Coal machine equipment remote monitoring operation and maintenance system and operation and maintenance method

Abstract

The invention discloses a coal machine equipment remote monitoring operation and maintenance system and operation and maintenance method, and relates to the field of intelligent monitoring and operation and maintenance management of coal mine mechanical equipment, and the system comprises a terminal layer, a gateway layer, a cloud platform layer and an AR remote cooperation support platform. Wherein the terminal layer is used for collecting operation parameters of coal machine equipment and auxiliary equipment and carrying out coding and noise reduction processing; the gateway layer is used for realizing protocol conversion, optimized transmission, anomaly detection and secure transmission of data; the cloud platform layer is used for centralized data storage, intelligent analysis, knowledge management and decision support; and the AR remote cooperation support platform is used for realizing real-time visual cooperation between the scene and the remote experts. Through a terminal-gateway-cloud platform three-layer system architecture, real-time state perception, intelligent diagnosis analysis, remote collaborative operation and maintenance and knowledge self-learning closed loop of coal mine mechanical equipment are realized, so that the conversion from passive maintenance to predictive maintenance is realized, and the operation and maintenance efficiency, safety and intelligent level of the coal mine mechanical equipment are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent monitoring and operation and maintenance management of coal mine machinery equipment, and particularly relates to a coal mine equipment remote monitoring and operation and maintenance system and method. BACKGROUND

[0002] As the core equipment of coal mine production, the running stability of coal mine equipment is directly related to the output and safety of the mine. However, the traditional coal mine equipment operation and maintenance mode relies on manual inspection and regular maintenance, which has the following significant problems. First, the information island phenomenon is serious. There are many types of coal mine equipment, and different brands lead to different data formats, which cannot realize effective data fusion and sharing. This makes it impossible to form a unified data platform when monitoring and analyzing the equipment, resulting in low management efficiency and inability to achieve accurate fault diagnosis and prediction. Second, data transmission is unstable. Due to the complex underground environment of coal mines, electromagnetic interference, signal attenuation and other factors seriously affect data acquisition and transmission. Sensors in the coal mine environment may be damaged due to harsh environments such as high temperature, high humidity, and gas corrosion, resulting in discontinuous data transmission and poor real-time performance, which cannot ensure the stability and accuracy of the monitoring system. Third, the fault response is lagging. The current fault diagnosis relies on manual experience for inspection and analysis, and lacks intelligent diagnosis mechanism, which leads to the inability to discover and handle equipment faults in time. This lagging fault response often leads to unplanned downtime of equipment, affecting production efficiency and increasing maintenance costs. Fourth, there is insufficient knowledge sedimentation. The existing coal mine equipment management and operation and maintenance process lacks effective knowledge management and sedimentation. The experience of maintenance personnel cannot be effectively recorded and systematically managed, resulting in repeated occurrence of the same faults, low operation and maintenance efficiency, and inability to accumulate valuable operation and maintenance data and experience.

[0003] Although some existing technologies have achieved remote monitoring and fault diagnosis of some coal mine equipment by introducing Internet of Things and artificial intelligence technologies, there are still the following deficiencies. First, the limitations of Internet of Things technology. Existing monitoring systems based on Internet of Things only realize basic data acquisition functions. Although they can collect real-time operation state data of coal mine equipment, they cannot effectively fuse and analyze the data, resulting in low intelligent level of the monitoring system. The accuracy of data acquisition, interconnection between devices, and stability of data transmission cannot be solved. Second, insufficient application of artificial intelligence. Although existing intelligent fault diagnosis systems have introduced artificial intelligence technology, most systems still cannot fully adapt to the complex environment of coal mines. AI diagnosis models often rely on a large amount of high-quality data for training, but data collection in the mine environment has noise and poor stability of sensors, resulting in low accuracy of fault diagnosis and difficulty in achieving accurate real-time fault warning and diagnosis.

[0004] Based on the above problems, the existing coal machine equipment remote monitoring system faces the following three technical problems. First, the data fusion problem, the data format of multiple brands and types of coal mine equipment is not unified, causing difficulty in data sharing, making it impossible to effectively fuse and analyze the data, and leading to the inability to form a comprehensive systematic management of the state monitoring and fault diagnosis of the equipment. Second, the transmission reliability problem, due to the particularity of the underground environment, the transmission link of the coal mine equipment monitoring system is often affected by electromagnetic interference, insufficient sensor precision and other factors, causing unstable data transmission, affecting the effectiveness and accuracy of real-time monitoring, and limiting the practicality and reliability of the monitoring system. Third, the intelligent diagnosis capability is insufficient, the current fault diagnosis technology still relies on manual experience for judgment, the intelligent level is low, and there is a lack of comprehensive monitoring and accurate diagnosis of the health status of the equipment, leading to a lag in the response to equipment failure, increasing the time and maintenance cost of unplanned downtime.

[0005] Although the monitoring system based on the Internet of Things and the artificial intelligence fault diagnosis technology have been proposed in the prior art, these systems mostly have the following limitations. Limitation one is that the existing Internet of Things system can realize basic data collection, but cannot solve the problems of data fusion, information sharing and intelligent analysis between devices. Limitation two is that the fault diagnosis technology based on artificial intelligence can provide intelligent analysis, but there is still a big gap in the intelligent level and adaptability, especially in the complex environmental conditions of the mine, the reliability and accuracy of intelligent fault diagnosis are low. Limitation three is that the existing remote monitoring system mostly fails to fully consider the special needs of the mine environment, lacks adaptability to the complex underground environment, and thus greatly reduces the stability and usability of the system.

[0006] Therefore, there is an urgent need for a new remote monitoring and operation system that integrates multiple advanced technologies and is specially optimized for coal mine scenarios. SUMMARY

[0007] In order to solve the problems of existing coal machine equipment monitoring and operation system, such as difficulty in fusing data of multiple brands of equipment, poor reliability of underground data transmission, reliance on manual experience for fault diagnosis, and difficulty in knowledge accumulation, the present application provides a coal machine equipment remote monitoring and operation system and operation method, which innovatively realizes real-time state perception, intelligent diagnosis and analysis, remote collaborative operation and knowledge self-learning closed loop of coal mine mechanical equipment by building a "terminal-gateway-cloud platform" three-layer architecture, thereby realizing the transition from "passive maintenance" to "predictive maintenance", and significantly improving the operation efficiency, safety and intelligent level of coal machine equipment.

[0008] The technical scheme adopted by the coal machine equipment remote monitoring and operation system and operation method of the present application is as follows: A coal machine equipment remote monitoring and operation system, comprising a terminal layer, a gateway layer, a cloud platform layer and an AR remote collaboration support platform; wherein, The terminal layer is arranged on the coal mining equipment and auxiliary equipment, and includes a sensor group and an industrial computer, and is used for collecting operation parameters of the coal mining equipment and auxiliary equipment and performing encoding and noise reduction processing; The gateway layer is arranged in the underground communication node, integrates a multi-protocol fusion communication module, a data preprocessing module, a breakpoint resume transmission module and an encrypted transmission module, and is used for realizing protocol conversion, optimized transmission, abnormality detection and safe transmission of data; and the gateway layer adopts a multi-protocol communication mechanism of fusion of an MQTT protocol and a Kafka stream processing framework; The cloud platform layer is based on a hybrid distributed architecture of HDFS+HBase+Redis, includes a data storage module, an intelligent fault diagnosis module, a dynamic knowledge base module and a visual display module, and is used for centralized data storage, intelligent analysis, knowledge management and decision support; wherein the intelligent fault diagnosis module adopts a double-engine mechanism combining a rule engine and a model engine. The AR remote collaboration support platform includes an AR helmet terminal, a cloud collaboration server and an expert terminal, and is used for realizing real-time visual collaboration between the scene and the remote experts.

[0009] The further improvement of the technical scheme of the present application is that the breakpoint resume transmission module of the gateway layer is internally provided with a local cache unit, which can cache the un-uploaded data when the communication is interrupted, and automatically retransmit the data after the communication is restored.

[0010] The further improvement of the technical scheme of the present application is that the data preprocessing module of the gateway layer adopts a data compression algorithm and a differential synchronization algorithm to compress and incrementally transmit the collected data, thereby reducing the network resource occupation in the underground low-bandwidth environment.

[0011] The further improvement of the technical scheme of the present application is that the rule engine of the intelligent fault diagnosis module realizes real-time alarm based on the equipment operation threshold, and the model engine adopts a fusion model of a random forest algorithm and a deep neural network algorithm, and is used for fault type identification, root cause positioning, trend prediction and equipment health scoring.

[0012] The further improvement of the technical scheme of the present application is that the average early warning time of the model engine is greater than or equal to 12 hours, the equipment health score is constructed based on the operation parameters, fault history and maintenance record, the scoring range is 0-100 points, and the early warning is triggered when the score is less than or equal to 60 points.

[0013] The further improvement of the technical scheme of the present application is that the dynamic knowledge base module stores historical fault cases, maintenance experience and diagnosis results, adopts a tagging management and self-learning closed-loop mechanism, the self-learning closed-loop mechanism includes four stages of diagnosis, verification, learning and optimization, and realizes dynamic updating and reuse of knowledge.

[0014] Further improvement of the technical scheme of the application is that the AR remote collaboration support platform supports real-time picture superposition prompt, fault component labeling, remote task assignment and collaboration process closed-loop management, and the collaboration response time is less than or equal to 3 seconds.

[0015] Further improvement of the technical scheme of the application is that a communication connection is established between the gateway layer and the cloud platform layer through a TLS secure channel, and an AES encryption algorithm is used for data transmission to prevent data leakage and tampering.

[0016] Further improvement of the technical scheme of the application is that the sensor group of the terminal layer includes a temperature sensor, a current sensor, a hydraulic pressure sensor, a vibration sensor, a flow sensor, a geographic positioning module and a gas concentration sensor, and the collected parameters include motor temperature, current, hydraulic pressure, vibration frequency, driving speed and gas concentration.

[0017] A coal machine equipment remote monitoring and operation method adopts the operation system and comprises the following steps: S1, the terminal layer collects equipment operation parameters through the sensor group, and the industrial computer encodes, packs and preliminarily denoises the parameters; S2, the gateway layer receives the terminal layer data, performs protocol conversion, data compression and encryption processing, and then uploads the data to the cloud platform layer through the MQTT protocol, simultaneously realizes high-concurrency transmission by using the Kafka framework, and starts breakpoint resume transmission when communication is interrupted; S3, the cloud platform layer stores the data in a classified manner, and the intelligent fault diagnosis module analyzes the equipment state through a double-engine mechanism and judges whether there is a fault or potential risk; S4, if there is a fault or risk, the cloud platform layer triggers an alarm and pushes the diagnosis result and processing suggestion to the operation personnel terminal; S5, the operation personnel handle the fault according to the suggestion, directly operate and feed back the result for simple faults, and connect experts to complete fault handling for complex faults by starting the AR remote collaboration module; S6, the cloud platform layer inputs the fault handling process and result into a dynamic knowledge base and updates the model and knowledge base through a self-learning closed-loop mechanism; S7, the cloud platform layer displays the equipment operation state through a visual interface, generates an operation report and provides decision support.

[0018] Thanks to the above technical scheme, the application has the following technical progress: Through the multi-protocol fusion mechanism of the gateway layer, the application supports multiple protocols such as MQTT, CoAP and HTTPS, can adapt to coal machine equipment of different brands and models, realizes unified format conversion and fusion transmission of multi-source data, constructs a unified data management platform, solves the data island problem and provides a data basis for comprehensive equipment state monitoring and comprehensive analysis.

[0019] The application adopts the transmission architecture combining the MQTT lightweight protocol with the Kafka high-concurrency message queue, and is supplemented by the data compression, differential synchronization and breakpoint resume mechanism, so that stable data transmission in a low-bandwidth and high-interference environment is realized.The data transmission success rate is improved, and the response delay is controlled within 500 ms, so that the continuous and stable transmission of real-time monitoring data is ensured, and the effectiveness of remote monitoring is ensured.

[0020] The application constructs a double-engine diagnosis mechanism combining a rule engine and a model engine, the rule engine realizes real-time alarm based on device running threshold, the model engine adopts random forest and deep neural network algorithm, and the historical data training and scene adaptation optimization are used to improve the fault diagnosis accuracy, and potential fault warning can be realized 12 hours in advance, so that the traditional passive maintenance is changed into predictive maintenance, and the non-scheduled downtime is reduced.

[0021] The dynamic knowledge base of the application realizes automatic absorption and reuse of knowledge by structured storage and tagging processing of historical maintenance cases and expert experience in combination with machine learning algorithm.Through the closed-loop mechanism of diagnosis-verification-learning-optimization, the system can update the fault model and maintenance scheme continuously, realize the sedimentation and iteration of operation and maintenance knowledge, reduce the dependence on artificial experience, and improve the overall operation and maintenance efficiency.

[0022] The AR remote collaboration platform of the application realizes real-time transmission and visual labeling of the scene picture through the AR helmet or mobile terminal, and the remote expert can accurately locate the fault component and issue operation guidance, and the cooperative response time is controlled within 3 seconds.This mode effectively makes up for the scarcity of on-site expert resources, shortens the fault response and processing time, and improves the maintenance efficiency and safety of complex faults.

[0023] The application realizes data transmission encryption of the gateway layer and the cloud platform layer through the TLS secure channel and the AES encryption algorithm, prevents data leakage and tampering, and has the functions of breakpoint resume and abnormality detection, so that data is not lost and the system runs stably when communication is interrupted or the device is abnormal, and the safety and reliability of the overall system are improved. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 It is a coal mining machine terminal data transmission diagram of the application; Figure 2 It is a schematic diagram of the MQTT protocol of the application; Figure 3 It is a big data platform architecture diagram of the application; Figure 4 It is a big data analysis model structure diagram of the application; Figure 5 It is a KAFKA function structure diagram of the application; Figure 6 Figure 1 is a system device list interface diagram of the present application; Figure 7 Figure 2 is a system device detailed interface diagram of the present application; Figure 8 Figure 3 is a system device fault list diagram of the present application; Figure 9 Figure 4 is a system fault knowledge base interface diagram of the present application; Figure 10 Figure 5 is a remote diagnosis and judgment interface diagram of the present application; Figure 11 Figure 6 is a remote diagnosis and judgment specific flow diagram of the present application; Figure 12 Figure 7 is a data processing visualization display interface diagram of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the present application more clear and obvious, the present application is further described in detail below in combination with specific embodiments and with reference to the drawings. In the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present application.

[0026] As shown in Figures 1-12 The present embodiment provides a coal machine equipment remote monitoring and operation and maintenance system, which comprises terminal layer deployment, gateway layer deployment, cloud platform layer deployment and AR remote collaboration device deployment, and each layer is communicatively connected through an underground industrial network and a public network.

[0027] The terminal layer is deployed at key positions of a coal mining machine, a scraper conveyor and a hydraulic support, and comprises a sensor group, an industrial computer and a data transmission interface, thereby realizing multi-dimensional acquisition and preliminary processing of equipment operation parameters.

[0028] The sensor group comprises more than 20 sensors such as temperature sensors, current sensors, hydraulic pressure sensors, vibration sensors, flow sensors, power detection modules, geographic positioning modules and gas concentration sensors, which are configured according to the monitoring requirements of different equipment, and the specific installation positions are as follows: A temperature sensor and a vibration sensor are installed on the cutting part motor housing of the coal mining machine, a current sensor and a frequency sensor are installed on the traction part, a hydraulic pressure sensor and an oil pump flow sensor are installed on the hydraulic system, and an inclination sensor and a geographic positioning module are installed on the machine body; a temperature sensor, a current sensor and a vibration sensor are installed on the motor end of the scraper conveyor, and a pressure sensor is installed on the tension part of the conveyor belt; a pressure sensor and a displacement sensor are installed on the hydraulic cylinder of the hydraulic support, and a flow sensor and an oil temperature sensor are installed on the hydraulic system. The acquisition frequency of each sensor is 10 Hz, and the acquisition accuracy is within ±0.5%, thereby ensuring the real-time and accuracy of data acquisition.

[0029] The industrial computer receives raw data of various sensors in real time through the bus interface, including more than 20 types of parameters such as motor temperature, current, hydraulic pressure, vibration frequency, driving speed, gas concentration, etc.; and uses LASC high-speed modulation technology to encode and package the raw data, the encoding format follows a self-defined industrial standard to ensure the efficiency and compatibility of data transmission, and finally the collected raw data is preliminarily denoised to eliminate obvious outliers (such as data exceeding the reasonable range caused by sensor failure), and is stored according to the device type and parameter category, waiting to be uploaded to the gateway layer.

[0030] The gateway layer in this embodiment is deployed in the explosion-proof junction box of the downhole communication node, serving as a data relay and intelligent filtering hub, integrating a multi-protocol converged communication module, a data preprocessing module, a breakpoint resume transmission module, and an encrypted transmission module.

[0031] The gateway layer is built-in with a multi-protocol conversion chip, supporting the access and conversion of MQTT, CoAP, HTTPS, etc. The industrial computer sends the packaged data to the gateway layer through the private LASC protocol, and the protocol analysis module of the gateway layer analyzes the LASC protocol data and extracts the original parameters; converts the analyzed original parameters into MQTT protocol format for efficient transmission in low-bandwidth environments; at the same time, a high-concurrency message queue is built through the Kafka stream processing framework to meet the demand of simultaneous data upload of multiple devices on the fully mechanized coal mining face.

[0032] At the same time, the gateway layer of this embodiment uses a data compression algorithm to compress the converted MQTT protocol data, with a compression ratio of 3:1, significantly reducing the bandwidth occupancy of data transmission; for periodically collected parameters (such as motor temperature, current), a differential synchronization algorithm is used to transmit only the difference data from the previous collection period, further reducing the data transmission volume and adapting to the low-bandwidth environment underground; based on the preset parameter threshold (such as the normal range of motor temperature 0-85℃), the data is detected in real time, the abnormal data is marked and transmitted preferentially, ensuring timely feedback of fault information.

[0033] The gateway layer is built-in with a local cache, which automatically stores the unuploaded data to the local cache when the communication with the cloud platform is interrupted; when the communication is restored, the cached data is retransmitted in sequence according to the data collection time, ensuring the success rate of data transmission; the gateway layer and the cloud platform layer establish a secure communication channel through the TLS protocol, and the data transmission is encrypted using the AES encryption algorithm to prevent data from being stolen or tampered with during transmission.

[0034] Through the above optimization, the data transmission response delay of the gateway layer is controlled within 500ms, meeting the needs of real-time monitoring; in an environment with electromagnetic interference intensity ≥80dB underground, the data transmission success rate still remains above 99.5%, adapting to the complex communication environment underground.

[0035] The cloud platform layer is deployed in the ground monitoring center of the coal mine, is constructed based on a hybrid distributed architecture of HDFS+HBase+Redis, and includes a data storage module, an intelligent fault diagnosis module, a dynamic knowledge base module, a visual display module, and a decision support module.

[0036] The HDFS (distributed file system) is used to store massive raw collected data, historical operation data, and video data, supports distributed storage and parallel reading of data, and meets the needs of multi-year data retention and backtracking analysis; the HBase (distributed column storage database) is used to store structured device parameters, fault records, maintenance cases, and the like, supports high-concurrency reading and writing and random queries, and the query response time is ≤100 ms; and the Redis (cache database) is used to cache hot data such as real-time device status, fault early warning information, and high-frequency access maintenance cases, to improve the data query and access speed and ensure the real-time performance of visual display and intelligent decision-making.

[0037] The intelligent fault diagnosis module adopts a double-engine mechanism combining a rule engine and a model engine; the rule engine has a built-in device running threshold library, the thresholds are set according to the technical parameters provided by the device manufacturers and the actual operation experience on site, for example, a coal mining machine motor temperature ≥85℃ triggers an “overheated motor” alarm, a hydraulic pressure ≤15MPa triggers a “hydraulic leakage” early warning, and a vibration amplitude ≥5mm / s triggers a “mechanical looseness” early warning. The rule engine receives real-time data from the cloud platform in real time, compares it with the threshold library, and automatically triggers an alarm of the corresponding level when the parameter exceeds the threshold, and pushes it to the visual display module and the operation and maintenance personnel terminal.

[0038] The model engine adopts a fusion model combining a random forest algorithm and a deep neural network (DNN) algorithm, first extracts device normal operation data and fault data from the historical data of the cloud platform, performs data cleaning (eliminates noise data and fills in missing values), and feature extraction (extracts vibration frequency features and current change trend features); then divides the preprocessed data set into a training set and a test set in a 7:3 ratio, the training set is used for model parameter training, and the test set is used for model precision verification; adjusts the number of decision trees of the random forest and the number of network layers of the DNN through iterative optimization, so that the model test accuracy is ≥92%; finally, the trained fusion model is deployed to the real-time analysis node of the cloud platform, the model receives the operation data uploaded by the terminal layer in real time, identifies the fault type (such as motor fault, hydraulic fault, mechanical fault, etc.), locates the fault root cause (such as specific fault components and fault triggering factors), predicts the fault trend (such as fault deterioration speed and predicted fault occurrence time), and generates a device health score (full score 100 points, less than 60 points triggers an early warning) and a maintenance plan suggestion.

[0039] The knowledge base is built based on an HBase database, and the storage content includes historical fault cases (device model, fault phenomenon, fault cause, treatment scheme, treatment result), expert experience (text description, operation video, schematic diagram), fault coding mapping table, etc. All data is processed by tagging, and the tags include device type, fault type, fault component, treatment method, etc., which facilitates quick retrieval and matching.

[0040] The dynamic knowledge base adopts a self-learning closed-loop mechanism, which includes four stages of diagnosis, verification, learning and optimization, realizes dynamic updating and reuse of knowledge, and specifically as follows: Diagnosis stage: the intelligent fault diagnosis module outputs fault diagnosis results and preliminary treatment suggestions; Verification stage: field operation and maintenance personnel handle faults according to the suggestions, and feed back the treatment process, actual fault cause and treatment effect to the cloud platform; Learning stage: the knowledge base module performs structured processing on the feedback information, extracts new fault features and treatment experience, and updates the parameters of the fault model and the retrieval rules of the knowledge base through machine learning algorithm; Optimization stage: when similar faults occur again, the system pushes more accurate treatment schemes based on the updated model and knowledge base, realizes the continuous iteration of "diagnosis-verification-learning-optimization", and at the same time, the knowledge base supports multi-user collaborative editing and labeling, and operation and maintenance personnel can upload personal experience and share it, improving the accumulation efficiency of overall operation and maintenance knowledge.

[0041] In this embodiment, the AR remote collaboration support platform includes an AR helmet terminal, a cloud collaboration server and an expert end command center. The field operation and maintenance personnel are equipped with AR helmets, and the helmets are built-in communication modules and downhole WiFi modules, which are suitable for downhole network environment; the cloud collaboration server is deployed in the cloud platform layer, supports video stream real-time transcoding, labeling instruction transmission and collaboration process management; the expert end command center is deployed in the ground monitoring center, and is equipped with high-performance computers, display screens and voice communication equipment.

[0042] When the field device fails and the operation and maintenance personnel cannot solve it independently, the operation and maintenance personnel start the AR remote collaboration module through voice instructions or buttons of the AR helmet; the AR helmet automatically sends a connection request to the cloud collaboration server through the underground network, and the request contains the device number, fault code and real-time video stream; the cloud collaboration server automatically matches the corresponding remote expert according to the fault type and device model, and sends a collaboration request to the expert end; after the expert confirms, the system establishes a real-time communication link between the AR helmet and the expert end; the expert views the real-time video through the display screen, marks the fault components on the screen, draws the operation path, and issues text or voice guidance; the operation and maintenance personnel view the marking information and guidance instructions through the display interface of the AR helmet, and handle the fault according to the requirements; after the fault handling is completed, the operation and maintenance personnel feed back the processing result through the AR helmet, the system automatically records the collaboration process and processing result, updates to the knowledge base, and completes the collaboration process closed loop.

[0043] Embodiment two The embodiment provides a coal machine equipment remote monitoring and operation and maintenance method, which comprises the following steps: S1, the terminal layer collects device operation parameters through a sensor group, and an industrial computer encodes, packs and preliminarily denoises the parameters; S2, the gateway layer receives the terminal layer data, performs protocol conversion, data compression and encryption processing, and then uploads to the cloud platform layer through the MQTT protocol, simultaneously realizes high-concurrency transmission by using the Kafka framework, and starts breakpoint resume when communication is interrupted; S3, the cloud platform layer stores the data in a classified manner, and an intelligent fault diagnosis module analyzes the device state through a double-engine mechanism and judges whether there is a fault or potential risk; S4, if there is a fault or risk, the cloud platform layer triggers an alarm and pushes the diagnosis result and processing suggestion to the operation and maintenance personnel terminal; S5, the operation and maintenance personnel handle the fault according to the suggestion, directly operate and feed back the result for simple faults, and start the AR remote collaboration module for complex faults to connect experts to complete the fault handling; S6, the cloud platform layer inputs the fault handling process and result into a dynamic knowledge base, and updates the model and knowledge base through a self-learning closed loop mechanism; S7, the cloud platform layer displays the device operation state through a visual interface, generates an operation and maintenance report, and provides decision support.

[0044] In the above embodiment, the present application provides a coal machine equipment remote monitoring operation and maintenance system and an operation and maintenance method. The present application supports access of multiple protocols such as MQTT, CoAP and HTTPS through a multi-protocol fusion mechanism of a gateway layer, can adapt to coal machine equipment of different brands and models, realizes unified format conversion and fusion transmission of multi-source data, constructs a unified data management platform, solves the problem of data island, provides a data basis for comprehensive equipment state monitoring and comprehensive analysis, adopts a transmission architecture combining MQTT lightweight protocol and Kafka high-concurrency message queue, and is supplemented by data compression, differential synchronization and breakpoint resume mechanism, thereby realizing stable data transmission in a low-bandwidth and high-interference environment, improving data transmission success rate, controlling response delay within 500 ms, ensuring continuous and stable transmission of real-time monitoring data, and ensuring effectiveness of remote monitoring. The present application constructs a double-engine diagnosis mechanism combining a rule engine and a model engine. The rule engine realizes real-time alarm based on equipment operation threshold, the model engine adopts random forest and deep neural network algorithms, and through historical data training and scene adaptation optimization, the accuracy of fault diagnosis is improved, potential fault warning can be realized 12 hours in advance, traditional passive maintenance is changed into predictive maintenance, and thus non-scheduled downtime is reduced. The dynamic knowledge base of the present application realizes automatic absorption and reuse of knowledge through structured storage and tagging processing of historical maintenance cases and expert experience in combination with machine learning algorithms. Through a closed-loop mechanism of “diagnosis-verification-learning-optimization”, the system can continuously update fault models and maintenance schemes, realize sedimentation and iteration of operation and maintenance knowledge, reduce dependence on artificial experience, and improve overall operation and maintenance efficiency. The AR remote collaboration platform of the present application realizes real-time transmission and visual labeling of a live picture through an AR helmet or a mobile terminal. Remote experts can accurately locate fault components and issue operation guidance. Collaboration response time is controlled within 3 seconds. This mode effectively makes up for the scarcity of on-site expert resources, shortens fault response and processing time, and improves the efficiency and safety of complex fault repair. The present application realizes data transmission encryption of the gateway layer and the cloud platform layer through a TLS secure channel and an AES encryption algorithm, prevents data leakage and tampering, and has functions such as breakpoint resume and abnormality detection, thereby ensuring that data is not lost and the system is stably operated when communication is interrupted or equipment is abnormal, and improving the safety and reliability of the overall system.

[0045] The above-described embodiments are merely preferred embodiments of the present application and do not limit the concept and scope of the present application. Various modifications and improvements to the technical solutions of the present application made by those skilled in the art without departing from the design concept of the present application shall fall within the protection scope of the present application. The technical content claimed by the present application has been fully recorded in the claims.

Claims

1. A remote monitoring and maintenance system for coal mining equipment, characterized in that: This includes the terminal layer, gateway layer, cloud platform layer, and AR remote collaboration support platform; among which, The terminal layer is set on the coal mining equipment and auxiliary equipment, including sensor groups and industrial control computers, and is used to collect the operating parameters of the coal mining equipment and auxiliary equipment and perform encoding and noise reduction processing; The gateway layer is set up at the underground communication node and integrates a multi-protocol fusion communication module, a data preprocessing module, a breakpoint resume module, and an encrypted transmission module. It is used to realize data protocol conversion, optimized transmission, anomaly detection, and secure transmission. The gateway layer adopts a multi-protocol communication mechanism that integrates the MQTT protocol and the Kafka stream processing framework. The cloud platform layer is based on a hybrid distributed architecture of HDFS+HBase+Redis, including a data storage module, an intelligent fault diagnosis module, a dynamic knowledge base module, and a visualization module, which are used for centralized data storage, intelligent analysis, knowledge management, and decision support; among them, the intelligent fault diagnosis module adopts a dual-engine mechanism that combines a rule engine and a model engine. The AR remote collaboration support platform includes an AR helmet terminal, a cloud collaboration server, and an expert terminal, which enables real-time visual collaboration between on-site and remote experts.

2. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that, The gateway layer's breakpoint resume module has a built-in local cache unit, which can cache unuploaded data when communication is interrupted and automatically retransmit it after communication is restored.

3. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The data preprocessing module of the gateway layer uses data compression and differential synchronization algorithms to compress and incrementally transmit the collected data, thereby reducing the network resource consumption in the low-bandwidth environment downhole.

4. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The rule engine of the intelligent fault diagnosis module implements real-time alarms based on equipment operating thresholds, and the model engine adopts a fusion model of random forest algorithm and deep neural network algorithm for fault type identification, root cause location, trend prediction and equipment health scoring.

5. The remote monitoring and maintenance system for coal mining equipment according to claim 4, characterized in that: The model engine has an average early warning time of ≥12 hours. The equipment health score is constructed based on operating parameters, fault history, and maintenance records. The score range is 0-100 points, and an early warning is triggered when the score is ≤60 points.

6. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The dynamic knowledge base module stores historical fault cases, maintenance experience, and diagnostic results. It adopts a tag-based management and self-learning closed-loop mechanism, which includes four stages: diagnosis, verification, learning, and optimization, to achieve dynamic updating and reuse of knowledge.

7. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The AR remote collaboration support platform supports real-time image overlay prompts, fault component labeling, remote task assignment, and closed-loop management of the collaboration process, with a collaboration response time of ≤3 seconds.

8. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The gateway layer and the cloud platform layer establish a communication connection through a TLS secure channel, and data transmission uses the AES encryption algorithm to prevent data leakage and tampering.

9. The remote monitoring and maintenance system for coal mining equipment according to claim 1, characterized in that: The sensor group of the terminal layer includes a temperature sensor, a current sensor, a hydraulic pressure sensor, a vibration sensor, a flow sensor, a geolocation module, and a gas concentration sensor. The parameters collected include motor temperature, current, hydraulic pressure, vibration frequency, driving speed, and gas concentration.

10. A method for remote monitoring and maintenance of coal mining equipment, characterized in that, The operation and maintenance system according to any one of claims 1-9 includes the following steps: S1. The terminal layer collects equipment operating parameters through the sensor group, and the industrial control computer encodes, packages and performs preliminary noise reduction on the parameters. S2. The gateway layer receives data from the terminal layer, performs protocol conversion, data compression, and encryption, and then uploads it to the cloud platform layer via the MQTT protocol. At the same time, it uses the Kafka framework to achieve high-concurrency transmission and starts breakpoint resume when communication is interrupted. S3 and cloud platform layers classify and store data, while the intelligent fault diagnosis module analyzes the equipment status through a dual-engine mechanism to determine whether there are faults or potential risks. S4. If a fault or risk exists, the cloud platform layer will trigger an alarm and push the diagnostic results and handling suggestions to the operation and maintenance personnel's terminal. S5. Maintenance personnel handle faults according to suggestions. For simple faults, they can operate directly and report the results. For complex faults, they can activate the AR remote collaboration module to connect with experts to complete the fault handling. S6, the cloud platform layer records the fault handling process and results into the dynamic knowledge base, and updates the model and knowledge base through a self-learning closed-loop mechanism; The S7 cloud platform layer displays the device's operating status through a visual interface, generates operation and maintenance reports, and provides decision support.

Images