A mine optical cable state monitoring system

Abstract

The present application provides a kind of mine optical cable state monitoring system, including self-organizing perception module, optical cable health assessment module, fault identification module, automatic repair module and block chain storage module.The present application can solve the problem of blind area in traditional optical cable monitoring and repair system, information lag problem and the problem of low repair efficiency, realize full coverage, efficient, real-time, automated optical cable monitoring and repair.These innovations not only improve the safety and stability of mine optical cable, but also significantly reduce the cost of manual maintenance and system fault repair time, provide a new, intelligent management scheme for mine communication infrastructure.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of underground, and particularly relates to a mine optical cable state monitoring system. BACKGROUND

[0002] As a crucial communication and data transmission facility in mines, mine optical cables bear the responsibility of providing real-time information flow and communication connection for various devices and operations in mines. Especially in underground mines, optical cables are not only used as transmission media for real-time monitoring and data acquisition, but also play an important role in remote monitoring, environmental monitoring and dispatching. However, the particularity of mine environment leads to many potential risks and problems for mine optical cables. Factors such as temperature, humidity, air flow changes, geological activities and chemical corrosion in mines often cause optical cables to be exposed to extremely complex and unstable conditions, which pose a serious threat to the operational life, stability and safety of optical cables. Traditional optical cable state monitoring and fault detection methods mostly rely on manual inspection, regular maintenance or local sensor detection, which have problems of monitoring blind area and time lag.

[0003] Traditional mine optical cable monitoring technology usually monitors the running state of optical cables by installing single-point sensors, but these sensors can only cover a limited range in the vast and complex environment of mines, and cannot achieve real-time monitoring of the entire optical cable. The failure or failure of a single sensor can cause information loss and monitoring blind spots in the system. In addition, traditional manual inspection methods often face a series of problems such as low efficiency, long response time, personnel safety, and cannot find small damage in real time, resulting in many potential optical cable faults that cannot be discovered in the early stage. On the other hand, even if monitoring is carried out through traditional sensors, existing fault diagnosis techniques often have great limitations, such as the early state of optical cable damage being difficult to detect in time, and in many cases, the location of the fault cannot be accurately located, resulting in maintenance work having to rely on cumbersome positioning processes, increasing mine maintenance costs and downtime.

[0004] In order to overcome these technical defects, researchers have explored the possibility of combining various new technologies into the field of mine optical cable monitoring in recent years, such as distributed sensor networks, wireless sensor nodes, Internet of Things technology, etc. Although these technologies have improved the real-time and breadth of monitoring to some extent, they still have problems such as poor interconnectivity between systems, information processing lag, and data authenticity and security. For example, remote monitoring through the Internet of Things has made breakthroughs in data transmission, but due to the complex communication environment in the mine, especially in deep wells or complex rock formations, signal coverage is not comprehensive or network instability still occurs from time to time. In addition, the current mine monitoring system mostly focuses on information collection and transmission, and does not fundamentally solve the problem of repairing after the damage of the mine optical cable, and the self-repairing capability of the existing technology is almost zero, which also leads to the fact that once the optical cable of the mine equipment is damaged, it usually needs a long time of manual intervention or scheduling, affecting the production efficiency of the mine.

[0005] Although these existing technologies have promoted the progress of mine monitoring systems to some extent, they have not yet completely solved the core problem of mine optical cable health monitoring and self-repairing, i.e. how to realize real-time optical cable health monitoring with full coverage, high efficiency and low delay, and automatically repair when damage occurs. Therefore, a new type of mine optical cable state monitoring and repair system is urgently needed, which not only can monitor the health status of the mine optical cable in real time and accurately, provide comprehensive fault diagnosis, but also needs to have automatic repair function, reduce the need for manual intervention, improve the efficiency and accuracy of optical cable repair, and ensure the long-term stable operation of the optical cable. SUMMARY

[0006] The purpose of the present application is to provide a mine optical cable state monitoring system, which fundamentally solves the problems of blind area, information lag and failure to repair in time in the monitoring of the state of the mine optical cable.

[0007] In order to solve the above technical problems, the embodiment of the present application provides a mine optical cable state monitoring system, which comprises a self-organizing perception module, an optical cable health evaluation module, a fault identification module, an automatic repair module and a blockchain storage module.

[0008] The self-organizing perception module is used to obtain sensor node data of the optical cable in the mine, and to construct a regional grid self-organizing perception network based on the sensor nodes, and to preprocess the sensor node data, and to fuse the preprocessed data based on a preset neural network model to obtain an optical cable health evaluation value; wherein the sensor node data includes temperature, humidity, vibration and strain data.

[0009] The optical cable health assessment module is configured to calculate an upload frequency of a node thereof according to the health assessment value, perform adaptive bandwidth allocation according to the upload frequency of the node and a total upload frequency of all nodes, and obtain a bandwidth allocation of the node, and identify a fault node by performing difference analysis on a current health assessment value of the node and a historical health assessment value of the node and avoiding false positives, and automatically increase the upload frequency of the node based on an abnormality identification result and re-allocate bandwidth once the node is marked as abnormal.

[0010] The fault identification module is configured to determine fault positioning by performing difference analysis on a current health assessment value of a node and a historical health assessment value of the node and a change rate, obtain a regional fault probability graph, and determine a priority of a repair task based on the regional fault probability graph.

[0011] The automatic repair module is configured to enable a micro robot to automatically perform repair on a fault node according to the regional fault probability graph and the priority of the repair task, generate repair status data, and generate feedback data in combination with a sensor of the micro robot.

[0012] The blockchain storage module is configured to input the repair status data and the feedback data into a blockchain as a smart contract, and automatically manage the repair task and a reward mechanism through the smart contract.

[0013] As a preferred embodiment, the regional grid self-organizing perception network forms a distributed topology through a dynamic self-organizing protocol, wherein each node in the network transmits data through a low-power communication protocol, and can automatically adjust a transmission frequency under different environmental conditions in combination with an adaptive communication frequency control mechanism, and each node is responsible for collecting environmental data of a region where the node is located and exchanging data with adjacent nodes through a self-organizing mechanism to adjust a network topology in real time.

[0014] As a preferred embodiment, the preprocessing includes: removing random noise in data by using a sliding average filtering or Gaussian filtering method; identifying and removing outliers in sensor data by using a threshold method or a standard deviation method; smoothing time series data by using a Kalman filter; and performing local data fusion on data at each node to obtain preprocessed data.

[0015] The preset neural network model structure is as follows:

[0016] An input layer: preprocessed data of each sensor;

[0017] A hidden layer: a nonlinear activation function is used to process complex mutual relationships between different sensor data and learn contribution weights of the data to health assessment;

[0018] An output layer: an output optical cable health assessment value, which is used to represent a health condition of the optical cable.

[0019] And through L2 regularization, limit the growth of neural network model weight, improve the generalization ability of neural network model.

[0020] As a preferred embodiment, the uploading frequency of the node is calculated according to the health assessment value, and the adaptive bandwidth allocation is performed according to the uploading frequency of the node and the total uploading frequency of all nodes to obtain the bandwidth allocation of the node, including:

[0021] According to the optical cable health assessment value, the uploading frequency is calculated in the form of Sigmoid function;

[0022] According to the uploading frequency, the total uploading frequency of all nodes in the current network is combined to obtain the bandwidth allocation corresponding to the current node;

[0023] The identification of the fault node includes:

[0024] Based on the difference analysis result of the current health assessment value and the historical health assessment value of the node, the change rate is calculated through trend analysis, and if the change rate of the change of the node health assessment value is greater than a preset threshold, the node is considered to have an abnormality and enters a warning state.

[0025] As a preferred embodiment, the false positive is avoided, including:

[0026] In the difference analysis, a regularization anomaly detection method is introduced, and the abnormal fluctuation of each node is controlled by introducing a regularization term ; wherein, the regularization term is expressed as: ;

[0027] wherein, is the health assessment fluctuation standard deviation of node , indicating the normal fluctuation range of the node; is the rate of change of the node health assessment value; is the total number of nodes.

[0028] As a preferred embodiment, the fault identification module performs steps including:

[0029] The difference analysis result and the change rate are obtained, a fault location model is constructed, and the position weight of the node is obtained; wherein the position weight of the node of the fault location model is related to the abnormal degree of the node and the position thereof;

[0030] The position weights of all nodes are integrated to obtain a regional fault probability graph , represents the probability of failure occurring in each region in the mine; the regional failure probability map is calculated by superimposing the node weights:

[0031] ;

[0032] wherein is an indicator function, indicating whether the coordinates of the node are within the current region ; is the number of all sensor nodes in the mine

[0033] As a preferred embodiment, when the failure region is determined, the repair task is automatically triggered, and the priority of the repair task is automatically determined according to the location of the failure region , and the calculation formula is: ;

[0034] wherein, is a scheduling sensitivity coefficient, controlling the relationship between the regional failure probability and the task location in task scheduling; represents the distance between the execution location of the repair task and the failure region;

[0035] If the priority of the repair task exceeds the set task triggering threshold, the repair task is automatically scheduled and the relevant personnel are notified.

[0036] As a preferred embodiment, the repair state data generated by the micro robot performing the repair on the failure node according to the regional failure probability map and the priority of the repair task includes

[0037] The failure probability of the regional failure probability map and the minimization of the distance from the current position of the micro robot to the failure node are taken as the optimization objective, and a corresponding optimization objective function is established;

[0038] The sensor data carried by the micro robot is collected, and the accurate position of the damage and the severity of the damage area are determined by fusing the sensor data carried by the robot; the severity of the damage area is confirmed based on the regional failure probability map through the indicator function;

[0039] When the micro robot completes the repair task, the repair effect is detected through the built-in sensor, and the health status after repair is fed back to the mine management system; the health status after repair is determined based on the change rate of the health status before repair and the severity of the damage area.

[0040] As a preferred embodiment, each block of the blockchain includes: stored task data, a hash value pointing to the previous block, and a timestamp of task execution;

[0041] wherein, whenever a repair task is completed, a new block is added to the blockchain containing the full data of the task;

[0042] wherein, the smart contract will define the rules by:

[0043] When the severity of the new damage area exceeds the preset threshold, the smart contract is automatically generated and activated to create a new repair task;

[0044] The execution of the repair task is triggered by the robot, and a new repaired health state is generated after the task is completed, and the result is submitted to the smart contract to trigger the repair completion event;

[0045] The smart contract automatically verifies the execution of the repair task. If the new repaired health state is within a certain threshold and meets the expectation, the contract execution is successful, and the repair completion is recorded;

[0046] wherein, in the smart contract, the task state is updated with the progress of the repair task, including: task not started, task started but not completed, and task completed.

[0047] As a preferred embodiment, once the smart contract completes the verification of the repair task and confirms the successful execution of the task, the relevant data is recorded in the blockchain, and the smart contract further includes a reward mechanism, which automatically allocates rewards according to the execution quality of each repair task; wherein, the reward of the repair task is determined by the repair quality and the completion of the task, and is calculated as follows:

[0048] ;

[0049] wherein, and are adjustment coefficients to control the influence of health state and task state on the reward; is the health state of the repaired node, indicating the effect of repair; is the task state, reflecting the completion of the task.

[0050] The beneficial technical effects of the present application are at least as follows:

[0051] The present application first introduces the concept of regional grid deployment and self-organizing perception network, divides the monitoring area of the mine optical cable into multiple independent and adaptive grids. Each grid is equipped with multiple sensor nodes, which can dynamically adjust the data collection strategy and working frequency according to the local conditions while monitoring the mine environment, ensuring that the system can achieve real-time and comprehensive monitoring even in the case of complex mine environment and wide optical cable coverage. The introduction of self-organizing perception network enables each sensor node to work independently and cooperate with surrounding nodes, enhancing the robustness of data acquisition and transmission and solving the problems of unstable information transmission and incomplete coverage in traditional sensor systems.

[0052] Another innovation of the present application is the introduction of a miniature robot repair mechanism. Through the autonomous positioning and repair function of the miniature robot, once the monitoring system detects an abnormality in the optical cable, the robot can respond in real time and automatically go to the fault area for on-site repair. The miniature robot can effectively repair different types of optical cable damage through the built-in self-healing material or repair tool, greatly improving the efficiency and accuracy of optical cable damage repair and avoiding the time delay and cost waste caused by manual repair in traditional systems.

[0053] In terms of data storage and management, the present application uses blockchain technology to encrypt and store all sensor data, monitoring reports and repair records, ensuring the non-tamperability and integrity of the data. At the same time, through the smart contract, the repair task is automatically scheduled and executed, and a series of operations such as automatic triggering of repair tasks, cost settlement and other operations can be automatically performed, reducing the need for manual intervention and improving the transparency, efficiency and security of the system. The application of blockchain further ensures the traceability of the whole process of mine optical cable state monitoring and repair, greatly improving the reliability and management efficiency of the system.

[0054] Through the above technical innovations, the present application can solve the problems of blind area, information lag and low repair efficiency in traditional optical cable monitoring and repair systems, and realize full coverage, high efficiency, real-time and automatic optical cable monitoring and repair. These innovations not only improve the safety and stability of the mine optical cable, but also significantly reduce the cost of manual maintenance and the time of system failure repair, providing a new and intelligent management scheme for mine communication infrastructure. BRIEF DESCRIPTION OF DRAWINGS

[0055] The present application is further illustrated by the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present application. For ordinary skilled persons in the art, other drawings can be obtained without creative labor on the basis of the following drawings.

[0056] Figure 1 A framework diagram of a mine optical cable state monitoring system of the present application. DETAILED DESCRIPTION

[0057] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of the present application.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application; the use of the terms "including," "comprising," "having" and "with" in the specification and claims herein are used to mean "including but not limited to," unless otherwise noted.

[0059] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise explicitly and specifically limited.

[0060] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearance of the phrase in various places in the specification does not necessarily all refer to the same embodiment, or necessarily alternatives to other embodiments. It will be explicitly and implicitly appreciated by those of ordinary skill in the art that embodiments described herein can be combined with other embodiments.

[0061] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present application generally represents an "or" relationship between the associated objects.

[0062] In the description of the embodiments of the present application, the term "multiple" refers to two or more (including two), and similarly, "multiple groups" refers to two or more groups (including two groups), and "multiple pieces" refers to two or more pieces (including two pieces).

[0063] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0064] like Figure 1 As shown in the figure, an embodiment of the present invention provides a mining optical cable status monitoring system, the system comprising: a self-organizing sensing module 100, an optical cable health assessment module 200, a fault identification module 300, an automatic repair module 400, and a blockchain storage module 500.

[0065] The self-organizing sensing module 100 is used to acquire sensor node data of optical cables in the mine, as well as a regional gridded self-organizing sensing network constructed based on the sensor nodes, and to preprocess the sensor node data and fuse the preprocessed data based on a preset neural network model to obtain the optical cable health assessment value; wherein the sensor node data includes temperature, humidity, vibration and strain data.

[0066] In a preferred embodiment, the regional gridded self-organizing sensing network forms a distributed topology through a dynamic self-organizing protocol. Each node in the network transmits data through a low-power communication protocol. Combined with an adaptive communication frequency control mechanism, the transmission frequency can be automatically adjusted under different environmental conditions. At the same time, each node is responsible for collecting environmental data of its area and exchanging data with neighboring nodes through the self-organizing mechanism to adjust the network topology in real time.

[0067] In a preferred embodiment, the preprocessing includes: removing random noise from the data using a moving average filter or a Gaussian filter; identifying and removing outliers from the sensor data using a threshold method or a standard deviation method; smoothing the time series data using a Kalman filter; and performing local data fusion at each node to obtain the preprocessed data.

[0068] The preset neural network model structure is as follows:

[0069] Input layer: Preprocessed data from each sensor;

[0070] Hidden layer: Uses non-linear activation functions to process the complex relationships between different sensor data and learn the contribution weight of each data point to health assessment;

[0071] Output layer: output the optical cable health assessment value, which is used to represent the health status of the optical cable;

[0072] And through L2 regularization, the growth of neural network model weight is limited, and the generalization ability of neural network model is improved.

[0073] Specifically, in this scheme, the present application takes the health monitoring of mine optical cable as the core, combines with the design of regional grid self-organizing perception network, and further utilizes the neural network model to fuse data and give early warning of faults.

[0074] Firstly, a comprehensive, flexible and efficient sensor network is needed in the mine to ensure that the full length of the optical cable and its surrounding environment can be monitored. For this purpose, a regional grid self-organizing perception network is designed and deployed to meet the monitoring needs in the complex environment of the mine. This network is composed of multiple types of sensor nodes, including temperature sensors, humidity sensors, strain sensors, vibration sensors, etc.

[0075] Network structure: this network forms a distributed topology through dynamic self-organizing protocol, which can adjust the network structure in real time according to the changes of the mine environment, avoiding the single point failure problem that may occur in traditional mine sensor networks. Specifically, each node in the network transmits data through a low-power communication protocol (such as LoRa), and combines with an adaptive communication frequency control mechanism to automatically adjust the transmission frequency under different environmental conditions, reduce the burden of communication bandwidth, and ensure the stability of data transmission.

[0076] Network training and maintenance: the preliminary layout of sensor nodes should consider the spatial structure of the mine and the key monitoring points. Each node is responsible for collecting environmental data in its area, and exchanges data with adjacent nodes through self-organizing mechanism to adjust the network topology in real time.

[0077] In the mine environment, sensor nodes may be affected by physical obstacles (such as walls, cables, rock layers, etc.), and the dynamic self-organizing protocol can optimize the connectivity of the network to prevent network interruption caused by communication failure between nodes.

[0078] Preferably, after collecting data, each sensor node first performs data preprocessing. Due to the influence of noise, interference and other factors, the data needs to be denoised, outlier removed, smoothed, etc. before transmission. The purpose of preprocessing operation is to ensure the accuracy of subsequent analysis.

[0079] Specific data preprocessing methods:

[0080] Denoising: use sliding average filtering or Gaussian filtering method to remove random noise in data and improve data quality.

[0081] Outlier removal: Identify and remove outliers in sensor data through threshold or standard deviation methods.

[0082] Smoothing: Apply Kalman filter to smooth time series data, further improving data usability.

[0083] Each node performs local data fusion, that is, according to the data of other nearby nodes, it performs simple data integration. This local fusion not only improves the stability of the data, but also reduces the dependence on the central processing system, and reduces the pressure on the network bandwidth. Data preprocessing formula:

[0084] ;

[0085] Where, represents the node The original data collected (temperature, humidity, strain, vibration). is the preprocessed data. is the data preprocessing function, including filtering and denoising steps to ensure data cleanliness and accuracy.

[0086] Preferably, the processed data is periodically uploaded to the central data processing system of the mine network through an adaptive upload frequency mechanism. The upload period of each node is dynamically adjusted according to the type of data collected, the collection frequency and the real-time situation. By reducing unnecessary uploads, energy consumption is reduced and data transmission efficiency is effectively improved.

[0087] Preferably, in the health monitoring of mine optical cables, the data collected by the sensors has different physical meanings and dimensions, and direct weighted average cannot provide effective fusion results. In order to better fuse these data, the present application proposes an adaptive data fusion model based on neural network. The core of this model is to process the data of different sensors through neural network in multiple levels, and convert all the original data output by the sensors into a standardized health evaluation value. The input layer of the network accepts the preprocessed data of each type of sensor, and through training, the neural network can automatically learn the contribution weight of each data to the final health evaluation.

[0088] Network structure of neural network:

[0089] Input layer: Preprocessed data of each sensor, such as temperature, humidity, strain, etc.

[0090] Hidden layer: Use nonlinear activation function to process the complex relationship between different sensor data.

[0091] Output layer: Output a health evaluation value, which is used to represent the health status of the optical cable.

[0092] The calculation formula of the comprehensive health assessment value is:

[0093]

[0094] The preprocessed temperature, humidity, vibration and strain data are respectively represented by T, H, V and S. The weight coefficients of each input data in the neural network are represented by W. The output function of the neural network model is represented by f.

[0095] In order to avoid overfitting of the neural network in the training process, the present application introduces L2 regularization to limit the growth of model weights and improve the generalization ability of the model:

[0096]

[0097] The number of training samples is represented by N, The true health assessment value is represented by Y, The predicted health assessment value of the neural network is represented by Y. The i-th weight in the neural network is represented by W, The regularization coefficient is represented by λ.

[0098] The optical cable health assessment module 200 is used to calculate the upload frequency of its node according to the health assessment value, and to perform adaptive bandwidth allocation according to the upload frequency of its node and the total upload frequency of all nodes to obtain the bandwidth allocation of the node. At the same time, by analyzing the difference between the current health assessment value and the historical health assessment value of the node, the faulty node is identified and false positives are avoided. Once the node is marked as abnormal, the upload frequency of the node will be automatically increased based on the abnormal identification result, and the bandwidth will be re-allocated.

[0099] As a preferred embodiment, the calculation of the upload frequency of its node according to the health assessment value, and the adaptive bandwidth allocation according to the upload frequency of its node and the total upload frequency of all nodes to obtain the bandwidth allocation of the node, comprises:

[0100] According to the optical cable health assessment value, the upload frequency is calculated in the form of a Sigmoid function;

[0101] According to the upload frequency, the bandwidth allocation corresponding to the current node is obtained in combination with the total upload frequency of all nodes in the current network; ​​​​​​​​​​​

[0102] The identified faulty nodes include:

[0103] Based on the difference analysis results between the current health assessment value and the historical health assessment value of the node, and combined with the rate of change, the rate of change of the node's health assessment value is calculated through trend analysis. If the rate of change of the node's health assessment value is greater than a preset threshold, the node is considered to have an anomaly and enters an early warning state.

[0104] Specifically, in mine monitoring systems, real-time data transmission and anomaly detection are crucial for ensuring stable system operation. By constructing an intelligent sensor network and performing effective anomaly detection during data transmission, this invention enables comprehensive monitoring of the mine environment, early identification of faults or hazards, and the issuance of warnings. This module directly inherits the results of data preprocessing and local analysis (node ​​health assessment data) from the first part. The goal of this invention is to design an innovative data transmission mechanism and ensure the system's efficiency and real-time performance through intelligent anomaly detection algorithms.

[0105] To address the limitations of communication bandwidth and node energy consumption in mines, this invention employs a dynamic, adaptive data transmission mechanism based on health assessment values. This mechanism determines when and how to transmit data based on the health status of each node, thereby balancing data transmission efficiency and energy consumption. The specific design is as follows:

[0106] Preferably, the present invention is based on the health assessment value of the node. To adjust its upload frequency This is to reduce energy consumption and improve the stability of data transmission. This adjustment follows the following relationship:

[0107] ;

[0108] in, It is a node Health assessment values. It is the threshold for health assessment, when When this threshold is exceeded, the upload frequency will increase significantly. It is a coefficient that adjusts the sensitivity of the transmission frequency, controlling the impact of the health assessment value on the frequency. This is a constant coefficient that adjusts the maximum frequency. The formula, through a sigmoid function, ensures that when a node's health assessment value is low (i.e., healthy), its upload frequency remains low to reduce unnecessary energy consumption; when the node's health assessment value approaches or exceeds a threshold, the upload frequency increases to ensure real-time transmission of critical health data.

[0109] Preferably, to ensure that data transmission does not cause congestion on the mine network, this invention employs an adaptive bandwidth allocation strategy. Bandwidth Allocation is proportional to the upload frequency of node and dynamically adjusted according to the current network load of the mine:

[0110]

[0111] wherein, represents the bandwidth allocation of node . is a constant coefficient for adjusting the overall allocation of bandwidth. represents the total upload frequency of all nodes in the network. This bandwidth allocation scheme ensures that when the node upload frequency increases, the bandwidth will be automatically allocated according to the overall load of the network, thereby avoiding local network overload.

[0112] Preferably, in this step, the present application designs an intelligent anomaly detection algorithm for data transmission. Based on the health assessment value and historical data of each node, the algorithm identifies anomalies by comparing current data with historical trends and issues timely failure warnings. The specific algorithm design is as follows:

[0113] Firstly, the present application calculates the difference between the current health assessment value of node and its historical health assessment value . This step helps to capture the rapid changes of the node within a short period of time:

[0114]

[0115] Based on the difference , the present application further adopts a trend analysis method to determine whether the change of the node health assessment value is within the normal fluctuation range. The specific method is to calculate the rate of change of the node health assessment value:

[0116]

[0117] wherein, is the difference between the current and historical health assessment values. is the time interval. If the change of exceeds the preset threshold , the node is considered to have an anomaly and enters the warning state.

[0118] Preferably, in order to avoid false positives and overfitting, the present application designs an innovative regularization anomaly detection method. Based on the health assessment difference analysis, combined with the normal health fluctuation range of each node, a regularization term is introduced to control the abnormal fluctuation of the node. The regularization term ​​​​​is expressed as:

[0119] ;

[0120] wherein, is the health assessment fluctuation standard deviation of the node , indicating the normal fluctuation range of the node. is the total number of nodes. This regularization term reduces false positives caused by normal fluctuations by normalizing the health changes of each node, improving the accuracy of anomaly detection.

[0121] Preferably, after anomaly detection, if the health assessment value of the node exceeds the threshold or its change rate exceeds the expected range, the system will start the data upload mechanism and send a failure warning signal through the central system of the mine network. The specific implementation steps are as follows:

[0122] Once the node is marked as abnormal, the system will automatically increase the upload frequency of the node based on the anomaly detection result, and re-allocate bandwidth according to the network bandwidth scheduling algorithm to ensure real-time data upload to the central processing system.

[0123] Preferably, through the mine management system, the system will send warning information to relevant staff and provide real-time node health status and failure type. Staff can take necessary repair or replacement operations based on this information.

[0124] In this module, the present application designs an innovative adaptive data transmission and anomaly detection framework, which can efficiently manage data flow in mine sensor networks. Through dynamic adjustment of upload frequency and real-time anomaly detection, the entire system can maintain low energy consumption and high efficiency while ensuring system stability and real-time performance. This scheme greatly improves the fault response speed of the mine health monitoring system through precise health assessment, difference analysis and regularization anomaly detection, and reduces the risk of false positives. It is an indispensable core part of mine sensor networks.

[0125] wherein, the fault identification module 300 is used to determine the fault location according to the difference analysis result between the current health assessment value and the historical health assessment value of the node, and the change rate, obtain the regional fault probability graph, and determine the priority of the repair task based on the regional fault probability graph.

[0126] Specifically, in the mine optical cable health monitoring system, the automatic triggering of fault location and repair tasks is the key to ensuring the reliability of mine facilities. The present application designs an intelligent fault location and task scheduling system, which can quickly locate the optical cable fault location based on the abnormal detection and data analysis results (such as health evaluation value and abnormal detection output) in the previous module, and automatically trigger the corresponding repair task. This scheme not only realizes the accurate positioning of faults, but also ensures the continuity and safety of the mine. The input of this step is abnormal data (such as node health evaluation value , health change rate , and fault alarm), based on these information, the present application can locate the potential fault area and trigger the repair task. The following are the detailed steps of fault location and repair task automatic triggering:

[0127] Based on the health evaluation difference and the change rate , the present application first identifies the area most likely to have a fault by analyzing the distribution of health evaluation values in each area of the mine. The present application defines the health evaluation difference of each sensor node as:

[0128] ;

[0129] Wherein, represents the current health evaluation value of node . represents the historical health evaluation value of node .

[0130] The definition of change rate is:

[0131] ;

[0132] Wherein, is the health evaluation difference. is the measurement time interval.

[0133] In order to convert these data into a fault location model, the present application introduces a location weighting model, which converts the fault location problem into a weighted average problem of each node in the area. The location weight of node is defined as , which is related to the abnormal degree of node ( and ) and its location (such as node coordinates and ):

[0134] ;

[0135] wherein, and is the maximum value of the health assessment difference and the rate of change in all nodes, used to normalize the data. is the Euclidean distance between the node and the center of the mine failure area. is the decay coefficient, which controls the influence of distance on the node weight. is the weighting factor, which is used to adjust the proportion of the influence of the health difference and the rate of change in the calculation of the position weight.

[0136] Through the above weighting calculation, the invention obtains a weight value of each node , and the greater the node weight, the more likely the node is in the failure area.

[0137] Preferably, using the weighted position model, the weights of all nodes are integrated to obtain a regional failure probability map , which represents the probability of failure in each region of the mine. The probability map is obtained by superimposing the node weights:

[0138] ;

[0139] wherein, is the indicator function, which indicates whether the coordinates of the node are within the current region . is the number of all sensor nodes in the mine.

[0140] Through this weighted probability mapping, the invention can obtain the failure probability values of each region of the mine, so as to accurately identify the failure area. The repair task of the mine can be prioritized according to the values of the regional failure probability map. For example, if the failure probability of a region exceeds a set threshold , it is considered that the region has failed.

[0141] Preferably, once the system determines the failure area, the next step is to automatically trigger the repair task. Through the task scheduling algorithm, the system can automatically determine the most suitable repair team, repair tool and repair time according to the location of the failure area.

[0142] Firstly, the invention defines the priority of the repair task, and the calculation formula is:

[0143] ;

[0144] wherein, is the scheduling sensitivity coefficient, which controls the relationship between the regional failure probability and the task position in the task scheduling. represents the distance between the execution location of the repair task and the fault area. If exceeds the set task trigger threshold , the system automatically schedules the repair task and notifies the relevant personnel. During the task scheduling process, the system also considers the task load and resource allocation in different areas of the mine to ensure efficient execution of the repair task.

[0145] Preferably, once the repair task is executed and completed, the system will feedback the repair effect and update the health monitoring model of the mine based on the actual repaired data. For example, if the health assessment value of the relevant sensor node returns to the normal range after the repair in the fault area is completed, the change in the health status of the mine can be reflected by updating the health assessment model, ensuring more accurate subsequent fault positioning and repair tasks. After the repair is completed, the health assessment value of the node will be recalculated and updated. The invention uses a weighted average method to update the node state and adjusts the next fault detection strategy through the change of .

[0146] The automatic repair module 400 is used to enable the micro robot to automatically perform repair on the fault node according to the regional fault probability graph and the priority of the repair task, generate repair status data, and generate feedback data in combination with the sensors of the micro robot itself.

[0147] As a preferred embodiment, the enabling the micro robot to automatically perform repair on the fault node according to the regional fault probability graph and the priority of the repair task, and generating repair status data includes:

[0148] The minimum of the fault probability of the regional fault probability graph and the distance from the current position of the micro robot to the fault node is taken as the optimization objective to establish a corresponding optimization objective function;

[0149] The sensor data of the micro robot is collected to determine the accurate position of the damage and the severity of the damage area by fusing the sensor data of the robot; the severity of the damage area is confirmed based on the regional fault probability graph through an indicator function;

[0150] When the micro robot completes the repair task, the repair effect is detected through the built-in sensor, and the health status after repair is fed back to the mine management system; the health status after repair is determined based on the change rate of the health status before repair and the severity of the damage area.

[0151] ​Specifically, in the mine fiber optic cable health monitoring system, microrobots undertake critical repair tasks. Based on the fault area and repair task scheduling results, this module aims to utilize autonomously moving microrobots for automated repair of fiber optic cable damage. Through precise positioning and efficient repair tools, the microrobots can quickly repair damaged areas, ensuring the stability of the fiber optic cable and the safety of the mine environment.

[0152] Preferably, the following are the detailed steps for repairing fiber optic cable damage using a microrobot:

[0153] Path planning and remediation target confirmation:

[0154] Based on the generated fault area probability map The microrobot first needs to plan its path to reach the location of the fiber optic cable damage. Different fault areas correspond to different repair priorities. The robot prioritizes repairing areas with high levels of damage.

[0155] During path planning, the robot will determine the path based on the damaged area. The robot employs a weighted path selection method, avoiding obstacles outside the fault area while considering energy consumption and repair efficiency. To this end, this invention defines an objective function that combines fault probability and path length for optimization. The robot selects the optimal path by optimizing the objective function, avoiding unnecessary detours and efficiently reaching the damaged area.

[0156] Preferably, the microrobot carries multiple sensors to monitor the extent of damage to the optical cable in real time. These sensors detect various indicators of the optical cable, such as strain, temperature, and vibration, and determine the precise location of the damage by fusing the sensor data. Using the fused sensor data, the robot can generate a detailed damage report.

[0157] Assuming the severity of the damage area is The present invention assesses the severity of damage in the following ways:

[0158] ;

[0159] in, For indicator functions, if node If the degree of damage exceeds the threshold, then Otherwise, the value is 0. When the robot detects a severely damaged area, it will select a repair tool (such as heat welding, coating repair, etc.) for targeted repair. The specific repair tool selection is adjusted based on the degree of damage and the type of optical cable.

[0160] Preferably, once the robot reaches the repair target location, the robot will automatically perform the repair task. During the repair process, the robot uses special tools to connect broken optical cables, repair surface coatings, and other operations. After the repair is complete, the robot detects the repair effect through the built-in sensor and reports the repair result to the mine management system through the health state feedback mechanism.

[0161] Assuming that the repair task is completed, the health state of the node is represented by . The health state after repair is calculated by the following formula:

[0162] ;

[0163] wherein is the health state before repair, indicating the original state of the optical cable at the node. is the degree of damage reduced by repair. This health assessment value reflects the effectiveness of the repair, and this data will serve as the basis for the next round of maintenance and monitoring decisions.

[0164] Preferably, the robot optimizes subsequent repair tasks through real-time feedback of repair results. For example, if the repair task fails to completely restore the health state of the optical cable, the system will automatically adjust the repair strategy or dispatch other robots for subsequent repair.

[0165] The system also monitors the path planning of the robot and the efficiency of the repair task execution during the task. According to the task execution result, the system will adjust the robot route, task priority and the use of repair tools to ensure the most efficient repair process.

[0166] Through the above steps, the micro robot can efficiently perform the optical cable damage repair task and continuously optimize the repair strategy according to the feedback data, thereby realizing the automated and intelligent maintenance of the optical cable system.

[0167] The blockchain storage module 500 is used to input the repair state data and feedback data into the blockchain as a smart contract, and automatically manage the repair task and reward mechanism through the smart contract.

[0168] As a preferred embodiment, each block of the blockchain includes: stored task data, a hash value pointing to the previous block, and a timestamp of task execution;

[0169] Wherein, whenever a repair task is completed, a new block will be added to the blockchain, which contains the complete data of the task;

[0170] Wherein, the smart contract will define rules in the following way:

[0171] When the severity of the new damage area exceeds a preset threshold, an intelligent contract is automatically generated and activated, creating a new repair task;

[0172] The execution of the repair task is triggered by the robot, and a new repaired health status is generated after the task is completed. The result is submitted to the intelligent contract, triggering a repair completion event;

[0173] The intelligent contract automatically verifies the execution of the repair task. If the new repaired health status is within a certain threshold and meets the expected value, the contract execution is successful, and the repair completion is recorded;

[0174] In the intelligent contract, the task status is updated as the repair task progresses, including: task not started, task started but not completed, and task completed.

[0175] Specifically, the core purpose of this module is to ensure the transparency, security and tamper-proofing of data storage, sharing and task execution in the process of mine optical cable health monitoring and repair through blockchain technology. The intelligent contract is used to automatically manage the mine monitoring and repair process, ensuring that all repair tasks are executed according to the predetermined rules, and the relevant data cannot be tampered with.

[0176] The following is a specific scheme for implementing blockchain data storage and intelligent contract management:

[0177] Each time the health status of the optical cable changes, fault detection, repair task execution, etc. Data needs to be stored in the blockchain. These data are generated in real time when the sensors and robots perform tasks in the mine and uploaded to the blockchain to ensure their transparency and tamper-proofing. Each node's data, including fault detection, repair status and robot execution records, needs to be recorded in the blockchain as input for the intelligent contract.

[0178] Assuming that the data for each repair task is , including the health status before and after repair and , the execution robot ID, repair tool type and other information. These information will be stored in each block.

[0179] Each block in the blockchain network is composed of the following contents:

[0180] : The stored task data ( ).

[0181] : The hash value of the previous block.

[0182] : The timestamp of the task execution.

[0183] Whenever a repair task is completed, a new block will be added to the blockchain containing the complete data of the task, ensuring the security and consistency of the data.

[0184] Preferably, the smart contract automates the execution of the repair task logic, enabling transparent management and execution of the task. Each repair task needs to be defined as an entry in the smart contract, and the triggering, execution, completion, and payment of the task are all controlled by the smart contract.

[0185] Preferably, to automatically manage repair tasks, the smart contract will define rules in the following ways:

[0186] Task triggering rules: when new damage detection data exceeds a certain threshold, the system will automatically generate and activate a smart contract, creating a new repair task.

[0187] Task execution rules: the execution of the repair task is triggered by the robot, and after the task is completed, the health status of the optical cable will be updated and the results will be submitted to the smart contract, triggering the repair completion event.

[0188] Task verification rules: the smart contract automatically verifies the execution of the repair task. If it meets the expectations within a certain threshold, the contract execution is successful, and the repair completion is recorded.

[0189] In the smart contract, the task status will be updated as the repair task progresses, and the status includes:

[0190] 0: Task not started.

[0191] 1: Task started but not completed.

[0192] 2: Task completed.

[0193] The smart contract dynamically adjusts the task status and reward mechanism according to the progress of the executed task, ensuring the reliable execution of the repair task.

[0194] Preferably, once the smart contract completes the verification of the repair task and confirms the successful execution of the task, the system will record the relevant data in the blockchain, ensuring that all repair records are tamper-proof. To further encourage the enthusiasm of robots and maintenance personnel, the smart contract also contains a reward mechanism, which will automatically allocate rewards based on the execution quality of each repair task. The reward for the repair task is linked to the repair quality and task completion, and the calculation formula is as follows:

[0195] ;

[0196] where, and is a tuning coefficient, controlling the influence of the health state and the task state on the reward. is a health state of the repaired node, indicating the effect of the repair. is a task state, reflecting the completion of the task.

[0197] The reward mechanism ensures transparent execution of the task and records each reward distribution through the blockchain.

[0198] By combining blockchain technology with smart contracts, the optical cable health monitoring and repair process in the mine not only realizes safe storage and transparent sharing of data, but also efficiently manages task execution, ensuring the reliability and automation of the entire process. All data and operations are recorded in the blockchain, ensuring tamper resistance and efficient task execution.

[0199] Unless otherwise specified, the relative steps, numerical expressions, and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present application.

[0200] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the system described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0201] In the description of the present application, it should be noted that the terms "upper", "lower", etc. indicate the orientation or positional relationship shown in the drawings, or the orientation or positional relationship commonly used when the product is used, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0202] In the description of the application, it also needs to be explained that, unless otherwise explicitly specified and limited, the terms "set", "install", "connect", "connect" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the application can be understood according to the specific circumstances.

[0203] Finally, it should be pointed out that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A mining optical cable status monitoring system, characterized in that, The system includes: a self-organizing sensing module, an optical cable health assessment module, a fault identification module, an automatic repair module, and a blockchain storage module; The self-organizing sensing module is used to acquire sensor node data of optical cables in the mine, as well as a regional gridded self-organizing sensing network constructed based on the sensor nodes, and to preprocess the sensor node data and fuse the preprocessed data based on a preset neural network model to obtain the optical cable health assessment value; wherein, the sensor node data includes temperature, humidity, vibration and strain data. The optical cable health assessment module is used to calculate and adjust the upload frequency of its nodes based on the health assessment value, and to perform adaptive bandwidth allocation based on the upload frequency of its nodes and the total upload frequency of all nodes to obtain the bandwidth allocation of the nodes. At the same time, by analyzing the difference between the current health assessment value of the node and its historical health assessment value, it identifies faulty nodes and avoids false alarms. Once a node is marked as abnormal, the upload frequency of the node will be automatically increased based on the abnormal identification result, and the bandwidth will be reallocated. The fault identification module is used to determine the fault location based on the results of the difference analysis and the rate of change, obtain a regional fault probability map, and determine the priority of the repair task based on the regional fault probability map. The automatic repair module is used to enable the microrobot to automatically repair faulty nodes based on the regional fault probability map and the priority of repair tasks, generate repair status data, and generate feedback data in conjunction with the microrobot's own sensors. The blockchain storage module is used to input the repair status data and feedback data into the blockchain as a smart contract, and to automatically manage the repair tasks and reward mechanism through the smart contract. The process involves the microrobot automatically repairing faulty nodes based on a regional fault probability map and the priority of repair tasks. The generated repair status data includes: With nodes The optimization objective is to minimize the regional failure probability and the distance from the current position of the microrobot to the failure node. A corresponding optimization objective function is then established: Data from the sensors on the microrobot is collected, and the precise location and severity of the damage are determined by fusing the data. The severity of the damage is confirmed based on a regional failure probability map using an indicator function. When the microrobot completes the repair task, it detects the repair effect through built-in sensors and reports the post-repair health status to the mine management system; the post-repair health status is determined based on the rate of change between the pre-repair health status and the severity of the damaged area. The fault identification module performs the following steps: The difference analysis results and change rate are obtained, a fault location model is constructed, and the position weights of the nodes are obtained; wherein, the position weights of the nodes in the fault location model are related to the degree of anomaly of the nodes and their positions. By integrating the location weights of all nodes, a regional fault probability map is obtained. This represents the probability of a fault occurring in each area of ​​the mine; the area fault probability map The result is obtained by superimposing the node weights: ； in, It is an indicator function, representing a node. coordinates Is it in the current area? Inside; This represents the total number of sensor nodes in the mine. For nodes Position weights.

2. The mining optical cable status monitoring system according to claim 1, characterized in that, The regional gridded self-organizing sensing network forms a distributed topology through a dynamic self-organizing protocol. Each node in the network transmits data through a low-power communication protocol. Combined with an adaptive communication frequency control mechanism, it can automatically adjust the transmission frequency under different environmental conditions. At the same time, each node is responsible for collecting environmental data of its area and exchanging data with neighboring nodes through the self-organizing mechanism to adjust the network topology in real time.

3. The mining optical cable status monitoring system according to claim 1, characterized in that, The preprocessing includes: removing random noise from the data using moving average filtering or Gaussian filtering; identifying and removing outliers from the sensor data using thresholding or standard deviation methods; smoothing the time series data using a Kalman filter; and performing local data fusion at each node to obtain the preprocessed data. The preset neural network model structure is as follows: Input layer: Preprocessed data from each sensor; Hidden layer: Uses non-linear activation functions to process the complex relationships between different sensor data and learn the contribution weight of each data point to health assessment; Output layer: Outputs the optical cable health assessment value, which is used to represent the health status of the optical cable; Furthermore, L2 regularization is used to limit the growth of neural network model weights and improve the generalization ability of the neural network model.

4. The mining optical cable status monitoring system according to claim 1, characterized in that, The process of calculating and adjusting the upload frequency of a node based on the health assessment value, and performing adaptive bandwidth allocation based on the upload frequency of that node and the total upload frequency of all nodes to obtain the bandwidth allocation for the node includes: Based on the optical cable health assessment values, the upload frequency is calculated using the Sigmoid function. Based on the upload frequency and the total upload frequency of all nodes in the current network, the bandwidth allocation corresponding to the current node is obtained; The identified faulty nodes include: Based on the difference analysis results between the current health assessment value and the historical health assessment value of the node, and combined with the rate of change, the rate of change of the node's health assessment value is calculated through trend analysis. If the rate of change of the node's health assessment value is greater than a preset threshold, the node is considered to have an anomaly and enters an early warning state.

5. The mining optical cable status monitoring system according to claim 4, characterized in that, The avoidance of false alarms includes: The difference analysis introduces a regularized anomaly detection method by introducing a regularization term. Abnormal fluctuations of nodes are controlled by combining the normal health fluctuation range of each node; wherein, the regularization term , represented as: ； in, For nodes The standard deviation of the health assessment fluctuation represents the normal fluctuation range of that node; The rate of change of the node's health assessment value; This represents the total number of nodes.

6. The mining optical cable status monitoring system according to claim 1, characterized in that, Once the fault area is identified, a repair task is automatically triggered, and the priority of the repair task is automatically determined based on the location of the fault area. The calculation formula is as follows: ； in, The scheduling sensitivity coefficient controls the relationship between the probability of regional failures and the task location during task scheduling. Indicates the distance between the location where the repair task is performed and the faulty area; If the priority of the repair task If the set task trigger threshold is exceeded, a repair task will be automatically scheduled and relevant personnel will be notified.

7. The mining optical cable status monitoring system according to claim 1, characterized in that, Each block of the blockchain includes: stored task data, a hash value pointing to the previous block, and a timestamp of task execution; Whenever a repair task is completed, a new block will be added to the blockchain, containing the complete data of the task; Smart contracts will define rules in the following ways: When the severity of a new damage area exceeds a preset threshold, a smart contract is automatically generated and activated to create a new repair task. The execution of the repair task is triggered by the robot. After the task is completed, a new repaired health status will be generated and the result will be submitted to the smart contract, triggering the repair completion event. The smart contract automatically verifies the execution of the repair task; if the new, repaired health status meets expectations within a certain threshold, the contract is executed successfully and the repair is recorded as complete. In smart contracts, the task status is updated as the repair task progresses. The statuses include: task not started, task started but not completed, and task completed.

8. The mining optical cable status monitoring system according to claim 1, characterized in that, Once the smart contract verifies the repair task and confirms its successful execution, it records the relevant data in the blockchain. The smart contract also includes a reward mechanism that automatically allocates rewards based on the execution quality of each repair task. The rewards are determined based on the quality of repairs and the completion status of the task, calculated as follows: ； in, and To adjust the coefficients, the impact of health status and task status on rewards is controlled; This represents the health status of the node after repair, indicating the effectiveness of the repair. This represents the task status, reflecting the task completion progress.

Images