Mineral resource safety monitoring and early warning method and system

Abstract

The invention relates to a mineral resource safety monitoring and early warning method and system, and the method comprises the steps: obtaining monitoring data of a plurality of monitoring points, timely finding potential safety hazards, and enabling the monitoring data to at least comprise ground surface displacement, rock stratum deformation parameters, and the water level, flowing direction and flowing speed of underground water; performing normalization processing on the monitoring data to obtain normalized data; performing feature extraction and model analysis on the normalized data to judge the safety condition of the mineral resources; and early warning is realized according to mineral safety conditions. According to the method disclosed by the invention, through seamless connection with a monitoring platform, automatic analysis, early warning and alarm can be realized according to monitoring data, and a scientific and accurate decision basis is provided for a mine management department.

Description

Technical Field

[0001] The present disclosure relates to the technical field of mine management, and particularly to a method and system for safety monitoring and early warning of mineral resources. Background Art

[0002] With the continuous development and utilization of global mining resources, mine safety has become an increasingly prominent issue. During the process of mineral resource extraction, surface displacement, rock formation deformation, groundwater level changes, etc. are key monitoring parameters, which are directly related to the safety and stability of the mine. In recent years, with the increase in mining depth and the improvement of mining intensity, the risk of geological disasters has also increased, making real-time monitoring and early warning systems particularly important. Traditional mine safety monitoring methods mainly rely on manual inspections and regular measurements. This method is not only inefficient but also difficult to detect potential safety hazards in a timely manner. In addition, due to the complexity and variability of the mine environment, traditional methods often have difficulty providing accurate and comprehensive safety information.

[0003] With the progress of technology, especially the rapid development of sensor technology, data transmission technology, and big data processing technology, new possibilities have been provided for mine safety monitoring. Modern monitoring systems can real-time monitor a variety of parameters of the mine surface and underground structures, such as surface displacement, groundwater level, flow direction and velocity, rock formation deformation, etc., so as to detect abnormal situations in a timely manner and provide strong support for preventing geological disasters. However, most of the current mine safety monitoring systems on the market can only monitor a single parameter, or although they can monitor multiple parameters, their data processing and analysis capabilities are limited and they cannot provide a comprehensive safety assessment and early warning. In addition, these systems often lack seamless docking with the monitoring platform and cannot achieve automatic data analysis and early warning.

[0004] Therefore, developing a safety monitoring and early warning system for mineral resources that can real-time monitor multiple key parameters, has powerful data processing and analysis capabilities, and can be seamlessly docked with the monitoring platform is of great significance for improving mine safety and preventing geological disasters. Summary of the Invention

[0005] To solve the problems of the low efficiency of traditional mine safety monitoring methods, difficulty in timely detecting potential safety hazards, and due to the complexity and variability of the mine environment, it is difficult to provide accurate and comprehensive safety information, etc. The present disclosure proposes a method for safety monitoring and early warning of mineral resources to solve the above problems.

[0006] According to one aspect of the present disclosure, there is provided a method for safety monitoring and early warning of mineral resources, including:

[0007] S10. Obtain monitoring data of multiple monitoring points, where the monitoring data at least includes surface displacement, rock formation deformation parameters, and the water level, flow direction, and velocity of groundwater;

[0008] S20. Normalize the monitoring data to obtain normalized data;

[0009] S30. Extract features and perform model analysis on the normalized data to judge the safety situation of mineral resources;

[0010] S40. Implement early warning according to the mineral safety situation.

[0011] Preferably, obtaining the monitoring data of multiple monitoring points includes: obtaining the surface displacement data obtained by regularly observing the three-dimensional coordinate changes of surface points by a GPS receiver or total station;

[0012] Obtaining the groundwater level data obtained by real-time monitoring of the dynamic changes of the groundwater level;

[0013] Obtaining the rock formation deformation parameters obtained by measuring the minute deformation of the rock formation;

[0014] Obtaining the flow direction and flow velocity of the water flow by real-time recording and transmitting the relevant data of the water flow.

[0015] Preferably, the multiple monitoring points include surface displacement monitoring points, groundwater level monitoring points and rock formation deformation monitoring points. The surface displacement monitoring points are set on the slopes and mining faces of the mine; the groundwater level monitoring points are set in the aquifers or water level wells that may be affected by mining activities; the rock formation deformation monitoring points are set in the pre-drilled holes.

[0016] Preferably, obtaining the monitoring data of multiple monitoring points further includes: setting the data acquisition frequency, data transmission method and early warning threshold parameters. The data acquisition frequency is set according to the actual situation and safety requirements of the mine. The data transmission method adopts wired or wireless methods. The early warning threshold parameters are set according to historical data and / or expert suggestions.

[0017] Preferably, normalizing the monitoring data to obtain normalized data includes: cleaning, denoising and standardizing the monitoring data.

[0018] Preferably, extracting features and performing model analysis on the normalized data includes: using machine learning algorithms to perform pattern recognition and / or trend prediction on the normalized data / extracting key features related to mine safety from the normalized data.

[0019] Preferably, when using a machine learning algorithm to perform pattern recognition and / or trend prediction on the normalized data / extract key features related to mine safety from the normalized data, it further includes: constructing a feature analysis graph corresponding to the normalized data / the key features; using a preset convolutional neural network to extract features from the feature analysis graph to obtain corresponding convolutional analysis features; based on the normalized data / the key features and the convolutional analysis features, using a machine learning algorithm to perform pattern recognition and / or trend prediction.

[0020] Preferably, the method of using a machine learning algorithm to perform pattern recognition and / or trend prediction based on the normalized data / the key features and the convolutional analysis features includes: splicing the normalized data / the key features and the convolutional analysis features to obtain a joint feature vector; based on the joint feature vector, using a machine learning algorithm to perform pattern recognition and / or trend prediction.

[0021] Preferably, before splicing the normalized data / the key features and the convolutional analysis features to obtain a joint feature vector, using a principal component analysis algorithm to fuse the convolutional analysis features to obtain convolutional analysis fusion features with a set number of fusion features; splicing the normalized data / the key features and the convolutional analysis fusion features to obtain a joint feature vector.

[0022] Preferably, achieving early warning according to the mineral safety situation includes: when detecting an abnormal parameter situation, immediately triggering an early warning mechanism and sending an alarm message to relevant personnel, and at the same time of sending the alarm message, generating an analysis report on the abnormal parameter.

[0023] Preferably, if the result corresponding to pattern recognition using a machine learning algorithm is that there is danger now and the result corresponding to trend prediction using a machine learning algorithm is that there is danger in the future, then trigger the highest-level early warning mechanism; if the result corresponding to pattern recognition using a machine learning algorithm is that there is danger now or the result corresponding to trend prediction using a machine learning algorithm is that there is no danger in the future, then trigger the intermediate-level current early warning mechanism; if the result corresponding to pattern recognition using a machine learning algorithm is that there is no danger now or the result corresponding to trend prediction using a machine learning algorithm is that there is danger in the future, then trigger the sub-intermediate-level future early warning mechanism; if the result corresponding to pattern recognition using a machine learning algorithm is that there is no danger now and the result corresponding to trend prediction using a machine learning algorithm is that there is no danger in the future, then trigger the low-level early warning mechanism.

[0024] Preferably, if the highest-level early warning mechanism or the current intermediate-level early warning mechanism is triggered, the mineral production shall be immediately suspended; if the future intermediate-level early warning mechanism is triggered, the mineral production shall be suspended before the moment corresponding to the future danger; if the low-level early warning mechanism is triggered, the current mineral production shall be maintained.

[0025] According to one aspect of the present disclosure, there is provided a safety monitoring and early warning system for mineral resources, including:

[0026] A monitoring data acquisition module for acquiring monitoring data of multiple monitoring points, where the monitoring data at least includes surface displacement, rock formation deformation parameters, and the water level, flow direction, and flow velocity of groundwater;

[0027] A monitoring data processing module for performing normalization processing on the monitoring data to obtain normalized data;

[0028] A mineral resources safety judgment module for performing feature extraction and model analysis on the normalized data to judge the safety situation of mineral resources;

[0029] A safety early warning module for realizing early warning according to the mineral safety situation.

[0030] According to one aspect of the present disclosure, there is provided an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor is configured to: execute the above-mentioned safety monitoring and early warning method for mineral resources.

[0031] According to one aspect of the present disclosure, there is provided a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above-mentioned safety monitoring and early warning method for mineral resources is realized.

[0032] Compared with the prior art, the beneficial effects of the present disclosure are:

[0033] 1) The present disclosure comprehensively and real-time monitors various safety parameters on the surface and underground of mines through a distributed monitoring system, timely identifies and predicts possible geological disasters, greatly improves the safety monitoring ability of mines, and thus avoids or reduces the occurrence of disasters.

[0034] 2) The present disclosure provides scientific and accurate decision-making basis for mine management departments by analyzing the monitored data, and more accurately evaluates the safety status of mines by comparing real-time monitoring data with historical data, and formulates more reasonable safety management strategies.

[0035] 3) By seamlessly connecting with an advanced monitoring platform, the present disclosure automatically analyzes the monitoring data. Once data anomalies are detected, the system will immediately trigger an early warning mechanism, which not only improves the response speed but also reduces misjudgments or omissions caused by human factors, thereby improving the production efficiency of the mine and reducing the operating costs.

[0036] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit the present disclosure.

[0037] Other features and aspects of the present disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to explain the technical solutions of the present disclosure.

[0039] Figure 1 Showing the flowchart of the method for safety monitoring and early warning of mineral resources

[0040] Figure 2 Showing the block diagram of the safety monitoring and early warning system for mineral resources. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The following will detail various exemplary embodiments, features, and aspects of the present disclosure with reference to the accompanying drawings. The same reference numerals in the drawings denote elements having the same or similar functions. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0042] The special term "exemplary" herein means "serving as an example, embodiment, or illustration". Any embodiment described as "exemplary" herein is not necessarily to be construed as superior to or better than other embodiments.

[0043] The term "and / or" in this document merely describes the association relationship of associated objects and indicates that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" in this document means any one of multiple or any combination of at least two of multiple. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set composed of A, B, and C.

[0044] In addition, to better illustrate the present disclosure, numerous specific details are given in the following detailed implementation manners. Those skilled in the art should understand that the present disclosure can still be implemented without certain specific details. In some instances, methods, means, elements, and circuits well-known to those skilled in the art are not described in detail so as to highlight the gist of the present disclosure.

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Embodiment 1

[0047] Based on the above idea, the present disclosure proposes a method for safety monitoring and early warning of mineral resources. Figure 1 The flowchart showing the method for safety monitoring and early warning of mineral resources is presented. The method includes:

[0048] S10. Obtain the monitoring data of multiple monitoring points, where the monitoring data at least includes surface displacement, rock formation deformation parameters, and the water level, flow direction, and flow velocity of groundwater;

[0049] S20. Perform normalization processing on the monitoring data to obtain normalized data;

[0050] S30. Perform feature extraction and model analysis on the normalized data to judge the safety situation of mineral resources;

[0051] S40. Implement early warning according to the mineral safety situation.

[0052] The embodiments of the present disclosure provide a method for safety monitoring and early warning of mineral resources, including the following steps:

[0053] S10. Obtain the monitoring data of multiple monitoring points, where the monitoring data at least includes surface displacement, rock formation deformation parameters, and the water level, flow direction, and flow velocity of groundwater.

[0054] In this embodiment, first, appropriate monitoring points are selected. These monitoring points should cover the key areas of the mine, including but not limited to surface displacement monitoring points, groundwater level monitoring points, and rock formation deformation monitoring points, etc. The surface displacement monitoring points are set on the slopes and mining faces of the mine; the groundwater level monitoring points are set in aquifers or water level wells that may be affected by mining activities; the rock formation deformation monitoring points are set in pre-drilled holes. The selection of these monitoring points should be based on the geological conditions, historical data, and risk assessment results of the mine, and should be able to comprehensively reflect the overall deformation of the mine and the groundwater dynamics, while avoiding human interference.

[0055] Then, sensors are installed. The sensors include surface displacement sensors, groundwater level sensors, flow velocity and flow direction sensors, and rock formation deformation monitoring devices. These devices will be installed at the key monitoring points of the mine to ensure that the safety status of the mine can be comprehensively and real-timely monitored. During the installation process, the accuracy, stability, and anti-interference ability of the sensors need to be considered. All sensors need to be precisely calibrated to ensure that the collected data is accurate. In addition, to ensure the stable operation of the system, all hardware devices also need to be regularly maintained and overhauled.

[0056] For surface displacement monitoring, a high-precision GPS receiver or total station can be used for real-time monitoring. The surface displacement monitoring points are usually arranged in key areas such as near the slopes and mining faces of the mine. For the monitoring of groundwater level, flow direction, and flow velocity, a water level sensor and a flow velocity sensor can be installed. The groundwater level monitoring points are arranged in aquifers that may be affected by mining activities according to hydrogeological conditions, and the dynamic changes of the groundwater level are real-timely monitored through a water level gauge installed in the water level well. In addition, to monitor the flow direction and flow velocity, flow velocity and flow direction meters are installed on key water flow channels, and these instruments can real-timely record and transmit the relevant data of the water flow. For the monitoring of rock formation deformation, sensors such as strain gauges or inclinometers can be used. These devices are installed in pre-drilled holes and can accurately measure the small deformations of the rock formation. All sensors should be connected to the data collector to ensure the real-time transmission of data.

[0057] Among them, the monitoring data of multiple monitoring points include: surface displacement data obtained by obtaining the three-dimensional coordinate changes of surface points regularly observed by the GPS receiver or total station; groundwater level data obtained by obtaining the dynamic changes of the real-time monitored groundwater level; rock formation deformation parameters obtained by measuring the small deformations of the rock formation; the flow direction and flow velocity of the water flow obtained by obtaining the relevant data of the real-time recorded and transmitted water flow.

[0058] Finally, after the installation of the sensors is completed, initialization is carried out, including setting the data acquisition frequency, data transmission method, and early warning threshold parameters. The data acquisition frequency should be set according to the actual situation and safety requirements of the mine. Usually, data can be acquired once per second, once per minute, or once per hour. The data transmission method can be wired or wireless to ensure the real-time and accuracy of the data. The early warning threshold parameters should be set according to historical data and / or expert advice to trigger an early warning in a timely manner when the parameters are abnormal.

[0059] During the initialization process, it is also necessary to establish a connection with the monitoring platform, connect the data collector to the monitoring platform through the network, and realize the real-time upload and analysis of data. Each sensor transmits parameters such as surface displacement, groundwater level, flow direction, flow velocity, and rock formation deformation monitored in real time to the data center. The monitoring platform should have functions of automatic analysis, early warning, and alarm to notify the mine management department in a timely manner when abnormal parameters are found.

[0060] In addition, to ensure the stability and reliability of abnormal monitoring, regular maintenance and inspection are also required. This includes checking the operating status of the sensors, the stability of data transmission, and the effectiveness of the early warning system, etc. Through regular maintenance and inspection, potential problems can be discovered and solved in a timely manner to ensure the normal operation of safety monitoring and early warning.

[0061] In this embodiment, a medium-sized mine is assumed as the application scenario. Corresponding sensors are deployed in each key area of the mine and initialized. Through real-time monitoring and data analysis, abnormal situations of parameters such as surface displacement, groundwater level, flow direction and velocity, and rock formation deformation can be discovered in a timely manner, providing a scientific and accurate decision-making basis for the mine management department.

[0062] It should be noted that in the specific implementation process, flexible adjustment and optimization should be carried out according to the actual situation and safety requirements of the mine. For example, in areas with complex geological conditions, the number and types of sensors can be increased to improve the monitoring accuracy and reliability; in terms of data transmission, more advanced communication technologies can be adopted to improve the data transmission speed and stability. In order to ensure the real-time and accuracy of the data, a highly reliable data transmission scheme is adopted. Inside the mine, high-speed data transmission between the sensors and the data center is achieved by laying optical cables or setting up wireless communication base stations. At the same time, to cope with possible communication failures, the system is also equipped with a data caching and retransmission mechanism to ensure the integrity of the data.

[0063] S20. Perform normalization processing on the monitoring data to obtain normalized data.

[0064] In this embodiment, automation and remote control technologies are adopted to transmit the monitoring data to the central monitoring center in real time via wireless transmission. To ensure the integrity and reliability of the monitoring data, data redundancy design and error detection mechanisms are employed. Meanwhile, it is also necessary to have the function of adaptively adjusting the data acquisition frequency so as to collect data more intensively in case of anomalies and provide richer information for the early warning system.

[0065] During the process of collecting monitoring data, special attention is also paid to the calibration and verification of the monitoring data. It is regularly compared with the manually measured data to ensure the accuracy of automatic acquisition. In addition, all the collected monitoring data will be normalized, and steps such as cleaning, denoising and standardization are carried out on the monitoring data to improve the quality of the monitoring data.

[0066] Through this series of deployment and collection work, not only can key parameters such as accurate surface displacement, groundwater level, flow direction and velocity, and rock stratum deformation be provided, but also strong support can be provided for subsequent data analysis, safety assessment and early warning.

[0067] S30. Extract features and perform model analysis on the normalized data to judge the safety situation of mineral resources.

[0068] In the embodiment of the present disclosure, machine learning algorithms are used to perform pattern recognition and / or trend prediction on the normalized data / extract key features related to mine safety from the normalized data.

[0069] In the embodiment of the present disclosure, the use of machine learning algorithms to perform pattern recognition and / or trend prediction on the normalized data / extract key features related to mine safety from the normalized data further includes: constructing a feature analysis diagram corresponding to the normalized data / the key features; using a preset convolutional neural network to extract features from the feature analysis diagram to obtain corresponding convolutional analysis features; based on the normalized data / the key features and the convolutional analysis features, using machine learning algorithms to perform pattern recognition and / or trend prediction.

[0070] In the embodiment of the present disclosure, the method of using machine learning algorithms to perform pattern recognition and / or trend prediction based on the normalized data / the key features and the convolutional analysis features includes: splicing the normalized data / the key features and the convolutional analysis features to obtain a joint feature vector; based on the joint feature vector, using machine learning algorithms to perform pattern recognition and / or trend prediction.

[0071] In the embodiments of the present disclosure, before splicing the normalized data / the key features and the convolutional analysis features to obtain a joint feature feature vector, the principal component analysis algorithm is used to fuse the convolutional analysis features to obtain convolutional analysis fused features with a set number of fused features; the normalized data / the key features and the convolutional analysis fused features are spliced to obtain a joint feature feature vector.

[0072] In this embodiment, the normalized data obtained after normalizing the collected monitoring data, key features related to mine safety are extracted from the normalized data and sent to the data analysis module for in-depth analysis. One or several advanced machine learning algorithms, such as support vector machine (SVM), random forest, decision tree, K-nearest neighbor, logistic regression, linear discriminant analysis, adaptive boosting, and multi-layer perceptron, are used to perform pattern recognition and / or trend prediction on parameters such as ground surface displacement, groundwater level, and rock formation deformation.

[0073] In the embodiments of the present disclosure and other possible embodiments, the ground surface displacement, the water level, flow direction, flow velocity of groundwater, and the rock formation deformation parameters are respectively normalized to obtain the normalized data corresponding to the ground surface displacement, the water level, flow direction, flow velocity of the groundwater, and the rock formation deformation.

[0074] In the embodiments of the present disclosure and other possible embodiments, if the monitoring data corresponding to the ground surface displacement, the water level, flow direction, flow velocity of groundwater, and the rock formation deformation parameters are the monitoring data at the first moment corresponding to multiple monitoring points at the same set mineral location, then based on the monitoring data at the first moment and the trend label corresponding to the second moment after the first moment, a trend prediction model for monitoring and early warning is constructed using a machine learning algorithm, and then the trend of the monitoring and early warning is predicted using the trend prediction model. Among them, the trend of the monitoring and early warning or the trend label can be configured as there is danger in the future (corresponding numerical value is configured as 0) and there is no danger in the future (corresponding numerical value is configured as 1); among them, the existence of danger in the future can also be configured as low danger in the future (corresponding numerical value is configured as 2), medium danger in the future (corresponding numerical value is configured as 3), and high danger in the future (corresponding numerical value is configured as 4).

[0075] In the embodiments of the present disclosure and other possible embodiments, the normalized data corresponding to the set number of the surface displacements, the water levels corresponding to the groundwater, the flow directions, the flow velocities, and the rock formation deformations are divided into a data set according to a set ratio (such as 7:3 or 8:2), to obtain a training set of monitoring data with the first number A (A×M, where M is the preset number of monitoring data, such as the preset number of monitoring data is configured to be 5 or other values corresponding to the surface displacement, the water level of the groundwater, the flow direction, the flow velocity, and the rock formation deformation parameters), and a test set of monitoring data corresponding to the second number B greater than the first number A (B×M, where M is the preset number of monitoring data); a key feature related to mine safety is selected from the training set of monitoring data by using a preset feature selection algorithm, to obtain key features corresponding to the third number N less than the preset number of the monitoring data (for example, the surface displacement, the water level of the groundwater, the flow velocity, and the rock formation deformation parameters, and at this time N = 4); the machine learning algorithm is trained based on the key features to obtain a trend prediction model; the corresponding third number of key features (B×N, and the key features are such as the surface displacement, the water level of the groundwater, the flow velocity, and the rock formation deformation parameters) are selected from the test set of monitoring data corresponding to the second number; the performance index of the trend prediction model is evaluated by using the corresponding third number of key features of the test set of monitoring data.

[0076] In the embodiments of the present disclosure and other possible embodiments, if the monitoring data corresponding to the surface displacement, the water level of the groundwater, the flow direction, the flow velocity, and the rock formation deformation parameters are the monitoring data corresponding to multiple monitoring points at multiple different set mineral locations, then based on the monitoring data and their corresponding pattern labels, a pattern recognition model for monitoring and early warning is constructed by using a machine learning algorithm, and then the pattern of the monitoring and early warning is recognized by using the pattern recognition model. Among them, the pattern of the monitoring and early warning or the pattern label can be configured to be currently dangerous (the corresponding value is configured to be 0) and currently not dangerous (the corresponding value is configured to be 1); among them, the currently dangerous situation can also be configured to be currently low in danger (the corresponding value is configured to be 2), currently medium in danger (the corresponding value is configured to be 3), and currently high in danger (the corresponding value is configured to be 4).

[0077] Similarly, in the embodiments of the present disclosure and other possible embodiments, the normalized data of the set number of the surface displacements, the water levels corresponding to the groundwater, the flow directions, the flow velocities, and the rock formation deformations are divided into data sets according to a set ratio (such as 7:3 or 8:2), to obtain a training set of monitoring data with the first number A (A×M, where M is the preset number of monitoring data, such as the preset number of monitoring data is configured to be 5 or other values corresponding to the surface displacement, the water level of the groundwater, the flow direction, the flow velocity, and the rock formation deformation parameters); a preset feature selection algorithm is used to select key features related to mine safety from the training set of monitoring data, to obtain key features corresponding to the third number N that is less than the preset number of the monitoring data (for example, the surface displacement, the water level of the groundwater, the flow velocity, and the rock formation deformation parameters, and at this time N = 4); the machine learning algorithm is trained based on the key features to obtain a pattern recognition model; the corresponding third number of key features (B×N, and the key features such as the surface displacement, the water level of the groundwater, the flow velocity, and the rock formation deformation parameters) are selected from the monitoring data test set corresponding to the second number; the performance indicators of the pattern recognition model are evaluated by using the corresponding third number of key features of the monitoring data test set.

[0078] In the embodiments of the present disclosure and other possible embodiments, the preset feature selection algorithm can be configured as one or several of feature selection algorithms such as the Recursive Feature Elimination algorithm, the Relief algorithm, the Featureweighting algorithms, the mutual information filtering feature selection algorithm, and the XgBoost feature selection algorithm.

[0079] In the embodiments of the present disclosure and other possible embodiments, the performance indicators may include one or several of: Accuracy, Precision, Recall, F1-score, and AUC. By comparing with a large model trained from historical data of mine accident monitoring, the abnormal conditions of the parameters are automatically detected, wherein the large model is statistically trained from historical data. Once an abnormality is detected, an early warning mechanism will be immediately triggered, and alarm information will be sent to mine management personnel through various methods such as text messages, emails, and audible and visual alarms.

[0080] In the embodiments of the present disclosure and other possible embodiments, the method for safety monitoring and early warning of mineral resources further includes: after respectively normalizing the surface displacement, the water level of groundwater, the flow direction, the flow velocity, and the rock formation deformation parameters to obtain the normalized data corresponding to the surface displacement, the water level of the corresponding groundwater, the flow direction, the flow velocity, and the rock formation deformation, constructing a feature analysis diagram corresponding to the normalized data corresponding to the surface displacement, the water level of the corresponding groundwater, the flow direction, the flow velocity, and the rock formation deformation; using a preset convolutional neural network to extract features from the feature analysis diagram to obtain corresponding convolutional analysis features (non-linear convolutional analysis features); based on the normalized data corresponding to the surface displacement, the water level of the corresponding groundwater, the flow direction, the flow velocity, and the rock formation deformation and the convolutional analysis features, using a machine learning algorithm to construct a pattern recognition model and / or a trend prediction model.

[0081] In the embodiments of the present disclosure and other possible embodiments, the preset convolutional neural network includes: a first convolutional unit and a plurality of second convolutional units connected in sequence; wherein, the input of the first convolutional unit is configured as the feature analysis diagram, the output of the first convolutional unit is connected to the first second convolutional unit among the plurality of second convolutional units, and the last second convolutional unit among the plurality of second convolutional units outputs convolutional analysis features.

[0082] Further, in the embodiments of the present disclosure and other possible embodiments, the first convolutional unit and the plurality of second convolutional units at least include: a convolutional layer; wherein, the convolutional layer can be configured as a convolutional kernel of a set size. At the same time, the first convolutional unit and the plurality of second convolutional units further include: a pooling layer connected to the convolutional layer; and / or, a batch normalization layer connected to the pooling layer. In addition, the first convolutional unit and the plurality of second convolutional units further include: a non-linear processing layer connected to the pooling layer or the batch normalization layer connected to the pooling layer. Wherein, the non-linear processing layer is configured with a non-linear function.

[0083] In the embodiments of the present disclosure and other possible embodiments, the method for constructing a feature analysis diagram corresponding to the normalized data corresponding to the surface displacement, the water level of the corresponding groundwater, the flow direction, the flow velocity, and the rock formation deformation includes: using the first number corresponding to the normalized data as the row or column, and using the numerical values corresponding to the surface displacement, the water level of the corresponding groundwater, the flow direction, the flow velocity, and the rock formation deformation of the normalized data as the column or row to construct a feature analysis diagram.

[0084] In the embodiments of the present disclosure and other possible embodiments, a method for constructing a pattern recognition model and / or a trend prediction model by using a machine learning algorithm based on the normalized data and convolution analysis features corresponding to the ground surface displacement, the water level, flow direction, flow velocity of the groundwater, and the rock formation deformation includes: splicing the normalized data corresponding to the ground surface displacement, the water level, flow direction, flow velocity of the groundwater, and the rock formation deformation and the convolution analysis features to obtain a joint feature vector; and constructing a pattern recognition model and / or a trend prediction model by using a machine learning algorithm based on the joint feature vector.

[0085] In addition, in the embodiments of the present disclosure and other possible embodiments, before splicing the normalized data corresponding to the ground surface displacement, the water level, flow direction, flow velocity of the groundwater, and the rock formation deformation and the convolution analysis features to obtain a joint feature vector, the convolution analysis features are fused by using a principal component analysis algorithm to obtain convolution analysis fusion features with a set number of fusion features; the normalized data corresponding to the ground surface displacement, the water level, flow direction, flow velocity of the groundwater, and the rock formation deformation and the convolution analysis fusion features are spliced to obtain a joint feature vector; and a pattern recognition model and / or a trend prediction model is constructed by using a machine learning algorithm based on the joint feature vector.

[0086] Similarly, in the embodiments of the present disclosure and other possible embodiments, the normalized data corresponding to the set number of the surface displacements, the water levels corresponding to the groundwater, the flow directions, the flow velocities, and the rock formation deformations are divided into data sets according to a set ratio (such as 7:3 or 8:2), to obtain a training set of monitoring data with a first number A (A×M, where M is the preset number of monitoring data, such as the preset number of monitoring data is configured as 5 or other values corresponding to the surface displacement, the water level of the groundwater, the flow direction, the flow velocity, and the rock formation deformation parameters) and a test set of monitoring data corresponding to a second number B greater than the first number A (B×M, where M is the preset number of monitoring data); a preset feature selection algorithm is used to select key features related to mine safety from the training set of monitoring data, to obtain key features corresponding to a third number N less than the preset number of the monitoring data (for example, the surface displacement, the water level of the groundwater, the flow velocity, and the rock formation deformation parameters, and at this time N = 4); a feature analysis diagram corresponding to the key features is constructed; a preset convolutional neural network is used to extract features from the feature analysis diagram, to obtain corresponding convolutional analysis features; based on the key features and the convolutional analysis features, a pattern recognition model and / or a trend prediction model is constructed by using a machine learning algorithm; wherein, the constructing the pattern recognition model and / or the trend prediction model by using the machine learning algorithm based on the key features and the convolutional analysis features includes: training the machine learning algorithm based on the key features and the convolutional analysis features, to obtain the pattern recognition model and / or the trend prediction model. Furthermore, key features and convolutional analysis features are selected from the test set of monitoring data corresponding to the second number; the performance indicators of the pattern recognition model are evaluated by using the key features and the convolutional analysis features corresponding to the test set of monitoring data.

[0087] In the embodiments of the present disclosure and other possible embodiments, the method for training the machine learning algorithm based on the key features and the convolutional analysis features to obtain the pattern recognition model and / or the trend prediction model includes: splicing the key features and the convolutional analysis features to obtain a joint feature vector; based on the joint feature vector, constructing the pattern recognition model and / or the trend prediction model by using the machine learning algorithm.

[0088] In addition, in the embodiments of the present disclosure and other possible embodiments, before splicing the key features and the convolutional analysis features to obtain the joint feature vector, a principal component analysis algorithm is used to fuse the convolutional analysis features to obtain convolutional analysis fused features with a set number of fused features; the key features and the convolutional analysis fused features are spliced to obtain the joint feature vector; based on the joint feature vector, the pattern recognition model and / or the trend prediction model is constructed by using the machine learning algorithm.

[0089] Further, in the embodiments of the present disclosure and other possible embodiments, the method of constructing a feature analysis diagram corresponding to the key feature and extracting features from the feature analysis diagram using a preset convolutional neural network includes: constructing a feature analysis diagram with the first number corresponding to the key feature as the row or column and the value corresponding to the key feature as the column or row.

[0090] Wherein, in the embodiments of the present disclosure and other possible embodiments, a machine learning algorithm is used for pattern recognition and / or trend prediction to obtain a first probability value corresponding to the current existence of danger and / or a second probability value corresponding to the future existence of danger; based on the first probability value and the first warning threshold, it is determined whether there is danger currently and whether there is no danger currently; based on the second probability value and the second warning threshold, it is determined whether there is danger in the future and whether there is no danger in the future.

[0091] Further, in the embodiments of the present disclosure and other possible embodiments, if the first probability value is greater than or equal to the first warning threshold, it is determined that there is danger currently; otherwise, it is determined that there is no danger currently. Similarly, if the second probability value is greater than or equal to the second warning threshold, it is determined that there is danger in the future; otherwise, it is determined that there is no danger in the future.

[0092] Wherein, in the embodiments of the present disclosure and other possible embodiments, the first warning threshold and the second warning threshold can be set or configured according to historical data and / or expert suggestions.

[0093] S40. Implement early warning according to the mineral safety situation.

[0094] In this embodiment, the designed early warning mechanism fully considers the actual needs of mine management. When abnormal parameter conditions are detected, the early warning mechanism is immediately triggered, and an alarm message is sent to relevant personnel. At the same time as the alarm is issued, a detailed analysis report is also generated, including specific values of abnormal parameters, time and location of the abnormality, possible causes, and recommended countermeasures and other information. This information helps mine management personnel to quickly respond and take effective measures to prevent potential safety accidents.

[0095] In the embodiments of the present disclosure, if the result of pattern recognition using a machine learning algorithm indicates a current danger and the result of trend prediction using a machine learning algorithm indicates a future danger, the highest-level warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm indicates a current danger or the result of trend prediction using a machine learning algorithm indicates no future danger, the intermediate-level current warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm indicates no current danger or the result of trend prediction using a machine learning algorithm indicates a future danger, the sub-intermediate-level future warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm indicates no current danger and the result of trend prediction using a machine learning algorithm indicates no future danger, the low-level warning mechanism is triggered.

[0096] In the embodiments of the present disclosure, if the highest-level warning mechanism or the intermediate-level current warning mechanism is triggered, the production of the mineral is immediately stopped; if the intermediate-level future warning mechanism is triggered, the production of the mineral is stopped before the time corresponding to the future danger; if the low-level warning mechanism is triggered, the current production of the mineral is maintained. In addition, the safety monitoring and warning method for mineral resources also has the ability of self-learning and optimization. By continuously accumulating and analyzing new monitoring data, its warning accuracy and timeliness are gradually improved. This characteristic of continuous learning and optimization enables the safety monitoring and warning method for mineral resources to adapt to various complex mine environments and geological conditions, providing a strong guarantee for the safe production of mines.

[0097] In actual operation, the management personnel can adjust the warning threshold according to the actual situation and feedback new data and warning situations. These data will be used to continuously optimize and improve the monitoring model and warning mechanism.

[0098] In addition, by regularly conducting a comprehensive performance evaluation and safety inspection of the system, its long-term stable operation is ensured and the requirements for mine safety monitoring are met.

[0099] Through the above specific implementation process, it can be seen that the safety monitoring and warning method for mineral resources in the embodiments of the present disclosure not only has a comprehensive monitoring function, but also has efficient data processing capabilities and a flexible warning mechanism. This will provide strong technical support for the safe production of mines.

[0100] With the continuous development and utilization of global mining resources, mine safety issues have become increasingly prominent, and the demand for safety monitoring and warning of mineral resources has become increasingly urgent. The safety monitoring and warning method for mineral resources proposed in this embodiment has broad application prospects. It can not only play an important role in mining production, but also is expected to promote the development of related industries and technological progress.

[0101] First, in the field of mining production, the application of this embodiment will greatly improve the safety and stability of mine production. By real-time monitoring of key parameters such as surface displacement, groundwater level, flow direction, flow velocity, and rock formation deformation, the mine management department can timely detect potential safety hazards and take corresponding preventive measures. This can not only reduce the occurrence of mine accidents, ensure the safety of miners' lives, but also avoid the huge economic losses and social impacts brought by mine accidents.

[0102] Secondly, the application of this embodiment also helps to promote the intelligent and automated process of mining production. Through seamless connection with the monitoring platform, the monitoring data can be automatically analyzed, pre-warned, and alarmed, providing scientific and accurate decision-making basis for the mine management department. This will greatly improve the efficiency and accuracy of mine management, reduce the safety risks caused by human factors, and at the same time provide strong support for the sustainable development of the mine.

[0103] In the long run, the application of this embodiment is also expected to promote the green transformation and sustainable development of the mineral resources development industry. Through precise monitoring and early warning, mineral resources can be more scientifically planned and utilized, reducing unnecessary mining and waste, thereby reducing the impact on the environment.

[0104] In summary, the mineral resources safety monitoring and early warning method in this embodiment has broad application prospects and huge market potential. It can not only ensure the safety and stability of mine production, but also promote the development of related industries and technological progress, providing strong support for the sustainable development and utilization of mining resources. With the continuous progress of technology and the increasing expansion of the market, it is believed that the application of this embodiment will be more and more extensive, bringing more well-being and benefits to mining production and social development.

[0105] Embodiment 2

[0106] As another aspect of the embodiments of the present disclosure, a mineral resources safety monitoring and early warning system 100 is further provided, as Figure 2 shown, including:

[0107] A monitoring data acquisition module 1, which acquires monitoring data of multiple monitoring points, and the monitoring data at least includes surface displacement, rock formation deformation parameters, and groundwater level, flow direction, and flow velocity;

[0108] A monitoring data processing module 2, which performs normalization processing on the monitoring data to obtain normalized data;

[0109] A mineral resources safety judgment module 3, which performs feature extraction and model analysis on the normalized data to judge the safety situation of mineral resources;

[0110] A safety early warning module 4, which realizes early warning according to the mineral safety situation.

[0111] Without conflict, the above modules in the system of the embodiments of the present disclosure can implement any of the above-described implementation manners of the method.

[0112] Based on the description of the above embodiments, the embodiments of the present disclosure can achieve the following technical effects:

[0113] 1) The present disclosure comprehensively and real-time monitors various safety parameters on the surface and underground of the mine through a distributed monitoring system, timely identifies and predicts possible geological disasters, greatly improves the safety monitoring ability of the mine, and thus avoids or reduces the occurrence of disasters.

[0114] 2) The present disclosure provides scientific and accurate decision-making basis for the mine management department by analyzing the monitored data, and more accurately evaluates the safety status of the mine by comparing the real-time monitoring data with historical data, and formulates a more reasonable safety management strategy.

[0115] 3) The present disclosure seamlessly docks with an advanced monitoring platform to automatically analyze the monitored data. Once data anomalies are found, the system will immediately trigger an early warning mechanism, which not only improves the response speed but also reduces misjudgment or missed judgment caused by human factors, thus improving the production efficiency of the mine and reducing the operating cost.

[0116] The embodiments of the present disclosure also propose an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor is configured to perform the above-described method for safety monitoring and early warning of mineral resources. The electronic device can be provided as a terminal, a server or other forms of devices.

[0117] The embodiments of the present disclosure also propose a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above-described method for safety monitoring and early warning of mineral resources is implemented. The computer-readable storage medium can be a non-volatile computer-readable storage medium.

[0118] Those skilled in the art can understand that in the above-described method and system for safety monitoring and early warning of mineral resources in the specific implementation manner, the writing order of each step does not mean a strict execution order and does not constitute any limitation to the implementation process. The specific execution order of each step should be determined according to its function and possible internal logic.

[0119] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks may in fact be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0120] The embodiments of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, the practical application, or the technical improvement of the technology in the market, or to enable other ordinary skill in the art in the technical field to understand the embodiments disclosed herein.

Claims

1. A mineral resource safety monitoring and early warning method, characterized in that: The steps include: S10, acquiring monitoring data of multiple monitoring points, wherein the monitoring data includes at least one or more of surface displacement, rock formation deformation parameters, and groundwater level, flow direction, and flow velocity; S20, normalizing the monitoring data to obtain normalized data; S30, performing feature extraction and model analysis on the normalized data to determine the safety of mineral resources; S40. Provide early warning based on the mine safety situation.

2. The method according to claim 1, characterized in that: Acquiring monitoring data of multiple monitoring points includes: acquiring surface displacement data obtained by recording three-dimensional coordinate changes of surface points regularly observed by a GPS receiver or a total station; Acquire groundwater level data obtained by real-time monitoring of dynamic changes in groundwater levels; Obtaining rock formation deformation parameters obtained by measuring the deformation of the rock formation; Obtain real-time recording and transmission of water flow related data to obtain the flow direction and velocity of the water flow.

3. The method according to any one of claims 1 to 2, characterized in that: The multiple monitoring points include surface displacement monitoring points, groundwater level monitoring points and rock deformation monitoring points. The surface displacement monitoring points are set on the slopes and mining faces of the mine; the groundwater level monitoring points are set in aquifers or water level wells that may be affected by mining activities; and the rock deformation monitoring points are set in pre-drilled holes.

4. The method according to any one of claims 1 to 3, characterized in that: Obtaining monitoring data from multiple monitoring points also includes: setting data collection frequency, data transmission mode and warning threshold parameters. The data collection frequency is set according to the actual situation and safety requirements of the mine, the data transmission mode adopts wired or wireless mode, and the warning threshold parameters are set according to historical data and / or expert advice.

5. The method according to any one of claims 1 to 4, characterized in that: Normalizing the monitoring data to obtain normalized data includes: cleaning, denoising and standardizing the monitoring data.

6. The method according to claim 5, characterized in that The feature extraction and model analysis of the normalized data includes: using a machine learning algorithm to perform pattern recognition and / or trend prediction on the normalized data / extract key features related to mine safety from the normalized data; and / or, The method of using a machine learning algorithm to perform pattern recognition and / or trend prediction on the normalized data / extracting key features related to mine safety from the normalized data also includes: constructing a feature analysis graph corresponding to the normalized data / the key features; using a preset convolutional neural network to perform feature extraction on the feature analysis graph to obtain corresponding convolution analysis features; based on the normalized data / the key features and the convolution analysis features, using a machine learning algorithm to perform pattern recognition and / or trend prediction; and / or, The method for pattern recognition and / or trend prediction based on the normalized data / the key features and the convolution analysis features using a machine learning algorithm comprises: concatenating the normalized data / the key features and the convolution analysis features to obtain a joint feature vector; based on the joint feature vector, using a machine learning algorithm to perform pattern recognition and / or trend prediction; and / or, Before the normalized data / the key features are spliced ​​with the convolution analysis features to obtain the joint feature vector, the convolution analysis features are fused using a principal component analysis algorithm to obtain convolution analysis fusion features with a set number of fusion features; the normalized data / the key features are spliced ​​with the convolution analysis fusion features to obtain the joint feature vector.

7. The method according to any one of claims 1 to 6, characterized in that: Implementing early warning according to the safety situation of the mine includes: when abnormal parameters are detected, immediately triggering the early warning mechanism and sending alarm information to relevant personnel, while sending the alarm information, generating an analysis report on the abnormal parameters; and / or, If the result of pattern recognition using a machine learning algorithm is that there is a danger now and the result of trend prediction using a machine learning algorithm is that there will be a danger in the future, the highest level of warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm is that there is a danger now or the result of trend prediction using a machine learning algorithm is that there will be no danger in the future, the medium level of current warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm is that there is no danger now or the result of trend prediction using a machine learning algorithm is that there will be a danger in the future, the next medium level of future warning mechanism is triggered; if the result of pattern recognition using a machine learning algorithm is that there is no danger now and the result of trend prediction using a machine learning algorithm is that there will be no danger in the future, the low level of triggering warning mechanism is triggered; and / or, If the highest level warning mechanism or the medium level current warning mechanism is triggered, the mine production will be shut down immediately; if the medium level future warning mechanism is triggered, the mine production will be shut down before the corresponding moment in the future when the danger exists; if the low level triggering warning mechanism is triggered, the current mine production will be maintained.

8. A mineral resources safety monitoring and early warning system, characterized in that: include: A monitoring data acquisition module, which acquires monitoring data of multiple monitoring points, wherein the monitoring data includes at least one or more of surface displacement, rock formation deformation parameters, and groundwater level, flow direction, and flow velocity; A monitoring data processing module, performing normalization processing on the monitoring data to obtain normalized data; A mineral resource safety judgment module performs feature extraction and model analysis on the normalized data to judge the safety of mineral resources; The safety early warning module can provide early warning based on the mine safety situation.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the mineral resource safety monitoring and early warning method described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the mineral resource safety monitoring and early warning method described in any one of claims 1 to 7 is implemented.

Images