System-level Fault Prediction and Structural Upgrade Method for Coal Mining Face

Abstract

The present invention discloses a system-level fault prediction and structure upgrade method for a coal mining face, which relates to the technical field of data processing and includes: Step 1: Obtain the operation data of each sub-device in the coal mining face, and predict the fault probability of each sub-device having various types of faults in the next time step; the fault types include communication faults and equipment faults; Step 2: If the predicted probability of the sub-device having an equipment fault in the next time step exceeds a set first fault threshold, simulate the operation of the sub-device and communicate and interact with other sub-devices; Step 3: If the predicted probability of the sub-device having a communication fault in the next time step exceeds a set second fault threshold, control all sub-devices to call the backup network and then repair the original network. The present invention can accurately predict the occurrence time of equipment faults and automatically perform corresponding system structure upgrades.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically to a method for system-level fault prediction and structural upgrade of a coal mining face system. Background Art

[0002] With the continuous growth of global energy demand and the increasing depletion of resources, coal mines, as one of the important energy resources, their mining efficiency and safety have become key issues in modern coal mine production. The complexity and high risk of coal mine excavation operations make the equipment management and fault prediction of coal mining faces particularly important. Coal mine equipment faces various risks such as overload, wear, corrosion, and equipment failures during long-term operation. These factors not only affect the production efficiency of coal mines but also pose a threat to the safety of miners. To improve the operation efficiency and safety of coal mines, more and more coal mine enterprises have begun to adopt automated and intelligent equipment management and fault diagnosis systems.

[0003] In the past few decades, traditional coal mine equipment management methods mainly relied on manual inspection and regular maintenance. These traditional methods relied on miners or equipment maintenance personnel to regularly check the equipment status or take repair measures when obvious equipment failures occurred. However, this experience-based approach has significant limitations. First, the occurrence of equipment failures is often sudden and difficult to detect in advance by manual means, resulting in equipment failures in high-load or harsh working environments, thus causing production stagnation or safety accidents. Second, manual inspections are often limited by factors such as time and environment, which may lead to the neglect of hidden equipment failures or delays in repair opportunities, further increasing the probability of failures. Finally, manual maintenance cannot efficiently process large-scale equipment monitoring data and lacks the ability to perform real-time analysis of equipment operating status. Therefore, traditional equipment management methods can no longer meet the high requirements of modern coal mines for production efficiency and safety.

[0004] To address these issues, in recent years, many intelligent device management systems based on automation and data analysis technologies have emerged. These systems can accurately predict equipment failures and optimize equipment operation parameters by monitoring equipment status in real time, collecting equipment operation data, and combining data analysis and algorithm prediction, thereby reducing the occurrence of equipment failures and improving the overall efficiency and safety of coal mine operations. However, there are still some problems in the equipment failure prediction systems in the existing technologies. The existing failure prediction systems mainly rely on sensors to collect equipment operation data such as temperature, power, vibration, load, etc. Although these data can reflect certain operation states of the equipment, due to the complex and dynamically changing equipment operation environment, the existing monitoring methods often can only monitor one dimension of the equipment. Many complex failures, especially those related to environmental factors or multi-device synergy effects, cannot be accurately predicted through single-dimensional data analysis. For example, the equipment in the coal mining face is not only affected by internal mechanical loads but also by various external environmental factors such as coal seam hardness, gas concentration, and temperature changes. Therefore, the existing technologies often cannot comprehensively and accurately evaluate the overall health status and failure risks of the equipment. Lack of comprehensive consideration of environmental interaction factors: The existing failure prediction methods often focus on the status monitoring of the equipment itself but ignore the impact of environmental factors on equipment operation. For example, environmental factors such as coal seam hardness, working face roof height, and gas concentration directly affect the load status of coal mining equipment. Excessive gas concentration and coal seam hardness may lead to equipment overload, increased wear, and even serious accidents such as explosions. In the existing technologies, the operation data of the equipment and environmental factors are usually processed separately, lacking systematic consideration of the impact of the environment on equipment failure risks, resulting in a decrease in the accuracy of failure prediction. Summary of the Invention

[0005] To solve the above technical problems, a method for system-level failure prediction and structure upgrade of the coal mining face is provided. This method can comprehensively and accurately evaluate the health status of the equipment by monitoring the operation data and environmental status of the equipment in real time and combining the historical failure data of the equipment, thereby predicting the occurrence of equipment failures. This multi-dimensional and multi-level failure prediction method significantly improves the accuracy and timeliness of prediction compared with traditional single-device monitoring and manual maintenance. It also significantly enhances the stability, reliability, and flexibility of the entire coal mining face system through intelligent virtual mirror technology and network structure optimization.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for system-level failure prediction and structure upgrade of the coal mining face, comprising:

[0008] Step 1: Obtain the operation data of each sub-device in the coal mining face, perform data analysis on the operation data of each sub-device, and predict the failure probability of each sub-device having various types of failures in the next time step; the failure types include: communication failure and equipment failure;

[0009] Step 2: If the predicted probability of the sub-device having an equipment failure in the next time step exceeds the set first failure threshold, isolate the sub-device, generate a virtual image device corresponding to the sub-device in the cloud to simulate the operation of the sub-device, and communicate and interact with other sub-devices;

[0010] Step 3: If the predicted probability of the sub-device having a communication failure in the next time step exceeds the set second failure threshold, control the sub-device to call the backup network to send or receive data; the backup network is a communication network with a blockchain structure, in which each sub-device is interconnected; when the number of sub-devices calling the backup network in the coal mining face exceeds the set quantity threshold, control all sub-devices to call the backup network, and then repair the original network.

[0011] Further, when all sub-devices in the coal mining face call the backup network, all sub-devices find the shortest communication path with other sub-devices through the routing addressing algorithm.

[0012] Further, when repairing the original network, adjust the structure of the original network to be a mirror image of the network formed by the actual communication connections of each sub-device in the backup network to complete the upgrade of the original network structure.

[0013] Further, the sub-devices at least include: coal shearer, fan, conveyor belt, hydraulic pump and valve.

[0014] Further, for the th sub-device, the obtained operation parameters include: coal seam hardness coefficient , in Mohs hardness; power , in kilowatts; temperature , in degrees Celsius; operating rate ; gas concentration ; support resistance , in kilonewtons; maximum support resistance , in kilonewtons; motor load rate ; optimal load rate ; maximum load rate ; safety gas concentration limit ; dust concentration , in milligrams per cubic meter; standard dust concentration limit , in milligrams per cubic meter; coal mining depth , in meters; working face roof height , in meters; equipment power , in kN; maximum power , in kN; equipment operation age , in years; equipment designed working life , in years; coal mining efficiency , in tons per hour; maximum designed coal mining efficiency , in tons per hour; hydraulic oil temperature , in degrees Celsius; maximum allowable hydraulic oil temperature , in degrees Celsius; bearing vibration value , in mm / s; normal vibration value , in mm / s; number of historical window fault warnings , in times; cooling system efficiency ; pick wear degree , in mm; running time after last maintenance , in hours; recommended maintenance interval , in hours; new pick standard , in mm.

[0015] Further, step 1 specifically includes:

[0016] Obtain the operation data of each sub - device in the coal mining face;

[0017] Perform data analysis on the operation data of each sub - device, and predict the failure probability of each sub - device for various types of failures in the next time step, including: based on the operation data, calculate the device status index of each sub - device and the environmental interaction status index of each sub - device; according to the device status index, environmental interaction status index and the number of historical window fault warnings for each type of device for various types of failures, calculate the failure type correlation factor; according to the failure type correlation factor, predict the failure probability of each sub - device for various types of failures in the next time step.

[0018] Further, based on the operation data, the device status index of each sub - device is calculated as:

[0019] ;

[0020] Wherein, is the device status index of the th sub - device at the th time step.

[0021] Further, based on the operation data, the environmental interaction status index of each sub - device is calculated as:

[0022] ;

[0023] Among them, is the environmental interaction status index of the th sub-device at the th time step;

[0024] Furthermore, according to the device status index, the environmental interaction status index, and the historical window fault warning times of various types of faults for each device, calculate the fault type correlation factor:

[0025] ;

[0026] Among them, is the fault type correlation factor of the th type of fault of the th sub-device at the th time step; when , it indicates a communication fault; when , it indicates a device fault; is the historical window fault warning times of the th sub-device.

[0027] Furthermore, according to the fault type correlation factor, predict the fault probability of each sub-device having various types of faults at the next time step as:

[0028] ;

[0029] Among them, is the fault probability of the th type of fault of the th sub-device at the th time step; is the fault type coefficient. When , is 1. When , is 2.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] The present invention can realize comprehensive status monitoring of devices. By introducing the device status index and the environmental interaction status index, not only the operating status of the device itself is considered, but also the impact of environmental factors on the device is systematically evaluated. For example, the impact of environmental factors such as coal seam hardness and gas concentration on device load and fault risk is comprehensively considered, thereby improving the accuracy of fault prediction.

[0032] Secondly, by combining the analysis of historical fault data, it is possible to better predict the types of faults that may occur in the future and their occurrence probabilities for the equipment. By updating the fault prediction model of the equipment in real time, the system can dynamically adjust the prediction results according to the operating status of the equipment to ensure that the prediction always remains accurate and sensitive.

[0033] In addition, through cloud computing and virtual mirroring technologies, the present invention can quickly isolate the faulty equipment when a fault occurs in the equipment and generate a virtual mirror to ensure that production is not affected. This intelligent and automated fault handling method reduces the equipment downtime and improves the safety and efficiency of the coal mining operation. When it is determined that a communication fault or equipment fault poses a threat to the original network, the structure of the original network can be adjusted to the network mirror formed by the actual communication connections of each sub-device in the standby network to complete the upgrade of the original network. This operation can repair the original network without completely shutting down the machine and improve the fault tolerance and stability of the system by adjusting the network topology structure, thereby optimizing the communication and equipment operation efficiency of the entire coal mining face. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flowchart of the method for automatically locking the target of video face swapping for target tracking proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0036] Referring to Figure 1 As shown, the method for system-level fault prediction and structure upgrade of the coal mining face in the embodiment of the present invention includes:

[0037] Step 1: Obtain the operation data of each sub-device in the coal mining face, perform data analysis on the operation data of each sub-device, and predict the fault probability of each sub-device having various types of faults in the next time step; the types of faults include: communication faults and equipment faults;

[0038] First, in the coal mining face, each sub-device (such as a coal shearer, a fan, a conveyor belt, etc.) is equipped with a variety of sensors, which will collect the working data of the equipment in real time, such as various parameters such as power, temperature, vibration, load, gas concentration, dust concentration, and support resistance. Through these sensors, the collected data can reflect the current working state of the equipment and various abnormal conditions that may occur during the long-term operation of the equipment. Therefore, the high-frequency real-time update of sensor data is the basis for fault prediction. These data not only reflect the immediate state of the equipment but also carry the historical working mode and potential change trends.

[0039] Next, by processing and analyzing the collected raw data, the system will establish a mathematical model based on the known operation rules and historical data of the equipment to predict the possible failure types and probabilities of each sub-equipment. In this process, the system will first calculate the equipment status indicators of each equipment, and combine the operating environment of the equipment to calculate the environmental interaction status indicators of the equipment. These two indicators play a crucial role in failure prediction. The equipment status indicators are usually obtained by weighted calculation of the key parameters (such as power, temperature, load, etc.) during the operation of the equipment, reflecting the load and stress conditions of the equipment under the current operating conditions. The environmental interaction status indicators evaluate the interactive effects between the equipment and its working environment. Factors such as external climate conditions and working surface environment will affect the operation of the equipment, and these factors need to be combined with the status of the equipment for comprehensive analysis.

[0040] However, relying solely on the analysis of the equipment status and environmental status is not enough. To improve the accuracy of prediction, the system also needs to conduct further analysis by combining the historical data of each failure type. These historical data include the frequencies and patterns of failures that occurred in each sub-equipment within different time windows. By analyzing these historical window data, the system can calculate the failure type correlation factor, which combines the equipment status, environmental interaction status, and historical failure warning data to comprehensively evaluate the likelihood of different types of equipment failures. For example, the historical failure records of a certain equipment may show that the probability of equipment failure is relatively high under a certain specific load condition, while it is relatively low under other load conditions. In this way, the system can not only identify the current failure risks of the equipment but also make dynamic predictions about future failures, discovering potential dangers in advance. Further, based on these analysis results, the system will predict the failure probabilities of each sub-equipment for various types of failures in the next time step. The key to this process is the establishment of the failure probability model. By applying advanced machine learning algorithms and statistical methods, the system can calculate the occurrence probabilities of each failure type (such as equipment failure or communication failure) based on the known equipment status, environmental interaction status, and historical failure records. This probability prediction can help operators take preventive measures in advance, such as adjusting the equipment operation strategy, optimizing the equipment load, and performing maintenance in advance, thus avoiding potential failures.

[0041] Step 2: If the probability of predicting that the sub-equipment will have an equipment failure in the next time step exceeds the set first failure threshold, isolate the sub-equipment, generate a corresponding virtual mirror device for the sub-equipment in the cloud to simulate the operation of the sub-equipment, and communicate and interact with other sub-equipments;

[0042] First, the generation of virtual mirror devices relies on the application of cloud technology. Cloud computing provides strong support for the real-time generation and update of virtual mirrors. When a sub-device is diagnosed with a high risk of failure, the system quickly creates a virtual mirror of the device in the cloud. This virtual mirror can not only replicate the functions and operations of the real device but also simulate the interaction behavior between the device and other sub-devices. The role of the virtual mirror is that it can replace the faulty device in real time to ensure that the system will not interrupt production due to the failure of a certain device. Especially in a complex and high-risk environment such as a coal mining face, equipment failures often lead to a significant decline in production efficiency and may even cause more serious safety accidents. Through the technical means of virtual mirror devices, the entire system can quickly isolate and replace faulty devices without affecting the production process.

[0043] The generation of virtual mirror devices is not just a simple replacement process. In fact, it uses simulation technology to simulate the reactions and behaviors of devices under different working conditions. Specifically, the virtual mirror device will dynamically adjust its internal parameters and operation logic according to the operating data of the real device and changes in the working environment to reproduce the actual working state of the device as realistically as possible. An important difference between the virtual mirror and the real device is that the virtual mirror does not directly participate in physical production but maintains the normal operation of the production system by exchanging data and controlling interactions with other devices in the system. In this way, the system can not only avoid downtime problems caused by equipment failures but also provide time and space for the repair and maintenance of subsequent equipment.

[0044] In addition to isolating and replacing faulty devices, virtual mirror devices also provide favorable conditions for the subsequent repair and inspection of faulty devices. During the actual repair process of a faulty device, the virtual mirror can seamlessly take over all operations of the device to ensure that production is not affected by the outage of the device. At the same time, engineers can analyze the root cause of the equipment failure in depth and perform more precise repairs and optimizations without affecting the production progress. The generation of such virtual mirror devices can not only avoid economic losses caused by equipment downtime but also effectively reduce the interference to the production process during equipment maintenance.

[0045] The application of virtual mirror technology is also very different from traditional fault repair methods. In the past, when equipment failed, it usually needed to be shut down and a large amount of manual repair was required, which not only caused production interruptions but also might lead to a series of safety hazards. The virtual mirror device in the present invention can be generated in real time in the cloud, completely independent of the time limit of physical repair, and avoids many problems in the traditional repair process. The quick replacement and real-time interaction capabilities of the virtual mirror make the operation of the entire coal mining face more flexible and efficient.

[0046] This fault isolation method based on virtual mirroring can not only improve the emergency response ability of the system, but also enhance the adaptive ability of the system. In the face of changing working environments and equipment loads, the virtual mirroring device can automatically adjust its operating mode and operation strategy according to the changes in real-time data, ensuring the stability and security of the entire system. The dynamic adjustment and precise simulation of the virtual mirroring device make the replacement process of faulty devices smoother and seamless, greatly improving the system's tolerance and adaptability to faults.

[0047] Step 3: If the probability that the predicted sub-device will have a communication fault in the next time step exceeds the set second fault threshold, control the sub-device to call the backup network to send or receive data; the backup network is a communication network with a blockchain structure, and in the backup network, each sub-device is interconnected with each other; when the number of sub-devices that call the backup network in the coal mining face exceeds the set quantity threshold, control all sub-devices to call the backup network, and then repair the original network.

[0048] First of all, the communication network plays a crucial role in the coal mining face system. The real-time data transmission and information interaction between devices are the basis for ensuring the efficient operation of the entire system. Especially in a dynamic environment where the operating states of devices and the external environment are constantly changing, the stability of the communication system is directly related to the smooth progress of the production process. When the system discovers through data analysis that the probability of communication faults in a certain sub-device or multiple sub-devices exceeds the set threshold, it is necessary to immediately activate the backup network to prevent further problems caused by communication interruptions. Traditional communication networks often rely on a centralized architecture, and once a part of the network fails, it is easy to cause the paralysis of the entire system. However, the present invention adopts a backup network with a blockchain structure, and this decentralized network design greatly enhances the fault tolerance and security of the system.

[0049] The activation of the backup network is not just a simple "backup" process. It is actually an intelligent fault transfer mechanism. When multiple sub-devices have communication faults and the number exceeds the preset threshold, the system will automatically switch to the backup network and control all sub-devices to perform data transmission through the backup network. The decentralized feature of the blockchain network enables each sub-device to be interconnected with other devices as an independent node, thus avoiding the impact that a single point of failure in a traditional centralized network may bring. In the backup network, all sub-devices are interconnected with each other, and each device can share data through the blockchain network to achieve the secure transmission of information. The immutability and consensus mechanism of the blockchain ensure the integrity and credibility of communication data, greatly improving the security and stability of system communication.

[0050] Different from traditional backup networks, the introduction of blockchain technology enables the backup network to not only be limited to the recovery of data transmission. It can also, through a decentralized mechanism, ensure that when a network fails, each sub-device can autonomously and seamlessly switch to the backup network without relying on a single server or control center. This decentralized design avoids the risk of single-point failures, enabling other devices to still operate normally even when some devices or communication lines have problems, ensuring the overall stability of the system.

[0051] When the backup network is activated, the system will monitor the status of each sub-device in real time to ensure the continuity and stability of the data stream. In the backup network, each sub-device can not only send and receive data but also interact with other devices through the network and participate in the operation of the entire system. This seamless switch ensures the high reliability and efficiency of the communication system, enabling the system to continue operating without interruption and production tasks to still be completed smoothly even when some devices have communication failures.

[0052] However, the activation of the backup network is not just a temporary measure to cope with failures. It also provides strong support for the repair and optimization of the system. During the operation of the backup network, the system can repair and adjust the original network. In traditional responses to communication failures, the repair process often takes a long time and is prone to the risk of system downtime. In the method of the present invention, the activation of the backup network provides sufficient time for the repair of the original network. During the repair process, the structure of the original network will be adjusted to the network mirror formed by the actual communication connections of each sub-device in the backup network, completing the structural upgrade of the original network. When the devices in the backup network find the shortest communication path through the routing algorithm, the original network will be gradually restored to the normal operating state, finally completing a smooth transition from the backup network to the original network. Through this intelligent repair mechanism, the system can minimize the impact caused by communication failures and ensure the stability and safety of the coal mining face system.

[0053] Furthermore, when all sub-devices in the coal mining face call the backup network, all sub-devices find the shortest communication paths with other sub-devices through the routing addressing algorithm.

[0054] First of all, the basic goal of the routing addressing algorithm is to find the shortest path from one device to another. Since each device in the backup network is an independent node and the connection between each device is realized through the blockchain network, the communication between devices does not depend on a single central node but is carried out through a distributed and decentralized network structure. In this way, the routing addressing algorithm must be able to dynamically adjust the path based on the status of each device in the network and ensure that the data transmission in the network is both fast and secure.

[0055] In this process, the selection of the routing and addressing algorithm is crucial. The most common routing algorithms include the distance vector routing algorithm and the link state routing algorithm, which determine the optimal path by calculating the distance between devices or the connectivity status of each device respectively. In the backup network of the present invention, considering the decentralized characteristics of the network, an improved version of the Shortest Path First (SPF) algorithm may be adopted. For example, the Dijkstra algorithm is used to find the shortest communication path between each pair of devices. The Dijkstra algorithm repeatedly updates the shortest path from each device to the target device in the network graph and finally determines an optimal path from the source device to the target device. In this way, in the backup network, the communication between any two sub-devices can be carried out through a shortest and lowest-latency path.

[0056] It should be noted that the backup network is not just a simple connection of physical devices. It also needs to handle possible dynamic failures and network changes through intelligent routing and addressing. Due to the extremely complex working environment of the coal mining face, the device status and communication conditions change at any time. Some devices in the network may fail, or the network bandwidth or transmission quality may decrease due to changes in the external environment (such as temperature, humidity, etc.). Therefore, the routing and addressing algorithm needs to have a high degree of adaptability, be able to respond to changes in the network state in real time, and dynamically adjust the communication path according to the latest device connection situation. For example, when a sub-device fails or its communication link is interfered with, the routing algorithm can immediately select other available paths to ensure that data transmission is not interrupted. In this way, even if a device failure or link interruption occurs in the backup network, the communication of the entire system can still proceed smoothly.

[0057] In addition, the routing and addressing in the backup network also need to consider the factor of load balancing. In practical applications, some devices may generate a relatively high load due to long-term data transmission or processing of a large amount of information, resulting in a decrease in data transmission speed. The routing and addressing algorithm not only needs to find the shortest path but also needs to consider the load situation of each node in the network and select paths with lower load and faster transmission speed to avoid overloading of some devices and ensure the stability and efficiency of the entire backup network. The load balancing mechanism can be achieved through multi-path selection, that is, selecting an optimal path among multiple feasible paths or dispersing the data flow to multiple paths to improve the overall network efficiency.

[0058] In the standby network, the routing and addressing algorithm also needs to have the ability of self - repair. Since there may be short - term dynamic changes in the network and some links may be temporarily unavailable, the routing algorithm needs to quickly detect these changes and restore the normal operation of the network by recalculating the optimal path. Specifically, when the system detects a fault in a certain communication link, the routing algorithm will automatically search for other available paths, quickly update the communication path, and direct the data traffic to the new path to avoid communication delays or data loss caused by link interruptions.

[0059] In the above - mentioned way, the routing and addressing algorithm can ensure that the communication between all sub - devices in the standby network always remains efficient and stable. Especially in a complex environment such as a coal mining face, factors such as equipment failures and environmental changes often affect the stability of the communication system. Therefore, an intelligent, efficient and adaptive routing and addressing algorithm is the key to ensuring that the standby network plays its maximum role. Through this algorithm, the entire coal mining face system can quickly resume normal operation in the face of sudden failures or communication obstacles, ensuring the continuity and safety of production.

[0060] Furthermore, when repairing the original network, the structure of the original network is adjusted to be a mirror image of the network formed by the actual communication connections of each sub - device in the standby network, completing the upgrade of the original network structure.

[0061] First of all, the goal of repairing the original network is to ensure that the original network can resume normal operation after a communication failure and can draw on the structure of the standby network during the recovery process to improve the fault tolerance and reliability of the system. In the standby network, due to its decentralized design, the communication between each sub - device and other devices is interconnected through nodes constructed by blockchain technology, and this decentralized structure provides a more flexible communication method than traditional networks. When the original network is repaired, the system will extract the actual communication connections of each sub - device from the standby network as a reference to ensure that the repaired network can adapt to the current communication needs and fault environment of the devices.

[0062] To achieve this process, the structure of the original network needs to be adjusted according to the actual connection situation in the standby network. Specifically, the system will automatically identify the connection paths between various sub-devices in the standby network and reconfigure the device connection methods in the original network based on these paths. During this process, the repaired original network will no longer rely on the traditional fixed structure but be reconstructed based on the optimal communication paths in the standby network. This means that by mirroring the structure of the standby network, the system can improve the communication efficiency and stability while restoring the original network. In addition, the key to this repair process lies in how to dynamically adjust the topology of the original network according to the communication connection situation in the standby network. Traditional communication networks usually use static connection methods, and the communication paths between devices are determined at the design stage and cannot flexibly adapt to device failures or network topology changes. In the present invention, however, the dynamic connection mechanism of the standby network enables the system to automatically adjust the connection relationship between devices according to factors such as real-time data traffic, device load, and communication delay. When the original network is repaired, it will refer to the structure of the standby network and adopt a more flexible and dynamic routing strategy to ensure that the communication paths between each sub-device always remain in an optimal state.

[0063] When performing network repair, the system will also optimize the load distribution of the original network according to the actual communication conditions of each device in the standby network. The device connections and communication traffic in the standby network are adaptively adjusted according to the real-time network conditions. This process can ensure that the system minimizes communication delay as much as possible when the load is low, and automatically distributes data traffic to the paths with lower load when the network load is high. By applying the load balancing strategy in the standby network to the repair process of the original network, the repaired original network will have better resource allocation capabilities and avoid the overload or bottleneck problems that may occur in traditional networks. The repaired original network is not simply "copied" from the structure of the standby network, but optimized and improved on the basis of the existing network architecture. By mirroring the communication paths of the standby network, the structure of the original network will be more flexible and able to automatically cope with various sudden failures and load changes. The advantage of this process is that it can greatly improve the stability and reliability of the network and avoid the risks of single-point failures and network interruptions in traditional network structures. In the complex environment of the coal mining face, communication between devices is a key factor, and any communication interruption may lead to the stagnation of production and even affect the safety of personnel. Therefore, by upgrading the original network structure to a decentralized and dynamically adjustable structure similar to the standby network, the system can quickly switch to the optimal communication mode when device failures or network problems occur, ensuring the continuity and safety of production.

[0064] The upgrade of the original network structure also includes improvements to network protocols and data transmission mechanisms. In the backup network, due to its decentralized characteristics, each device can maintain good communication connections with other devices through self-management and coordination. By referring to this feature, the repaired original network can be optimized in communication protocols, adopting more intelligent routing selection, data transmission, and error repair mechanisms, thereby improving the overall robustness of the network. In addition, the upgraded network can also monitor the network operation status in real time through a more intelligent control method, automatically identify and respond to potential communication problems.

[0065] Furthermore, the sub-devices at least include: coal shearers, fans, conveyor belts, hydraulic pumps, and valves.

[0066] First of all, the coal shearer is one of the core devices in the coal mining face, and its function is to directly cut and excavate the coal seam. The coal shearer needs to operate in a high-load and high-intensity working environment. Therefore, any minor fault may lead to the stagnation of the production progress or damage to the equipment. The stable operation of the coal shearer depends on the coordinated work of its power system, drive system, cutting system, etc. In the fault prediction system of the present invention, the operating state of the coal shearer is monitored in real time through a variety of sensors, including parameters such as temperature, vibration, power, and load. Through the analysis of these data, the types of equipment faults that the coal shearer may encounter in the future can be predicted, such as motor faults, drive system faults, cutting system blockages, etc.

[0067] The fan plays a crucial role in the coal mining face, mainly responsible for providing power for the ventilation system, ensuring the air circulation in the mine, and preventing the accumulation of harmful gases (such as gas). The operation of the fan is directly related to the safety and air quality in the mine. After long-term operation, the fan may malfunction due to reasons such as excessive motor load, abnormal vibration, or bearing wear. In fault prediction, the key parameters of the fan include wind speed, current, vibration frequency, and temperature, etc. These parameters can help the system evaluate whether the fan is in a normal working state. If it is found that the probability of the fan malfunctioning is relatively high, the system can predict and isolate the fan to ensure that the mine ventilation is not affected and avoid safety accidents caused by gas accumulation.

[0068] The conveyor belt is a transportation tool in the coal mining and excavation process, responsible for transporting the excavated coal from the excavation site to the next link. Since the conveyor belt needs to operate under long-term and high-load conditions, its faults usually lead to the stagnation of coal mining operations. Common conveyor belt faults include belt breakage, motor overload, conveyor belt deviation, etc. The operating state of the conveyor belt is monitored by multiple sensors, such as tension sensors, temperature sensors, and current sensors, etc. Through the data of these sensors, the system can evaluate the load and wear of the conveyor belt in real time, predict possible faults, and take isolation measures in time to avoid affecting the production of the entire coal mining face.

[0069] Hydraulic pumps and valves are mainly used in coal mining faces to provide hydraulic power and drive various mechanical equipment (such as supports, shearers, etc.). Faults in the hydraulic system usually lead to abnormal operation of the equipment and even pose safety hazards. The operating state of the hydraulic pump is closely related to parameters such as its oil temperature, pressure, and flow rate. The control accuracy and response time of the hydraulic valves also affect the normal operation of the equipment. Therefore, real-time monitoring and fault prediction of the hydraulic system are particularly important. By monitoring the key parameters of the hydraulic pump and valves, the system can identify in advance problems such as leaks, overloads, or control failures that may occur in the hydraulic system. Timely fault prediction and isolation can effectively prevent the impact of hydraulic system failure on the entire coal mining operation.

[0070] The coordinated operation of these sub-devices forms the basis of the coal mining face system. The failure of each device will affect the efficiency and safety of the entire system. Therefore, the coal mining face system must be able to monitor the operating state of these sub-devices in real time and conduct fault prediction. In the present invention, by analyzing the states of each sub-device in real time, the system can predict the probability of each sub-device failing at the next time step and take preventive measures in a timely manner. For example, when the fault probability of the shearer exceeds the threshold, the system can quickly isolate it and generate a virtual mirror device to simulate its operation to ensure that the coal mining operation will not be stalled. Similarly, if there is a risk of failure of the fan or conveyor belt, the system will activate the backup network in advance for data transmission to avoid the impact of communication interruption.

[0071] Furthermore, for the th sub-device, the operating parameters obtained include: the coal seam hardness coefficient , in Mohs hardness; the power , in kilowatts; the temperature , in degrees Celsius; the operating rate ; the gas concentration ; the support resistance , in kilonewtons; the maximum support resistance , in kilonewtons; the motor load rate ; the optimal load rate ; the maximum load rate ; the safety gas concentration limit ; the dust concentration , in milligrams per cubic meter; the standard dust concentration limit , in milligrams per cubic meter; the coal mining depth , in meters; the working face roof height , in meters; the equipment power , in kilonewtons; the maximum power , in kilonewtons; the equipment operating age , in years; the designed working life of the equipment , in years; coal mining efficiency , in tons per hour; the maximum designed coal mining efficiency , in tons per hour; the temperature of the hydraulic oil , in degrees Celsius; the maximum allowable temperature of the hydraulic oil , in degrees Celsius; the bearing vibration value , in millimeters per second; the normal vibration value , in millimeters per second; the number of fault warnings in the historical window , in times; the efficiency of the cooling system ; the wear degree of the pick , in millimeters; the running time after the last maintenance , in hours; the recommended maintenance interval , in hours; the standard of new picks , in millimeters.

[0072] Coal seam hardness coefficient (unit: Mohs hardness) is an important physical parameter in the process of coal mining. It reflects the hardness of the coal seam. The greater the hardness of the coal seam, the greater the load on the coal mining equipment, which may lead to easier wear or overload of the equipment. Therefore, is a key factor used to predict equipment load, wear and faults. Power (unit: kilowatt) is the electricity consumed by each equipment during operation. It directly reflects the working load of the equipment. The fluctuation of power can be used to analyze whether the equipment is in an overloaded state or there is a problem of decreased operating efficiency. For example, if the power of the equipment remains high during normal operation, it may indicate that there is a fault inside the equipment or other abnormal working conditions.

[0073] Temperature (unit: degrees Celsius) is an important equipment health indicator. Excessive or too low temperature may have an adverse impact on the performance and life of the equipment. Especially in equipment sensitive to temperature such as hydraulic systems, motors or transmission systems, the temperature exceeding the normal range usually means that there may be problems such as overheating, poor lubrication or failure of the heat dissipation system. Operating rate is a measure of the movement speed or working efficiency of the equipment. For example, for a coal shearer, it refers to the rotation speed of its cutting tool or the cutting depth of the coal seam. This parameter reflects the working load and coal mining efficiency of the equipment. When the rate is abnormal, it is usually necessary to check whether there is equipment failure, increased system resistance or other working environment factors.

[0074] Gas concentration Refers to the gas concentration around the equipment, especially the gas concentration in coal mines. Excessive gas concentration poses a great risk to the normal operation of the equipment and may lead to serious safety accidents such as explosions. Therefore, gas concentration is an important parameter that cannot be ignored in the safe operation of the equipment. Support resistance (Unit: kilonewton) and the maximum support resistance (Unit: kilonewton) reflect the pressure borne by the support, which is crucial for the stability of the support system. When the support resistance approaches the maximum value, it may indicate that the support is already under excessive pressure and there is a risk of failure.

[0075] Motor load rate Represents the ratio of the equipment motor load to the maximum load capacity, and is usually used to evaluate the working state of the motor. Optimal load rate and the maximum load rate are the expected values and the maximum safe load values during the normal operation of the equipment. When the load rate exceeds these ranges, the system needs to be vigilant about motor overload or other potential failures. Safe gas concentration limit and dust concentration (Unit: milligram per cubic meter) are important environmental parameters related to mine ventilation and safety. If the gas or dust concentration exceeds the set safety value, it may cause equipment failures or safety hazards, affecting the safety of coal mining operations.

[0076] Coal mining depth (Unit: meter) and the working face roof height (Unit: meter) are important physical parameters related to the mining operation environment. As the coal mining depth increases, the working load of the equipment usually increases, and the resistance faced by the shearer may also increase. In addition, the height of the working face roof affects the use efficiency of the support and the stability of the equipment. Equipment power (Unit: kilonewton) and the maximum power (Unit: kilonewton) are used to evaluate the output power generated by the equipment when performing tasks. Changes in equipment power may reveal the working efficiency or operation problems of the equipment. If the equipment power remains at a high or low level for a long time, it may indicate equipment overload or reduced efficiency.

[0077] Equipment operation age and the designed working life (Unit: year) are important parameters related to the service life of the equipment. The longer the equipment is used, the higher the risk of failure. Therefore, these parameters are crucial for predicting long-term equipment failures and maintenance requirements. Coal mining efficiency (Unit: ton per hour) is a standard indicator for measuring the working efficiency of the equipment, reflecting the production capacity of the coal mining equipment per unit time. Maximum designed coal mining efficiency It represents the maximum working capacity of the device, and exceeding this value may cause device failure.

[0078] Hydraulic oil temperature (unit: degree Celsius) and the maximum allowable hydraulic oil temperature (unit: degree Celsius) are used to monitor the working state of the hydraulic system. An excessively high oil temperature in the hydraulic system usually means that the hydraulic pump or other components may face problems such as failure or reduced efficiency. Bearing vibration value (unit: mm / s) and the normal vibration value (unit: mm / s) are important parameters for detecting the mechanical state of the device. Excessive vibration usually means that there is wear or looseness in the device, and inspection and maintenance may be required immediately.

[0079] Number of fault warnings in the historical window (unit: times) reflects the fault history of the device and provides the frequency and trend of device failures. This data can help the system evaluate the long-term stability and fault probability of the device. Cooling system efficiency refers to the ability of the cooling system to keep the device temperature within the normal range. Low cooling efficiency may cause the device to overheat and increase the risk of failure. Degree of pick wear (unit: mm) and the running time since the last maintenance (unit: hours) are used to evaluate the wear degree and maintenance cycle of the shearer. Excessively worn picks will directly affect the coal mining efficiency and the safety of the device. New pick standard (unit: mm) is the standard wear degree of the pick at the first use to ensure that the device is in good working condition.

[0080] Further, step 1 specifically includes:

[0081] Obtain the operation data of each sub-device in the coal mining face;

[0082] Perform data analysis on the operation data of each sub-device to predict the fault probability of each sub-device having various types of faults in the next time step, including: based on the operation data, calculate the device state index of each sub-device and the environmental interaction state index of each sub-device; according to the device state index, environmental interaction state index, and the number of fault warnings in the historical window of each type of device having various types of faults, calculate the fault type correlation factor; according to the fault type correlation factor, predict the fault probability of each sub-device having various types of faults in the next time step.

[0083] When obtaining the operation data of each sub-device in the coal mining face, the system needs to monitor various working parameters of each sub-device in real time. These parameters cover the working state of the device, environmental factors, and the interaction between the device and the external environment. Usually, the system uses sensors to collect these data. The sensors can measure physical quantities such as temperature, pressure, power, load, current, vibration, gas concentration, etc., and can also collect parameters closely related to the working environment such as coal seam hardness, coal mining depth, and device operation rate. All these parameters are key information for reflecting the current working state of the device and are also the basic data source for fault prediction. Next, the system will perform data analysis on the operation data of each sub-device, specifically including two aspects of calculation and analysis. First is the calculation of the device status index. The device status index reflects the performance and health status of the device and can reveal whether there are problems such as overload, overheat, and wear. For example, the device status index of the shearer can be obtained through the comprehensive calculation of parameters such as power, temperature, and load. Through the weighted analysis of these parameters, the system can evaluate the health status of the device under different working conditions and detect potential fault hazards in advance. Second, the system also needs to calculate the environmental interaction status index of each sub-device. The environmental interaction status index considers the interaction between the device and its working environment, such as the influence of factors such as coal seam hardness, gas concentration, dust concentration, and working face roof height. These environmental factors are directly related to the working load and risk of the device, so they need to be analyzed and predicted separately.

[0084] According to the calculation results of the device status index and the environmental interaction status index, the system will then calculate the fault type correlation factor in combination with the number of historical window fault warnings. The fault type correlation factor is a comprehensive index that combines the operation data of the device, environmental factors, and historical fault data to reflect the potential probability of different types of faults occurring in the device. Specifically, the fault type correlation factor is jointly affected by multiple factors, including the device status index , the environmental interaction status index , and the frequency of historical faults of the device For example, the worse the workload and environmental conditions of the device are, and the higher the frequency of historical failures, the greater the probability of the device failing. This factor is crucial in fault prediction as it helps the system identify which devices are likely to experience a specific type of failure in the future. The calculation result of the fault type correlation factor will be used to predict the fault probability of each sub-device for various types of failures in the next time step. By applying advanced statistical and machine learning methods, the system can predict the probability of device failures within a certain period in the future based on the value of the fault type correlation factor. For example, when the fault type correlation factor is high, the system will determine that the probability of the device failing is large, and vice versa. This prediction process can promptly identify potential faulty devices and provide a basis for subsequent fault prevention and repair.

[0085] In this prediction process, the system not only relies on static device data but also dynamically adjusts the prediction model according to the changes in the device's operating state. For example, as the operating time of the device increases or the environmental conditions change, the device state and fault probability will also change accordingly. Therefore, it is necessary to update the prediction results in real-time based on the latest operating data. In addition, the system also needs to consider the synergy between multiple sub-devices and system-level fault prediction to ensure a comprehensive and integrated analysis of the fault prediction for the entire coal mining face system. Specifically, the analysis model in step 1 combines three factors: device state, environmental state, and historical fault data, forming a multi-dimensional prediction model. This model can help mine managers understand the possible future fault conditions of each sub-device and take corresponding preventive measures based on these predictions. For example, for a certain device, when the prediction result shows a high probability of device failure, the system will automatically start a fault isolation program, adjust the device's workload in a timely manner, or even generate a virtual mirror device to replace the faulty device to ensure that the entire coal mining face system is not affected.

[0086] Furthermore, based on the operating data, the device state index of each sub-device is calculated as:

[0087] ;

[0088] where is the device state index of the th sub-device at the th time step.

[0089] In the formula, the device state index is obtained by performing a weighted calculation by combining multiple device operation and environmental factors. First, the coal seam hardness coefficient reflects the hardness of the coal seam, which is one of the core factors affecting the operating load of the equipment. An increase in the coal seam hardness means that the cutting resistance faced by the coal mining equipment during operation also increases, which will lead to an increase in the working load of the equipment. The higher the coal seam hardness, the more likely the equipment is to experience problems such as overload and wear. Therefore, this parameter plays a crucial role in the formula. By introducing the coal seam hardness coefficient, the system can more accurately evaluate the working load and failure risk of the equipment under coal seams of different hardnesses. Then, the power also occupies an important position among the equipment status indicators. Power is a measure of the energy consumed by the equipment during operation, and it is directly related to the working load and efficiency of the equipment. When the equipment power is high, it usually means that the equipment is in a high-load state or the working environment is more severe. The increase in power may lead to excessive wear of the equipment and increase the risk of equipment failure. For example, when the shearer is running, if the power remains high for a long time, it may indicate that the coal seam hardness is large or there is a fault in the equipment, resulting in the equipment bearing an abnormal load. By introducing the factor of power into the equipment status indicators, the system can monitor the working load of the equipment in real time and predict whether the equipment will be overloaded or fail.

[0090] Equipment temperature is also an indispensable parameter among the equipment status indicators. Excessive temperature usually means that the equipment is overloaded or there is a problem with the equipment's heat dissipation system. For sensitive components such as motors and hydraulic systems, an increase in temperature will accelerate their aging and wear, and may even cause permanent damage to the equipment. Temperature is closely related to the operating efficiency of the equipment. Excessive temperature often means that the equipment is in an abnormal working state. By monitoring the equipment temperature, the system can detect the situation of excessive temperature in time and give an early warning, thus avoiding failures caused by equipment overheating. Operating speed is also an important factor affecting the health of the equipment. The speed reflects the working efficiency or operation speed of the equipment. Excessive operating speed often increases the load on the equipment, while too low speed may lead to low operation efficiency. Especially in coal mining operations, the equipment needs to work in a high-load and high-intensity environment. If the operating speed is not appropriate, it may lead to equipment failures or a decrease in production efficiency. By monitoring the operating speed of the equipment, the system can evaluate the working load of the equipment and provide a basis for adjusting the operation strategy to ensure that the equipment operates at the optimal load.

[0091] Support resistance and the maximum support resistance reflect the pressure borne by the support system, which is very crucial in coal mining operations. The operation of the shearer requires the support of the support to hold up the roof of the coal mine to prevent cave-ins. If the support resistance approaches or exceeds the maximum support resistance , which means there may be a risk of failure of the support, threatening the stability and safety of the equipment. By monitoring the support resistance, the system can detect the load condition of the support in real time and make timely adjustments when the support is overloaded to ensure the safety of the working area. Motor load rate is an important indicator of the motor's working state, which reflects the ratio between the motor load and its maximum load capacity. Excessive motor load usually means that the equipment exceeds the designed load-bearing capacity of the motor during operation, which may cause the motor to overheat, burn out or be damaged. Therefore, monitoring the motor load rate is crucial for the health management of the equipment. Optimal load rate and maximum load rate are the ideal and limit working loads designed for the motor. When the motor load rate exceeds these ranges, the risk of equipment failure increases significantly. Gas concentration and dust concentration are parameters closely related to environmental safety in the coal mining face. Excessive gas concentration increases the risk of explosion, while excessive dust concentration may cause equipment wear and even lead to health problems. In an environment with high gas or dust concentration, the operation of the shearer and other equipment may be severely affected, increasing the probability of failure. Therefore, the system needs to monitor these environmental parameters in real time to ensure that the equipment operates in a safe environment. Equipment status indicator is obtained by comprehensively analyzing these operation data. Its calculation method not only considers the operation state of the equipment itself but also the interaction between the equipment and the external environment. By integrating these parameters, the system can obtain a comprehensive evaluation of the equipment's health status, providing data support for fault prediction. High values usually mean that the equipment is in a high-load or abnormal working state, with a risk of failure, while low values indicate that the equipment is operating normally, with a lower probability of failure.

[0092] Furthermore, based on the operation data, the environmental interaction status indicator of each sub-equipment is calculated as:

[0093] ;

[0094] where is the environmental interaction status indicator of the th sub-equipment at the th time step;

[0095] Coal mining depth is a key factor affecting the working conditions of the equipment. In the coal mining face, as the coal mining depth increases, the working load that the equipment needs to bear will also increase accordingly. The greater the depth, the greater the hardness of the coal seam may be, and the working environment may also be more complex, which will lead to an increase in the pressure when the equipment is working. The coal mining depth is closely related to the equipment load and failure risk. Therefore, the change of the coal mining depth has a direct impact on the health status of the equipment. By introducing the coal mining depth parameter, the system can monitor the progress of the coal mining operation in real time and adjust the equipment operation strategy according to different depths, so as to prevent the failure risk brought by the increase in depth.

[0096] The height of the working face roof is a parameter related to the structure of the coal mining face. The roof height reflects the height of the working face space and is usually closely related to the design of the support system, the operation space of the equipment, and the equipment stability. In deep mines, the roof height may affect the stress condition of the support and the stability of the support system. The higher the roof height of the working face, the greater the load borne by the support system, and the greater the pressure faced by the equipment during operation. In the formula, through the processing of , an evaluation of the equipment operation under different roof heights is introduced. Especially in deep mine operations, the roof height directly affects the working load of the equipment and the potential failure risk.

[0097] The following part is the relationship between the equipment power and the maximum equipment power . The equipment power reflects the actual thrust or working force generated by the equipment during operation, and it is directly related to the load state of the equipment. As the operation load increases, the equipment power increases, which may lead to equipment overload and thus increase the probability of failure. By introducing the ratio of the equipment power to the maximum equipment power into the formula, the system can quantify the working state of the equipment under different working loads. Especially in the case of overload, the power of the equipment may approach its maximum bearing capacity, which means an increased risk of failure. The sine function is used in the formula to describe the relationship between the equipment power and the failure risk, indicating the impact of the ratio between the power and the maximum power on the failure prediction.

[0098] In addition, the parameters and represent a specific performance coefficient of the equipment and the actual operating age of the equipment respectively. The performance coefficient of the equipment reflects the gap between the current performance of the equipment and the design standard. The lower the performance coefficient, the worse the actual working efficiency of the equipment and the higher the failure risk. The operating age This directly affects the health status of the equipment. As the service life of the equipment increases, it usually faces problems such as wear and aging, which increases the probability of failures. By introducing the operating age and performance coefficient of the equipment, the system can adjust the fault prediction model in real time to reflect the fault risks of the equipment under different service years and working conditions.

[0099] In the formula, and are the standard performance coefficient and the designed working life of the equipment respectively. The standard performance coefficient refers to the working efficiency of the equipment under ideal conditions, while represents the theoretical working life designed for the equipment. The differences between the actual operating status of the equipment and these standard parameters can often reveal the health status of the equipment. If the performance coefficient of the equipment is significantly lower than the standard value, or the actual service life of the equipment is close to the designed life, the risk of failure is usually high.

[0100] Finally, the ratio of the coal mining efficiency and the maximum designed coal mining efficiency is also incorporated into the calculation. The coal mining efficiency directly reflects the working efficiency of the equipment. The lower the efficiency, the more likely it indicates that there are signs of failures, aging or improper operation of the equipment. The maximum designed coal mining efficiency is the maximum working capacity of the equipment under ideal conditions. By introducing the ratio of the coal mining efficiency to the designed efficiency, the system can evaluate the gap between the actual working capacity of the equipment and its designed capacity, and then infer the possible fault risks of the equipment. When the coal mining efficiency is close to the maximum designed coal mining efficiency, it usually indicates that the load of the equipment has approached its limit, and the probability of failure is also high.

[0101] Furthermore, according to the equipment status index, the environmental interaction status index, and the historical window fault warning times of each type of failure for each piece of equipment, the fault type correlation factor is calculated:

[0102] ;

[0103] where is the fault type correlation factor of the th sub-equipment for the th type of failure at the th time step; when , it indicates a communication failure; when , it indicates an equipment failure; is the historical window fault warning times of the th sub-equipment.

[0104] The equipment status index and the environmental interaction status index in the formula It is calculated by comprehensively considering multiple device operation and environmental parameters. The device status index reflects the overall health of the device, taking into account key operation parameters such as the device's power, temperature, load, and motor load rate. These parameters can reveal whether the device is in a state of potential faults such as overload, high temperature, and abnormal vibration. The environmental interaction status index reflects the impact of the external environment on the device operation, such as the combined effect of factors like coal mining depth, working face roof height, coal seam hardness, and gas concentration. These factors directly affect the device's workload and potential fault risks. Therefore, the device status index and the environmental interaction status index comprehensively measure the device's operation status under the current working conditions from two dimensions: the device itself and the external environment. The core role of the fault type correlation factor is to calculate the probabilities of different types of faults that the device may occur in the future based on the device status and environmental status, combined with the device's historical fault data. The first term in the formula is , which cubes the device status index . The role of this term is to amplify the impact of the device status index on the fault type correlation factor. If the value of the device status index is large (i.e., the device has a heavy load, high temperature, high power, etc.), this term will cause an increase in the fault type correlation factor, indicating a higher risk of the device failing. By cubing the device status index, the formula gives greater weight to the load status of the device, reflecting the importance of the device status in fault prediction.

[0105] The next second term is , which squares the environmental interaction status index after dividing it by 5. This term reflects the impact of the external environment on device faults. External environmental factors (such as coal seam hardness, gas concentration, working face roof height, etc.) play an important role in the device's working state. Especially in deep mines or high-risk areas, environmental changes may cause significant fluctuations in the device's load, thus increasing the likelihood of faults. By squaring the value, the formula emphasizes the role of environmental factors in device faults. Especially when the environment changes drastically, the device fault risk will increase significantly. The third term in the formula is , which combines two parameters: the hydraulic oil temperature and the bearing vibration value . The hydraulic oil temperature and the bearing vibration They are all important parameters reflecting the mechanical state of the equipment. An excessively high hydraulic oil temperature or abnormal bearing vibration usually indicates signs of wear, aging, or abnormal operation of the equipment. By introducing the hydraulic oil temperature and bearing vibration values, the formula can more accurately reflect the mechanical condition of the equipment. Especially in the case of high temperature or excessive vibration, the probability of equipment failure increases significantly. This item normalizes these parameters to ensure that the influence of the hydraulic oil temperature and vibration values has the same weight as other parameters.

[0106] Finally, in the formula reflects the number of historical fault warnings of the equipment on the influence of the fault type correlation factor. The number of historical fault warnings represents the frequency of equipment failures in the past period. Usually, for equipment with a relatively large number of historical fault occurrences, its fault risk is higher. By dividing the number of historical fault warnings by 100, adding 1, and then taking the square root, the formula appropriately amplifies the influence of the historical fault frequency of the equipment on the future fault risk. Historical fault data can not only help the system understand the past operating conditions of the equipment but also provide an important basis for future fault prediction. If the equipment has had multiple fault warnings in the past, the fault type correlation factor will gradually increase with the increase in historical data, thus reflecting the high probability of future equipment failures.

[0107] Furthermore, according to the fault type correlation factor, the fault probability of each sub-equipment having various types of faults at the next time step is predicted as:

[0108] ;

[0109] where the -th sub-equipment at the -th time step for the -th type of fault; is the fault type coefficient. When , is 1, and when , is 2.

[0110] The first part of the formula is: ; This part describes the influence of the fault type correlation factor on the fault probability. The fault type correlation factor reflects the indication of the current state of the equipment (including the equipment health state, environmental impact, and historical fault data) on future faults. By squaring it and dividing by 50, this operation amplifies the fault type correlation factor, making equipment with a larger fault correlation factor have a higher fault probability. The exponential function is a common non-linear function that can make when When it is relatively large, the failure probability increases rapidly, reflecting the non-linear characteristics of equipment failures. This term strengthens the prediction of future failures for larger failure type correlation factors, meaning that the probability of failure will increase sharply when the equipment is under high load, has a high failure history, or is in adverse environmental conditions.

[0111] Next is the second part of the formula: ; This part uses a typical S-shaped (Sigmoid) function to adjust the failure probability according to the value of the failure type correlation factor. The role of the Sigmoid function is to map the value of the correlation factor to a reasonable probability range (usually between 0 and 1). Among them, is the failure type coefficient. When , it represents the failure coefficient for communication failures and takes a value of 1; when , it represents the failure coefficient for equipment failures and takes a value of 2. By adjusting the value of , the system can provide different weights for different types of failures. For example, the impacts of communication failures and equipment failures in the system are different, so different weightings are performed for the calculation of the failure probabilities of these two types of failures, reflecting the adaptive adjustment of the system to different failure types. On the other hand, the expression represents the difference between the failure type correlation factor and a reference value of 1.5. This value is usually set empirically, and the activation point of the failure probability can be controlled by adjusting this reference. The shape of the Sigmoid function ensures that when is close to 1.5, the failure probability starts to change smoothly, and when is much greater than or much less than 1.5, the failure probability will quickly approach 1 or 0.

[0112] The third part is: ; This term represents the impact of the operating time or usage cycle of the equipment on the failure probability. represents the operating time of the equipment since the last maintenance, and is the maximum recommended operating time or service life of the equipment. By taking the square root of this ratio, the system can ensure that the impact of the equipment operating time on the failure probability is smooth. The failure probability of the equipment after long-term operation will gradually increase with the increase of the operating time, but the increasing speed is relatively slow (reflected by the square root function). This part reflects the impact of equipment aging. The longer the equipment is used, the gradually rising the probability of failure.

[0113] The last term is: ; This part reflects the impact of the wear degree of pick teeth on the failure probability of the equipment. The wear of pick teeth directly affects the cutting efficiency and operation ability of coal mining equipment. The more serious the wear, the higher the failure risk of the equipment. is the actual value of the current pick wear of the device, is the standard wear degree of a new pick, is the maximum allowable wear degree of the pick. By normalizing the wear degree to a ratio, the system can reflect the impact of the wear degree on equipment failures. The square root normalization operation compresses the impact of wear on the failure risk, such that when the equipment is excessively worn, the failure probability increases rapidly. Additionally, multiplying by a coefficient of 0.5 further controls the impact of this term on the failure probability, enabling it to not only reflect the significance of pick wear on the failure probability but also avoid having an overly large impact on the overall failure probability.

[0114] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification are only the principles of the present invention. Without departing from the spirit and scope of the present invention, various changes and improvements will occur to the present invention, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting system-level faults and upgrading structure of a coal mining face, characterized in that: include: Step 1: Obtain the operating data of each sub-equipment in the coal mining face, perform data analysis on the operating data of each sub-equipment, and predict the failure probability of each sub-equipment in the next time step for various types of failures; the failure types include: communication failure and equipment failure; Step 2: If the probability of a device failure in the next time step is predicted to exceed the set first failure threshold, the sub-device is isolated, and a virtual mirror device corresponding to the sub-device is generated in the cloud to simulate the operation of the sub-device and communicate and interact with other sub-devices; Step 3: If the probability of a communication failure in a sub-device in the next time step is predicted to exceed the set second failure threshold, the sub-device is controlled to call the backup network to send or receive data; the backup network is a communication network with a blockchain structure, in which each sub-device is interconnected; when the number of sub-devices calling the backup network in the coal mining face exceeds the set number threshold, all sub-devices are controlled to call the backup network, and then the original network is repaired; For The operating parameters of each sub-device include: coal seam hardness coefficient , unit is Mohs hardness; power , in kilowatts; temperature , in degrees Celsius; operating rate ; Gas concentration ; Bracket resistance , in kN; maximum support resistance , unit is kN; motor load rate ;Optimum load factor ; Maximum load rate ;Safe gas concentration limit ; Dust concentration , in mg / m3; standard dust concentration limit , in mg / m3; coal mining depth , in meters; height of top plate of working face , in meters; equipment power , in kilonewtons; maximum power , in kilonewtons; Equipment operating age , in years; equipment design service life , in years; coal mining efficiency , in tons / hour; maximum design coal mining efficiency , in tons / hour; hydraulic oil temperature , in degrees Celsius; Maximum allowable hydraulic oil temperature , in degrees Celsius; bearing vibration value , in mm / s; normal vibration value , in millimeters per second; number of fault warnings in the historical window , the unit is times; cooling system efficiency ; Wear degree of pick , in millimeters; running time since last maintenance , in hours; recommended maintenance interval , in hours; New pick standard , in millimeters; Based on the operation data, the equipment status index of each sub-equipment is calculated as: ； in, For the The sub-device is in The device status indicator for each time step.

2. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 1, characterized in that: When all sub-devices in the coal mining face call the backup network, all sub-devices find the shortest communication path with other sub-devices through the routing addressing algorithm.

3. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 2, characterized in that: When repairing the original network, the structure of the original network is adjusted to a mirror image of the network formed by the actual communication connections of each sub-device in the backup network, thereby completing the upgrade of the original network structure.

4. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 3, characterized in that: The sub-equipment includes at least: coal mining machine, fan, conveyor belt, hydraulic pump and valve.

5. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 4, characterized in that: Step 1 specifically includes: Obtain the operating data of each sub-equipment in the coal mining face; Data analysis is performed on the operating data of each sub-device to predict the failure probability of each sub-device having various types of failures in the next time step, including: calculating the device status index of each sub-device and the environmental interaction status index of each sub-device based on the operating data; calculating the failure type correlation factor according to the device status index, the environmental interaction status index and the number of historical window failure warnings of various types of failures of each device; predicting the failure probability of each sub-device having various types of failures in the next time step based on the failure type correlation factor.

6. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 5, characterized in that: Based on the operation data, the environmental interaction status index of each sub-device is calculated as: ； in, For the The sub-device is in The environmental interaction state indicator for each time step.

7. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 6, characterized in that: According to the equipment status indicators, environmental interaction status indicators and the number of historical window fault warnings of various types of faults for each equipment, the fault type correlation factor is calculated: ； in, For the The sub-device is in The time step The fault type correlation factor of the fault type; when When , it indicates a communication failure; when When , it indicates that the equipment is faulty; For the The number of historical window fault warnings for each sub-device.

8. The method for system-level fault prediction and structural upgrade of a coal mining face according to claim 7, characterized in that: According to the fault type association factor, the predicted fault probability of each sub-device having various types of faults in the next time step is: ； in, No. The sub-device is in The time step The failure probability of each type of failure; is the fault type coefficient, when hour, is 1, when hour, is 2.

Images