to receive the video data, analyze the video data using a preset algorithm model, and determine whether an abnormal operating condition has occurred based on the video data. The control module is signal-connected to the edge analysis module and is used to adaptively adjust the operating parameters of the coal mining equipment when abnormal operating conditions occur, based on the analysis results of the edge analysis module, so as to perform linkage control on the coal mining equipment.

[0006] Specifically, the sensing module includes multiple intrinsically safe mining cameras deployed on the coal mining equipment. These cameras are used to collect video data of the working face inside the coal mine from different angles.

[0007] Specifically, the control module includes: The data acquisition layer is connected to the sensing module and is used to perform multi-source data fusion of the video data and existing system data. Then, a preset algorithm model is used to analyze the multi-source data and determine whether any abnormal operating conditions occur. The support service layer is connected to the data acquisition layer by signal, and is used to provide the data acquisition layer with a unified algorithm model and application service support; The business application layer is connected to the data acquisition layer and the support service layer respectively, and is used to adaptively adjust the operating parameters of the coal mining equipment in order to perform linkage control of the coal mining equipment when abnormal operating conditions occur.

[0008] Specifically, the business application layer includes an equipment management module, which is used to customize coal mining equipment linkage rules and alarm rules. When abnormal operating conditions occur, the equipment management module controls the coal mining equipment in accordance with the coal mining equipment linkage rules and issues alarm information according to the alarm rules.

[0009] Specifically, the business application layer also includes an open API interface, which connects to other information platforms in the coal mine and is used to transmit the alarm information.

[0010] Specifically, the edge analysis module includes: The coal mining equipment drum rock cutting identification module is used to identify whether the coal mining equipment drum has cut into the upper or lower rock layers based on the preset algorithm model and the video data collected by the sensing module.

[0011] Specifically, the system also includes a power supply module, which is connected to an intrinsically safe DC regulated power supply via a mining power line and a mining explosion-proof power supply tee box to power the sensing module; the intrinsically safe DC regulated power supply has a built-in backup power supply.

[0012] According to a second aspect of the present invention, an intelligent visual control and recognition method for coal mine mining faces is provided, comprising: S100: Real-time video data of the working face in the coal mine is collected through the sensing module; S200: Receive the video data, analyze the video data using a preset algorithm model, and determine whether an abnormal working condition has occurred based on the video data; S300: When abnormal operating conditions occur, the operating parameters of the coal mining equipment are adaptively adjusted to enable linkage control of the coal mining equipment.

[0013] Specifically, the step of analyzing the video data using a preset algorithm model and determining whether abnormal operating conditions have occurred based on the video data includes: The system uses a pre-defined algorithm model to analyze the video data collected by the sensing module and identify whether the coal mining equipment drum is cutting into the upper or lower rock layers. If the drum is found to be cutting into the upper or lower rock layers, an abnormal working condition is determined to have occurred; otherwise, no abnormal working condition is determined to have occurred.

[0014] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, which, when executed by the processor, implement the above-described intelligent visual control and recognition method for coal mine mining faces.

[0015] Beneficial effects: This invention provides an intelligent visual control and recognition system and method for coal mine working faces. Through integrated visual monitoring and intelligent analysis, it achieves real-time monitoring and automatic response to abnormal working conditions at coal mine working faces, effectively improving the safety and efficiency of coal mining. By utilizing multi-angle video data acquisition and deep learning algorithms, it accurately judges abnormal situations such as rock cutting by the coal mining equipment drum, adjusts equipment operating parameters in a timely manner, avoids production accidents, reduces losses during coal mining, and also reduces the need for manual intervention, providing strong technical support for the intelligent transformation of coal mines. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the composition of the intelligent visual control and recognition system for coal mine data acquisition faces provided in a specific embodiment of the present invention; Figure 2 This is a flowchart of the intelligent visual control and recognition method for coal mine data acquisition working face provided in a specific embodiment of the present invention; Figure 3 This is a general framework diagram of the intelligent visual control and recognition system for coal mine mining faces provided in a specific embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Other similar embodiments obtained by those skilled in the art based on the embodiments in this application without creative effort should all fall within the scope of protection of this application. Furthermore, directional terms mentioned in the following embodiments, such as "up," "down," "left," and "right," are only for reference to the directions in the accompanying drawings; therefore, the directional terms used are for illustrative purposes and not for limiting the invention.

[0018] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0019] Example 1 Please see Figure 1 This embodiment provides an intelligent visual control and recognition system for coal mine mining faces, including coal mining equipment, a sensing module 100, an edge analysis module 200, and a control module 300; The sensing module 100 is installed on the coal mining equipment to collect video data from the working face inside the coal mine. Specifically, the sensing module 100 includes multiple intrinsically safe mining cameras deployed at different locations on the coal mining equipment to collect video data from the working face inside the coal mine from different angles. This multi-angle acquisition method can comprehensively capture the real-time status of the working face, improving the integrity and accuracy of the data.

[0020] Understandably, intrinsically safe mining cameras are camera devices suitable for use in the internal working environment of coal mines. They possess explosion-proof and intrinsically safe characteristics, ensuring safe use in flammable and explosive environments. With high resolution and adjustable focus, they provide clear video data for real-time monitoring of the working face within the coal mine. Coal mining equipment refers to various mechanical devices used for coal mining operations at the working face within a coal mine, including but not limited to coal mining machines and scraper conveyors. The working face within a coal mine refers to the actual workplace in the underground mining area of ​​the coal mine. It is the main site for coal mining and typically includes roadways, goafs, and coal faces. The working face environment is harsh, with dim lighting, high dust and noise levels, making it a key monitoring area for coal mine safety. Video data is a continuous sequence of images captured and recorded by camera equipment. It can be a collection of static images or a dynamic video stream. In this system, video data is used for real-time monitoring and intelligent analysis of the operations at the working face within the coal mine.

[0021] Specifically, the edge analysis module 200 is signal-connected to the sensing module 100 and receives video data from the sensing module 100. The edge analysis module 200 analyzes this video data using a preset algorithm model and determines whether abnormal operating conditions have occurred based on the video data. The edge analysis module 200 includes a coal mining equipment drum cutting rock identification module, which identifies whether the coal mining equipment drum has cut into the upper or lower rock layers based on the preset algorithm model and the video data collected by the sensing module 100. When the drum cuts into the rock layer, the system can promptly detect this abnormal situation, preventing equipment damage and safety accidents.

[0022] Specifically, the control module 300 is connected to the edge analysis module 200 by signal. Based on the analysis results of the edge analysis module 200, when abnormal working conditions occur, the operating parameters of the coal mining equipment are adaptively adjusted to perform linkage control of the coal mining equipment.

[0023] It is understandable that the above-described design in this embodiment not only improves the monitoring efficiency of coal mine production operations but also plays a crucial role in safety. It can provide timely warnings of potential safety hazards, such as equipment malfunctions and personnel violations, thereby enabling effective measures to prevent accidents and ensuring the safety and continuity of coal mine production. Furthermore, the high pixel count and flexible focal length adjustment range of the intrinsically safe mining camera ensure clear and detailed images even in poor lighting conditions underground, providing high-quality raw data for subsequent analysis. Real-time analysis of the video stream using a preset algorithm model allows the system to respond immediately. For example, when abnormal conditions such as the coal mining machine drum cutting into upper or lower rock layers or the scraper conveyor chain breaking and tilting are detected, an alarm mechanism is immediately activated, triggering corresponding emergency procedures, such as adjusting equipment operating parameters or suspending equipment operation. This effectively avoids equipment damage and safety hazards, reducing maintenance costs and risks in coal mine production.

[0024] See Figure 3 In this embodiment, the control module 300 includes a data acquisition layer, a support service layer, and a business application layer, constructing a multi-level intelligent control system designed to monitor and respond to abnormal working conditions at the coal mine face in real time. The data acquisition layer is signal-connected to the sensing module 100, fusing video data with existing system data from multiple sources. A preset algorithm model is then used to analyze the multi-source data and determine whether abnormal working conditions have occurred, such as rock cutting by the coal mining machine's drum or chain breakage and tilting of the scraper conveyor. This multi-source data fusion technology comprehensively analyzes information from different sensors and systems, improving the accuracy of the judgment.

[0025] Furthermore, the support service layer is signal-connected to the data acquisition layer, providing a unified algorithm model and application service support for the data acquisition layer. The support service layer contains various algorithm libraries and service components, among which the application services provide unified basic development capabilities and dependencies for upper-layer applications. The algorithm model enables the unified construction of basic technical capabilities and improves technology reuse, thereby ensuring the efficiency and accuracy of data analysis.

[0026] Furthermore, the business application layer is connected to the data acquisition layer and the support service layer via signals. When abnormal operating conditions occur, the operating parameters of the coal mining equipment are adaptively adjusted to enable coordinated control of the coal mining equipment. For example, when the rock cutting of the coal mining machine drum is detected, the system automatically adjusts the cutting depth or speed of the coal mining machine to avoid equipment damage, and notifies relevant personnel to intervene. This overall architecture not only improves the safety of coal mine production, but also reduces production costs and improves resource utilization through intelligent control of equipment operation.

[0027] Specifically, the business application layer in this embodiment also includes an equipment management module, which is used to customize the linkage rules and alarm rules for coal mining equipment. When abnormal operating conditions occur, the equipment management module controls the coal mining equipment according to the linkage rules and issues alarm information according to the alarm rules. For example, when it is detected that the drum is cutting into the rock strata, the system may automatically adjust the height of the coal mining machine or reduce the cutting speed, and at the same time issue an alarm to remind the operator. The open design of the system in this embodiment allows mine management personnel to independently configure equipment linkage rules and alarm thresholds, enhancing the system's flexibility and adaptability, enabling coal mines to more effectively cope with various emergencies.

[0028] Understandably, based on the aforementioned technical solution, mine management personnel can utilize the equipment management module to customize equipment linkage and alarm rules, thereby achieving intelligent linkage and reverse control of on-site coal mining equipment. This makes the system more aligned with actual operational needs, enhancing its adaptability and practicality. When the edge analysis module 200 in this embodiment identifies abnormal operating conditions through video monitoring, such as the coal mining machine's drum cutting rock or the scraper conveyor's chain breaking and tilting, the equipment management module responds rapidly according to preset rules. On one hand, it coordinates and controls the coal mining equipment, such as adjusting the coal mining machine's operating parameters or stopping it. On the other hand, it issues alarm information according to alarm rules, notifying relevant personnel to handle the situation immediately. This real-time monitoring and automated control based on AI technology greatly improves the safety and efficiency of coal mine production, reduces the risk of production interruptions due to equipment failure, and also reduces the workload of miners, promoting a comprehensive improvement in the level of coal mine intelligence. Furthermore, the alarm information and data analysis results generated by the system are shared with other information platforms at the mine through open API interfaces, achieving cross-platform information integration and application expansion, further optimizing resource scheduling and safety management.

[0029] Furthermore, the business application layer also includes an open API interface, which connects to other information platforms in the coal mine to transmit alarm information. In this way, abnormal situations can be quickly notified to relevant management systems and personnel, enabling rapid response.

[0030] In a preferred embodiment, alarm data generated by this system can be shared in real time to the coal mine safety management platform via an open API interface. This not only enhances the overall safety monitoring capabilities of the coal mine but also enables information exchange and coordinated response between multiple systems. This allows safety management personnel to quickly receive notifications of abnormal situations and take timely measures to prevent potential safety accidents. Furthermore, the open API interface can push equipment operating status and production data to the production scheduling platform, helping the production scheduling department make more rational resource allocation decisions based on real-time operating conditions. This improves coal mine production efficiency and reduces costs. This design of the present invention promotes the intelligentization and integration of coal mine production operations, strengthens the digital management capabilities of coal mines, and provides strong support for safe and efficient coal mine production.

[0031] Preferably, in this embodiment, the edge analysis module 200 includes a coal mining equipment drum rock cutting identification module, which is used to identify whether the coal mining equipment drum has cut into the upper or lower rock layers based on a preset algorithm model and video data collected by the perception module 100.

[0032] In this embodiment, the edge analysis module 200 integrates the function of identifying rock cutting by the coal mining equipment drum. Based on a preset algorithm model and real-time video data from the perception module 100, it can accurately determine whether the coal mining machine drum is touching the upper or lower rock strata. This module is designed based on deep learning technology and, through extensive sample training, achieves highly sensitive identification of rock texture, color, and dynamic changes when the coal mining machine drum contacts the rock strata. In practical applications, when the system detects rock cutting by the drum, it immediately triggers an early warning mechanism, reminding operators to take measures to adjust the working parameters of the coal mining machine or stop the equipment, avoiding equipment damage and improving coal mining efficiency. This immediate response mechanism not only reduces unplanned downtime but also lowers equipment maintenance costs caused by rock cutting, ensuring the continuity and safety of coal mine production. Furthermore, the module's algorithm model can be continuously iterated and optimized. By continuously learning new samples during mine operations, it further improves the accuracy and robustness of identification, enabling the system to better adapt to various complex underground mining environments. Furthermore, the edge analysis module 200 in this embodiment can be extended to identify more types of abnormal situations, such as broken scraper conveyor chains and tilting, providing more comprehensive safety assurance for underground operations. The integrated design of the edge analysis module 200 simplifies the system architecture, reduces data transmission latency, and ensures real-time decision-making, making it an important component of intelligent coal mine management.

[0033] In some specific embodiments, the system also includes a power supply module, which is connected to a mine-use intrinsically safe DC regulated power supply via a mine-use power supply cable and a mine-use explosion-proof power supply tee box to power the sensing module 100. The mine-use intrinsically safe DC regulated power supply has a built-in backup power supply to ensure that the system can continue to operate for a period of time when the main power supply is interrupted, thereby improving the reliability and safety of the system.

[0034] Understandably, the power supply module connects to a mine-use intrinsically safe DC regulated power supply via a mine-use power supply line and a mine-use explosion-proof power supply tee junction box, providing stable power support for the entire sensing module 100. The mine-use intrinsically safe DC regulated power supply has a built-in backup power supply, ensuring that the system can continue to operate for at least 4 hours in the event of an unexpected interruption of the main power supply. This design greatly enhances the reliability and continuity of the system, ensuring that video monitoring and AI analysis functions are unaffected by power supply fluctuations during mine operations. It guarantees the continuous acquisition and analysis of critical data, thereby making the overall performance of the intelligent visual control and recognition system for coal mine working faces more stable and improving the safety management and intelligence level of coal mine production.

[0035] See Figure 3 The working principle of the intelligent visual control and recognition system for coal mine mining faces in this embodiment will be illustrated below with a specific example: This example provides an AI-based intelligent visual control and recognition system for coal mine mining faces, including an intrinsically safe camera for mining, an edge analysis module 200 (AI edge inference server), an industrial ring network, and a control module 300. The control module 300 is divided into a data acquisition layer, a support service layer, and a business application layer. Data Acquisition Layer: Supports six acquisition methods: file data, database data, PLC sensor data, OPC industrial protocol data, interface data, and message queue data. It enables data fusion and analysis with existing systems, and has video access, recording, storage, and forwarding functions, as well as AI-based video analysis functions. Supporting service layer: includes application services and algorithm models. Application services provide unified basic development capabilities and dependencies for upper-layer applications, while algorithm models realize the unified construction of technical basic capabilities and improve technology reuse. Business application layer: Enables centralized management of models, business processes, and permissions. Through a closed-loop process of data collection, storage, local inference, and business application, it ensures flexible deployment of applications and business support. Mine management personnel can customize equipment linkage rules and alarm rule configurations through the equipment management module to achieve on-site equipment linkage and reverse control; Alarm information generated at the mining site can be accessed by other information platforms at the mining site through an open API interface. Two intrinsically safe mining cameras are installed in one set. The cameras have 4 megapixels, a focal length of 2.8-12mm, a horizontal field of view of 104°-38°, a vertical field of view of 54°-21°, and a diagonal field of view of 125°-45°. They are installed using quick-connect interfaces and use intrinsically safe mining data transmission interfaces as data exchange equipment. The data transmission interfaces are customized with 3 optical ports and 7 electrical ports. The devices are connected in series via 40m armored quick-connect optical cables and finally connected to a ring network switch to complete data transmission and interaction. The power supply method is to draw 127V from the integrated protection power supply and then pass it through 4 The 2.5mm² mining power cord is connected to the mining potted and intrinsically safe DC regulated power supply through the mining explosion-proof power supply tee junction box to power the switch and camera. The mining potted and intrinsically safe DC regulated power supply is small in size and light in weight, and has a built-in backup power supply that can support the equipment to operate for no less than 4 hours. The power cord adopts a ready-made quick-connect connection.

[0036] The model for identifying the cutting of upper and lower rock strata by the front and rear drums is integrated into the existing AI application platform. During the data acquisition and training phase, based on the rock cutting time period feedback from underground operations, the rock cutting video is retrieved and played back using a hard disk recorder. The samples are then transmitted to the AI ​​training center to train and learn sample data of normal coal cutting and abnormal rock cutting. The system performs real-time analysis through a camera installed on the support. When it is detected that the coal mining machine drum has cut into the upper and lower rock strata during operation, an alarm is triggered in a timely manner.

[0037] In operation, all parts of the system work collaboratively. The data acquisition layer continuously collects various types of data, and video data is analyzed using AI technology to capture real-time dynamics of the working face. AI-based video analysis constantly monitors the operating status of equipment such as the coal mining machine and scraper conveyor, as well as personnel's work behavior. Once an anomaly is detected, such as abnormal scraper conveyor trajectory or personnel entering a dangerous area, the relevant data is immediately transmitted to the business application layer. The support service layer provides a solid technical guarantee for the stable operation of the entire system. The application service provides development interfaces and a runtime environment for upper-layer business applications, ensuring smooth integration and operation of each functional module. The algorithm model performs in-depth analysis and processing of the collected data, continuously optimizing recognition accuracy and early warning capabilities. For example, through learning from a large amount of historical data, the algorithm model can more accurately predict equipment failures, issue early warnings, and reduce equipment downtime. The business application layer receives the data from the data acquisition layer... Upon receiving abnormal data, the system reacts according to preset equipment linkage and alarm rules. If a broken chain or tilting of the scraper conveyor is detected, the system automatically triggers an alarm, notifying relevant personnel for timely handling. Simultaneously, according to preset rules, it controls the lighting equipment near the scraper conveyor to illuminate, providing a better working environment for maintenance personnel, and suspends the operation of related equipment to prevent further escalation of the fault. Mine management personnel can view equipment status and modify linkage and alarm rules at any time through the equipment management module to adapt to different work scenarios and needs. The intrinsically safe mining camera features high resolution and a wide field of view, capturing images of the working face from all angles, providing clear and accurate image data for AI analysis. The quick-connect interface facilitates equipment installation and maintenance, improving work efficiency. The intrinsically safe mining data transmission interface and 40m armored quick-connect optical cable ensure high-speed and stable data transmission, ensuring no information loss or delay even in complex underground environments.

[0038] Regarding the identification of upper and lower rock strata cut by the front and rear drums of the coal mining machine, the system continuously monitors the working status of the coal mining machine drums. When it is detected that the coal mining machine drums are cutting into the upper or lower rock strata, an alarm signal is immediately issued to remind the operator to adjust the operating parameters of the coal mining machine in a timely manner, avoid excessive wear and tear on the equipment, reduce equipment maintenance costs, and ensure efficient coal mining. The alarm information and various data generated by the system are available for other information platforms at the mine end to call through open API interfaces. The safety alarm information is synchronized to the coal mine safety management platform, which makes it easier for safety management personnel to have a comprehensive understanding of the mine's safety status. The system also provides equipment operation data to the production scheduling platform to provide data support for the formulation and adjustment of production plans, realize intelligent management of the entire coal mine production and operation, improve the safety and efficiency of coal mine production, and reduce production costs.

[0039] In a preferred embodiment, the implementation steps for deep analysis and processing of the collected data by a preset algorithm model deployed in the edge analysis module are as follows: Step 1: Input data Multiple intrinsically safe mining cameras deployed on key equipment such as coal mining machines and hydraulic supports collect multiple real-time video streams, which are high-definition (such as 1080P or 4 million pixels) video frames, continuously capturing the cutting interface between the coal mining machine drum and the coal / rock wall, the running status of the scraper conveyor chain, and the surrounding environment of the equipment. The video streams acquired in real time are preprocessed: The video streams first pass through a preprocessing module to enhance the images. For the low-light environment in the well, adaptive histogram equalization (CLAHE) or a low-light enhancement algorithm based on deep learning is used to improve the visibility of the image. Physical models or convolutional neural networks (CNN) are applied to reduce the interference of dust and water mist on the images. Existing system data, such as data from industrial IoT platforms and PLCs, is integrated through the data acquisition layer for fusion analysis. Specific data includes equipment status parameters (coal mining machine motor current, voltage, power, traction speed; scraper conveyor chain tension, chain breakage sensor signal), environmental parameters (gas concentration, dust concentration, roof pressure displacement sensor data), and process parameters (preset mining height, cutting depth, etc.).

[0040] Step 2: Core Algorithm Model and Processing Flow The analysis process adopts a hierarchical model architecture of detection-identification-judgment-decision to achieve high efficiency and high accuracy.

[0041] Phase 1: Key target detection and localization. Lightweight deep convolutional neural network target detection models, such as YOLOv5s or MobileNet-SSD, are used. These models can perform real-time inference on edge computing devices, quickly and accurately delineating the target areas that need to be analyzed in each frame of video, such as the front drum of the coal mining machine, the rear drum of the coal mining machine, and the cutting section (coal wall / rock wall). The bounding boxes of the targets and the preliminary category confidence scores are output.

[0042] The second stage: refined scene understanding and anomaly identification. In this stage, a dedicated lightweight deep learning model is used for refined analysis of different target areas. The drum rock cutting recognition model can employ semantic segmentation network variants based on U-Net or DeepLabv3+, specifically optimized for downhole environments. Inputting the target area image of the drum and the cutting section captured in the first stage, it classifies each pixel in the target area image as either a coal seam, rock stratum, or drum. Its recognition principle lies in learning the deep visual feature differences between coal seams and rock strata under specific lighting and dust conditions downhole, such as texture features. Rock strata typically exhibit a coarser, more irregular, and higher-contrast texture. The output is a segmentation mask image that visually displays the rock stratum region. Simultaneously, key indicators are calculated. Rock cutting area ratio = (number of pixels in rock layer / total number of pixels in the cut section); The scraper conveyor operation status recognition model can use a temporal convolutional network (TCN) or a 3D CNN to analyze continuous video frames to identify abnormalities such as chain breakage, chain jamming, and skewing. By learning the regular optical flow or pixel-level change patterns of the chain movement during normal operation, when motion stagnation, irregular jumping, or abrupt changes in geometric shape are detected, it is judged as an abnormality.

[0043] The personnel intrusion detection model can combine object detection (YOLO) and posture estimation (such as OpenPose Lightweight) to detect in real time whether there are people in the preset restricted area, and can roughly judge whether the personnel's work behavior is standardized (such as squatting, climbing).

[0044] The third stage involves multi-source information fusion and abnormal operating condition decision-making. This stage inputs the outputs of all the aforementioned visual analysis models (rock cutting indicators, equipment status identification results), as well as auxiliary sensor data (such as sudden current surges). Thresholds are set based on the visual model outputs. For example, if the rock cutting area ratio is >15% and persists for more than 3 seconds, a primary rock cutting anomaly alarm is triggered. The visual alarm is cross-validated with sensor data. For instance, if a rock cutting anomaly visual alarm is accompanied by an abnormal increase in the coal mining machine motor current, the confidence level of the alarm is greatly increased, reducing false alarms.

[0045] Step 3: Output the decision results The edge analysis module outputs clear, actionable, and structured decision commands to the control module, rather than raw video or simple alarm signals.

[0046] Step 4: Model Training and Optimization Video clips from historical normal production periods and known anomalies (such as rock cutting and malfunction recordings) are extracted from hard disk video recorders at coal mine sites. Experts then perform fine-grained pixel-level annotations or bounding box annotations on coal seams, rock strata, equipment, and abnormal behaviors in the video frames to form a high-quality training set. Using a model pre-trained on large general datasets (such as ImageNet and COCO) as a foundation, fine-tuning is performed using a coal mine-specific dataset to simulate the underground environment. Random enhancements such as adding fog, dust, brightness adjustment, and blurring are applied to the training data to improve the model's robustness.

[0047] After system deployment, new abnormal samples that have been manually confirmed can be encrypted and sent back to the cloud training center for periodic model iteration updates to enhance the model's recognition accuracy.

[0048] It should be noted that the intelligent visual control and recognition system in this embodiment achieves real-time monitoring of the coal mine mining face and automatic handling of abnormal working conditions through video data acquisition, edge computing analysis and intelligent control linkage, which greatly improves the safety and efficiency of coal mine production. The modular design of this system gives it good scalability and compatibility, and it can be seamlessly connected with other information systems in the mine.

[0049] Example 2 Please see Figure 2 This embodiment provides an intelligent video control and recognition method for coal mine working faces. This method achieves automatic identification and processing of abnormal working conditions by real-time monitoring and intelligent analysis of video data from the coal mine working face, thereby improving the safety and efficiency of coal mining operations, effectively enhancing the level of intelligence in coal mine operations, realizing timely response and processing of abnormal working conditions, reducing production stoppages caused by equipment failures, and ensuring the safety of coal miners. It achieves the technical effect of improving the intelligence, safety, and efficiency of coal mine production, and solves the problems of low efficiency of manual inspection and untimely equipment failure warnings in traditional coal mine monitoring systems.

[0050] Specifically, in this embodiment, the method includes the following steps: S100: Real-time acquisition of video data from the working face within the coal mine via the sensing module 100.

[0051] It should be noted that in this step, the sensing module 100 includes multiple intrinsically safe mining cameras installed at different locations on the coal face. These cameras are explosion-proof and dustproof, enabling them to operate stably in the harsh environment of a coal mine. The video data collected by the cameras includes key information such as the operating status of the coal mining machine drum, the interface between the coal seam and rock strata, and the status of the support equipment. The video acquisition resolution can be set to 1080P and the frame rate to 30fps to ensure image clarity meets the requirements of subsequent analysis. The sensing module 100 is also equipped with an infrared imaging device, allowing for effective monitoring even in low-light conditions.

[0052] S200 receives video data, analyzes the video data using a preset algorithm model, and determines whether any abnormal operating conditions have occurred based on the video data.

[0053] It should be noted that in this step, the edge analysis module 200 uses a preset algorithm model to analyze the video data collected by the perception module 100 and identify whether the coal mining equipment drum has cut into the upper or lower rock strata. Specifically, the edge analysis module 200 first preprocesses the received video data, including image enhancement, noise reduction, and stabilization. Then, it uses a deep learning algorithm to identify the contact between the coal mining machine drum and the coal and rock strata in the video in real time. This algorithm is based on a convolutional neural network and learns the differences in color and texture features between the coal and rock strata to accurately identify whether the drum has cut into the upper or lower rock strata.

[0054] When the edge analysis module 200 detects that the coal mining equipment drum is cutting into the upper or lower rock strata, the system determines that an abnormal working condition has occurred; if it does not detect that the coal mining equipment drum is cutting into the upper or lower rock strata, it determines that no abnormal working condition has occurred. In addition, this embodiment will also combine historical data of the coal-rock interface and current coal mining parameters for comprehensive judgment to reduce the false judgment rate. This process focuses on identifying whether the coal mining equipment drum has cut into the upper or lower rock strata. The algorithm model uses deep learning and pattern recognition to analyze the video stream captured by the camera, quickly distinguishing between normal coal cutting operations and abnormal rock cutting situations. If the model determines that the coal mining equipment drum has cut into the upper or lower rock strata during operation, it is considered an abnormal condition, and the system immediately activates the alarm mechanism to notify on-site operators to take appropriate measures. Conversely, if no abnormal rock cutting situation is found in the video data, the system maintains normal monitoring and continues to monitor subsequent video data. This real-time intelligent identification not only effectively prevents equipment damage and extends equipment lifespan but also provides timely warnings of potential safety risks, ensuring the safe operation of coal mines. It achieves intelligent visual control of the coal mining face, improves the level of intelligence and safety of operations, and reduces production costs caused by equipment damage. This AI-based analysis method can improve the accuracy of identification as the algorithm model continues to learn and optimize, thereby enhancing the system's early warning capabilities and response speed.

[0055] S300: When abnormal operating conditions occur, the operating parameters of the coal mining equipment are adaptively adjusted to enable linkage control of the coal mining equipment.

[0056] It should be further explained that in this step, when the system determines that an abnormal operating condition has occurred, namely when the coal mining equipment drum cuts into the upper or lower rock strata, the control system will immediately execute an adaptive adjustment strategy. First, the system will automatically adjust the height of the coal mining machine drum according to the severity of the abnormality, returning it to a suitable cutting position. At the same time, the system will also adjust the forward speed of the coal mining machine, and in severe abnormal situations, it may even trigger an emergency shutdown procedure.

[0057] Preferably, the adaptive adjustment employs a fuzzy control algorithm, taking corresponding adjustment measures based on the degree of anomaly. For example, when the drum slightly contacts the rock strata, the system will slightly adjust the drum height; when the drum deeply penetrates the rock strata, the system will simultaneously adjust the drum height and the coal mining machine speed, and issue an alarm to remind the operator to intervene if necessary.

[0058] It should be noted that this embodiment provides an intelligent visual control and recognition method for coal mine mining faces. This adaptive adjustment mechanism enables intelligent linkage control of coal mining equipment, effectively avoiding equipment damage, increased dust, and safety risks caused by cutting into rock strata, while simultaneously improving coal quality and mining efficiency. The adjusted operating parameters are recorded and used for continuous system optimization, enabling the system to continuously improve recognition accuracy and control precision during long-term operation.

[0059] Example 3 In a preferred embodiment, this application also provides an electronic device, the electronic device comprising: The computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, implement the intelligent visual control and recognition method for coal mine mining faces. The computer device can be broadly categorized as a server, terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, memory, network interface, communication interface, etc., connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and internal memory. The non-volatile storage medium may store an operating system, computer programs, etc. The internal memory can provide an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface and communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, it performs the steps of the method of the present invention.

[0060] When the processor of this electronic device executes computer-readable instructions, the intelligent visual control and recognition method for coal mine mining faces is the same as the method described in Embodiment 2. It includes real-time acquisition of video data of the coal mine mining face through the sensing module 100; analysis of the video data and determination of whether abnormal working conditions occur through the edge analysis module 200; and adaptive adjustment of the operating parameters of the coal mining equipment to perform linkage control of the coal mining equipment when abnormal working conditions occur, based on the analysis results of the edge analysis module 200.

[0061] In this embodiment, the electronic device can be an industrial control computer, an embedded controller, or a dedicated intelligent control terminal for coal mines. The processor can be a multi-core CPU or GPU, with sufficient computing power to process high-definition video streams and run deep learning algorithms. The memory includes ROM and RAM, where ROM is used to store the operating system and basic software, and RAM is used for runtime data processing.

[0062] Preferably, the electronic device is also equipped with a high-speed data interface for establishing a stable connection with the camera of the sensing module 100, ensuring real-time transmission of video data. Simultaneously, the electronic device has an industrial control bus interface, enabling direct communication with the control system of the coal mining equipment to achieve parameter adjustment and the issuance of control commands.

[0063] Preferably, the electronic device adopts an explosion-proof design, meets the safety requirements for electrical equipment in underground coal mines, and can operate stably in high dust and high humidity environments. The device also has remote communication capabilities, enabling it to transmit working face status information and abnormal alarms to the ground monitoring center in real time, facilitating remote supervision and decision support for management personnel.

[0064] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0065] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, provided that such combination does not contain contradictions.

[0066] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. An intelligent visual control and recognition system for coal mine mining faces, comprising coal mining equipment, characterized in that, include: A sensing module, installed on the coal mining equipment, is used to collect video data from the working face inside the coal mine. An edge analysis module, which is signal-connected to the sensing module, is used to receive the video data, analyze the video data using a preset algorithm model, and determine whether an abnormal operating condition has occurred based on the video data. The control module is signal-connected to the edge analysis module and is used to adaptively adjust the operating parameters of the coal mining equipment when abnormal operating conditions occur, based on the analysis results of the edge analysis module, so as to perform linkage control on the coal mining equipment.

2. The intelligent visual control and recognition system for coal mine mining faces according to claim 1, characterized in that, The sensing module includes multiple intrinsically safe mining cameras deployed on the coal mining equipment. These cameras are used to collect video data of the working face inside the coal mine from different angles.

3. The intelligent visual control and recognition system for coal mine mining faces according to claim 1, characterized in that, The control module includes: The data acquisition layer is connected to the sensing module and is used to perform multi-source data fusion of the video data and existing system data. Then, a preset algorithm model is used to analyze the multi-source data and determine whether any abnormal operating conditions occur. The support service layer is connected to the data acquisition layer by signal, and is used to provide the data acquisition layer with a unified algorithm model and application service support; The business application layer is connected to the data acquisition layer and the support service layer respectively, and is used to adaptively adjust the operating parameters of the coal mining equipment in order to perform linkage control of the coal mining equipment when abnormal operating conditions occur.

4. The intelligent visual control and recognition system for coal mine mining faces according to claim 3, characterized in that, The business application layer includes an equipment management module, which is used to customize coal mining equipment linkage rules and alarm rules. When abnormal operating conditions occur, the equipment management module controls the coal mining equipment in accordance with the coal mining equipment linkage rules and issues alarm information according to the alarm rules.

5. The intelligent visual control and recognition system for coal mine mining faces according to claim 4, characterized in that, The business application layer also includes an open API interface, which connects to other information platforms in the coal mine and is used to transmit the alarm information.

6. The intelligent visual control and recognition system for coal mine mining faces according to claim 1, characterized in that, The edge analysis module includes: The coal mining equipment drum rock cutting identification module is used to identify whether the coal mining equipment drum has cut into the upper or lower rock layers based on the preset algorithm model and the video data collected by the sensing module.

7. The intelligent visual control and recognition system for coal mine mining faces according to claim 1, characterized in that, The system also includes a power supply module, which is connected to an intrinsically safe DC regulated power supply via a mining power line and a mining explosion-proof power supply tee box to power the sensing module; the intrinsically safe DC regulated power supply has a built-in backup power supply.

8. A method for intelligent visual control and recognition of coal mine working faces, characterized in that, include: S100: Real-time video data of the working face in the coal mine is collected through the sensing module; S200: Receive the video data, analyze the video data using a preset algorithm model, and determine whether an abnormal working condition has occurred based on the video data; S300: When abnormal operating conditions occur, the operating parameters of the coal mining equipment are adaptively adjusted to enable linkage control of the coal mining equipment.

9. The intelligent visual control and recognition method for coal mine mining faces according to claim 8, characterized in that, The step of analyzing the video data using a preset algorithm model and determining whether abnormal operating conditions have occurred based on the video data includes: The system uses a pre-defined algorithm model to analyze the video data collected by the sensing module and identify whether the coal mining equipment drum is cutting into the upper or lower rock layers. If the drum is found to be cutting into the upper or lower rock layers, an abnormal working condition is determined to have occurred; otherwise, no abnormal working condition is determined to have occurred.

10. An electronic device, characterized in that, include: Memory; The processor, wherein the memory stores computer-readable instructions, which, when executed by the processor, implement the intelligent visual control and recognition method for coal mine mining faces according to any one of claims 8 to 9.

Images