Downhole operation monitoring method and device and readable storage medium

Abstract

The invention discloses a downhole operation monitoring method and device and a readable storage medium. The method comprises the steps that a monitoring data set of a downhole operation environment is obtained, the downhole operation environment is used for representing the environment where downhole operation is located, and the monitoring data set comprises data monitored by the downhole operation environment under multiple monitoring dimensions; based on the monitoring data set, a target monitoring point model is constructed, and the target monitoring point model is used for indicating the incidence relation between different monitoring points in the downhole operation and the data transmission rule between the different monitoring points; based on the target monitoring point model, a target monitoring strategy is determined, and the target monitoring strategy is used for indicating a rule for monitoring underground operation; and monitoring the underground operation according to the target monitoring strategy. The technical problem that the safety monitoring efficiency of the underground operation environment is low is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic fences, and in particular to a monitoring method, device, and readable storage medium for underground operations. Background Art

[0002] At present, in the field of coal mining, especially when facing steeply inclined and thick coal seams or complex geological conditions, dynamic disasters such as rock burst and rock burst are important threats to mine safety. These dynamic disasters are often caused by stress concentration and energy accumulation in coal and rock masses, posing a serious threat to the life safety of miners and mine facilities. Therefore, how to effectively relieve pressure to reduce the stress concentration of coal and rock masses has become an urgent problem to be solved in mining.

[0003] In the relevant technology, the safety monitoring system of the underground working environment mainly relies on traditional physical fences, sensors and control systems to monitor the working area. The traditional monitoring system usually adopts a single-layer monitoring method. The physical layer is responsible for monitoring the switch status of the fence, the sensor layer is responsible for collecting environmental parameters, and the control layer triggers alarms or executes relevant control instructions through preset conditions. Although this type of monitoring method can realize real-time monitoring of the working environment to a certain extent, with the complexity of the underground working environment, the traditional single-layer monitoring system has been difficult to meet the actual needs. The existing monitoring system has obvious limitations in processing multi-level and multi-source data. The data of the physical layer, sensor layer and control layer often run independently and act independently. It is impossible to effectively integrate the related information between different levels, resulting in bottlenecks in multi-dimensional data processing and comprehensive analysis, making it difficult for the system to respond to potential dangers or abnormal situations in a timely and accurate manner, and there are certain safety hazards. Therefore, there is a technical problem of low efficiency in safety monitoring of the underground working environment.

[0004] With respect to the above-mentioned technical problem of low efficiency in safety monitoring of underground working environment, no effective solution has been proposed so far. Summary of the invention

[0005] The embodiments of the present invention provide a method, a device, and a readable storage medium for monitoring underground operations, so as to at least solve the technical problem of low efficiency in safety monitoring of underground operation environments.

[0006] According to one aspect of an embodiment of the present invention, a method for monitoring downhole operations is provided. The method may include: obtaining a monitoring data set of a downhole operation environment, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment in multiple monitoring dimensions; based on the monitoring data set, constructing a target monitoring point model, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; based on the target monitoring point model, determining a target monitoring strategy, wherein the target monitoring strategy is used to indicate the rules for monitoring the downhole operation; and monitoring the downhole operation according to the target monitoring strategy.

[0007] Optionally, the monitoring method for underground operations also includes: obtaining multiple monitoring data of the underground operating environment, wherein the monitoring data includes at least one of the following: data of each monitoring point or multi-layer data, and the multi-layer data includes at least: control layer data, sensor layer data and physical layer data; obtaining a monitoring data set, including: integrating multiple monitoring data to obtain a monitoring data set.

[0008] Optionally, based on the monitoring data set, a target monitoring point model is constructed, including: extracting multiple monitoring point data and multi-layer data from the monitoring data set; determining each monitoring point data as a model node, and determining the multi-layer data as a model edge; based on the model nodes and model edges, constructing an initial monitoring point model; adjusting the initial monitoring point model to obtain a target monitoring point model.

[0009] Optionally, the initial monitoring point model is adjusted to obtain a target monitoring point model, including: extracting attribute features from model nodes and model edges, wherein the attribute features are used to indicate features related to the attributes of the model nodes and model edges; generating multi-dimensional data based on the attribute features; and adjusting the initial monitoring point model based on the multi-dimensional data to obtain a target monitoring point model.

[0010] Optionally, based on the multi-dimensional data, the initial monitoring point model is adjusted to obtain the target monitoring point model, including: based on the multi-dimensional data, determining the weights between each model node, wherein the weights are used to indicate the importance of each model node; based on the weights, the initial monitoring point model is adjusted to obtain the target monitoring point model.

[0011] Optionally, the method for monitoring downhole operations further includes: identifying abnormal data from the multi-dimensional data; and outputting the abnormal data in response to the existence of the abnormal data.

[0012] According to another aspect of an embodiment of the present invention, a monitoring device for underground operations is also provided. The device may include: an acquisition unit, used to acquire a monitoring data set of an underground operation environment, wherein the underground operation environment is used to represent the environment in which the underground operation is located, and the monitoring data set includes data monitored for the underground operation environment in multiple monitoring dimensions; a construction unit, used to construct a target monitoring point model based on the monitoring data set, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the underground operation, and the data transmission rules between different monitoring points; a determination unit, used to determine a target monitoring strategy based on the target monitoring point model, wherein the target monitoring strategy is used to indicate the rules for monitoring the underground operation; and a monitoring unit, used to monitor the underground operation according to the target monitoring strategy.

[0013] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the method for monitoring downhole operations in an embodiment of the present invention.

[0014] According to another aspect of an embodiment of the present invention, a processor is provided, which is used to run a program, wherein when the program is run, the method for monitoring downhole operations in an embodiment of the present invention is executed.

[0015] According to another aspect of the embodiment of the present invention, a computer program product is also provided, which includes computer instructions, and when the computer instructions are executed by a processor, the method for monitoring downhole operations in the embodiment of the present invention is implemented.

[0016] In an embodiment of the present invention, a monitoring data set of a downhole operation environment is obtained, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment under multiple monitoring dimensions; based on the monitoring data set, a target monitoring point model is constructed, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; based on the target monitoring point model, a target monitoring strategy is determined, wherein the target monitoring strategy is used to indicate the rules for monitoring the downhole operation; and the downhole operation is monitored according to the target monitoring strategy. That is to say, the embodiment of the present invention processes the downhole operation environment data set and constructs a target monitoring point model, thereby monitoring the downhole environment according to the target monitoring strategy determined by the target monitoring point model, thereby achieving the purpose of monitoring the safety of the downhole operation environment, thereby solving the technical problem of low efficiency in monitoring the safety of the downhole operation environment, and achieving the technical effect of improving the efficiency of monitoring the safety of the downhole operation environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 is a flow chart of a method for monitoring downhole operations according to an embodiment of the present invention;

[0019] Figure 2 is a flow chart of a multi-level electronic fence monitoring method based on an underground working environment according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of a multi-level electronic fence monitoring system based on an underground operation environment according to an embodiment of the present invention;

[0021] Figure 4 The present invention is a flowchart of a method and device for monitoring underground operations according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, functional component or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, functional components or devices.

[0024] According to an embodiment of the present invention, an embodiment of a method for monitoring downhole operations is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] Figure 1 is a flow chart of a method for monitoring downhole operations according to an embodiment of the present invention, such as Figure 1 As shown, the method may include the following steps:

[0026] Step S101, obtaining a monitoring data set of the underground operating environment.

[0027] In the technical solution provided in step S101 of the present invention, the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored on the downhole operation environment in multiple monitoring dimensions. The monitoring data set can also be called a multi-level monitoring data set.

[0028] In this embodiment, a monitoring data set of the underground operating environment is obtained. For example, the monitoring data set of the underground operating environment is obtained through a monitoring and acquisition system. This is only an illustrative example and does not limit the specific method of obtaining the monitoring data set of the underground operating environment.

[0029] For example, a multi-level monitoring data acquisition system is constructed in an underground operation environment and multi-level monitoring data sets are collected.

[0030] Step S102: construct a target monitoring point model based on the monitoring data set.

[0031] In the technical solution provided in the above step S102 of the present invention, the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation and the data transmission rules between different monitoring points. The target monitoring point model can also be called a hypergraph neural network model.

[0032] In this embodiment, after the monitoring data set of the underground operating environment is acquired in step S101, a target monitoring point model is constructed according to the monitoring data set.

[0033] Optionally, a neural network model is constructed for the monitoring data set, and the hypergraph neural network model is trained using multi-dimensional data to obtain a target monitoring point model.

[0034] For example, the multi-level monitoring data sets from multiple monitoring points are transmitted through a local area network, and the local area network covers multiple monitoring points in the underground operation area; a hypergraph neural network model is constructed for the multi-level monitoring data sets collected from multiple monitoring points, and each monitoring point and each level of data are used as nodes in the hypergraph, and the correlation between the physical layer, sensor layer and control layer data is used as a hyperedge; features are extracted from the nodes and hyperedges in the hypergraph neural network model to generate a multi-dimensional data representation, and the data relationship between each monitoring point and the level is modeled based on the hypergraph neural network model, the relationship weights between each node are calculated, and the hypergraph neural network model is optimized through iterative training; the electronic fence system is optimized using the multi-dimensional data representation, and the hypergraph neural network model is further trained using a small amount of labeled data and a large amount of unlabeled data under a weakly supervised learning framework to obtain an optimized hypergraph neural network model, that is, the target monitoring point model.

[0035] Step S103: determining a target monitoring strategy based on the target monitoring point model.

[0036] In the technical solution provided in the above step S103 of the present invention, the target monitoring strategy is used to indicate the rules for monitoring underground operations. The monitoring strategy may also be called an electronic fence monitoring strategy or an electronic fence monitoring strategy.

[0037] In this embodiment, after the target monitoring point model is constructed in step S102, the target monitoring strategy is determined according to the target monitoring point model.

[0038] Optionally, the target monitoring point model is used to optimize the initial monitoring strategy to obtain a target monitoring strategy.

[0039] For example, the optimized hypergraph neural network model can be used to dynamically adjust the electronic fence monitoring strategy in the underground working environment, identify potential safety hazards based on the multi-dimensional data representation of the physical layer, sensor layer and control layer, and provide real-time warning for abnormal data.

[0040] Step S104: monitor the underground operation according to the target monitoring strategy.

[0041] In the technical solution provided in the above step S104 of the present invention, after the target monitoring strategy is determined in step S103, the downhole operation is monitored according to the target monitoring strategy.

[0042] In this embodiment, the downhole operation is monitored by linking the data of multiple monitoring points.

[0043] For example, by utilizing optimized monitoring strategies and coordinating data linkage of multiple monitoring points through multi-dimensional data representation, safety protection of electronic fences for underground operations can be achieved.

[0044] It should be noted that the above embodiments can be implemented through a multi-level electronic fence monitoring system based on the underground working environment.

[0045] In the above steps S101 to S104 of the present invention, a monitoring data set of the downhole operation environment is obtained, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment under multiple monitoring dimensions; based on the monitoring data set, a target monitoring point model is constructed, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; based on the target monitoring point model, a target monitoring strategy is determined, wherein the target monitoring strategy is used to indicate the rules for monitoring the downhole operation; and the downhole operation is monitored according to the target monitoring strategy. That is to say, the embodiment of the present invention processes the downhole operation environment data set and constructs the target monitoring point model, thereby monitoring the downhole environment according to the target monitoring strategy determined by the target monitoring point model, thereby achieving the purpose of monitoring the safety of the downhole operation environment, thereby solving the technical problem of low efficiency of safety monitoring of the downhole operation environment, and achieving the technical effect of improving the efficiency of safety monitoring of the downhole operation environment.

[0046] The above method of this embodiment is further introduced below.

[0047] As an optional implementation method, the monitoring method for underground operations also includes: obtaining multiple monitoring data of the underground operating environment, wherein the monitoring data includes at least one of the following: data of each monitoring point or multi-layer data, and the multi-layer data includes at least: control layer data, sensor layer data and physical layer data; obtaining a monitoring data set, including: integrating multiple monitoring data to obtain a monitoring data set.

[0048] In this embodiment, multiple monitoring data of the underground operating environment are obtained, and the multiple monitoring data are integrated to obtain a monitoring data set. For example, different monitoring data are collected by different sensors, and the multiple monitoring data are combined to obtain a monitoring data set.

[0049] For example, a multi-level monitoring data acquisition system is constructed in an underground operation environment. Sensors deployed in the multi-level monitoring data acquisition system are used to collect multi-level monitoring data of the physical layer, sensor layer, and control layer to construct a multi-level monitoring data set D total , including the following parts: Fence status data D of the physical layer phy , represents the physical status information of the underground fence, which is collected by the physical fence sensors installed around the underground working environment. The status information S collected by the physical fence sensors phy Contains the opening, closing and damage information of the electronic fence; real-time environmental data of the sensor layer D sens, the sensor network in the underground environment collects environmental parameters in real time; the feedback command data D of the control layer ctrl , through the command feedback information of the control equipment in the underground operation environment, the control layer feedback data is generated by the multi-level monitoring data acquisition system according to the abnormal information fed back by the sensor, including emergency braking and alarm activation status. The data at each level is combined into a multi-level monitoring data set, as shown in the following formula (1):

[0050] D total ={D phy ,D sens ,D ctrl} (1)

[0051] As an optional implementation method, a target monitoring point model is constructed based on a monitoring data set, including: extracting multiple monitoring point data and multi-layer data from the monitoring data set; determining each monitoring point data as a model node, and determining the multi-layer data as a model edge; constructing an initial monitoring point model based on the model nodes and model edges; and adjusting the initial monitoring point model to obtain a target monitoring point model.

[0052] In this embodiment, a plurality of monitoring point data and multi-layer data are extracted from the monitoring data set; each monitoring point data is determined as a model node, thereby obtaining a node set.

[0053] For example, for a multi-level monitoring dataset D collected from multiple monitoring points total A hypergraph neural network model H is constructed, and each monitoring point and its associated data at each level are taken as nodes in the hypergraph. The node set is shown in the following formula (2):

[0054] V={v1,v2,…,v n} (2)

[0055] Among them, each node v i Corresponding to a monitoring point or a certain level of data, node v i It can be expressed by the following formula (3):

[0056]

[0057] in, It can be used to represent the i-th data of the physical layer, sensor layer and control layer respectively. Represent the weight parameters of the physical layer, sensor layer and control layer data, α i is the environmental dynamic adjustment factor for each monitoring point i.

[0058] Optionally, after determining the node, multiple layers of data are determined as model edges, for example, the correlation between the fence status data of the physical layer, the real-time environment data of the sensor layer, and the feedback instruction data of the control layer is used as the hyperedge E.

[0059] For example, the initial hypergraph H0 in combination with the downhole operation environment is defined as the following formula (4):

[0060] H0=(V,E,W) (4)

[0061] Where W is the weight matrix, the initial value is set according to the preliminary association between the monitoring point and the data at each level, and E represents the edge of the hypergraph; the association between the fence status data of the physical layer, the real-time environment data of the sensor layer, and the feedback instruction data of the control layer is represented as the hyperedge E, and the hyperedge set is recorded as E = {e1, e2, …, e m}, each hyperedge e j Connecting a monitoring point and its data at all levels, the association strength between hyperedges is expressed as the following formula (5) based on the distance weight:

[0062]

[0063] Among them, D i is the data of different levels of monitoring points, α i It is the adjustment parameter of the hierarchical association.

[0064] Optionally, an initial monitoring point model is constructed based on the model nodes and model edges, and the initial monitoring point model is adjusted to obtain a target monitoring point model. The initial monitoring point model may also be referred to as an initial feature matrix.

[0065] For example, according to the multi-level dataset D of each monitoring point total Generate the initial feature matrix X = {x1, x2, ..., x n}, x i For node v i The feature vector of .

[0066] As an optional implementation method, the initial monitoring point model is adjusted to obtain a target monitoring point model, including: extracting attribute features from model nodes and model edges, wherein the attribute features are used to indicate features related to the attributes of the model nodes and model edges; generating multi-dimensional data based on the attribute features; and adjusting the initial monitoring point model based on the multi-dimensional data to obtain a target monitoring point model.

[0067] In this embodiment, attribute features are extracted from model nodes and model edges, thereby generating multi-dimensional data, that is, a feature matrix, based on the attribute features. For example, attribute features are extracted from multi-level monitoring data to form a node feature matrix.

[0068] For example, feature extraction is performed on the node set V and the hyperedge set E in the hypergraph neural network model, and the feature extraction of each node v i The eigenvector x i , the initial feature vector is composed of the corresponding multi-level monitoring data to form a node feature matrix X. Based on the hypergraph structure, the feature mapping function The node feature vector x i Projected into d-dimensional space, generating a multi-dimensional feature representation h of the node i It is the following formula (6):

[0069]

[0070] Among them, W (0) is the initial weight matrix, σ is the nonlinear activation function, The node v is at the 0th iteration. i The feature representation of .

[0071] Optionally, the initial monitoring point model is optimized and adjusted according to the multi-dimensional data to obtain a target monitoring point model. The specific optimization method is described as follows.

[0072] As an optional implementation method, the initial monitoring point model is adjusted based on multi-dimensional data to obtain a target monitoring point model, including: determining the weights between each model node based on the multi-dimensional data, wherein the weights are used to indicate the importance of each model node; based on the weights, the initial monitoring point model is adjusted to obtain a target monitoring point model.

[0073] In this embodiment, the weights between the model nodes are determined based on the multi-dimensional data. For example, the data relationship between the nodes is modeled using the association relationship in the hyperedge E, and the relationship weight matrix between the nodes is calculated.

[0074] For example, the data relationship between nodes is modeled using the association relationship in the hyperedge E, and the relationship weight matrix A between each node is calculated. ij}, relationship weight a ij The calculation of is shown in the following formula (7):

[0075]

[0076] Among them, E ij To include node v i and v j The hyperedge set of , w(e) is the weight of hyperedge e, Represents node v i and node v j Cosine similarity between feature vectors, k is the number of iterations.

[0077] Optionally, the initial monitoring point model is optimized and adjusted according to the weights to obtain the target monitoring point model. and relationship weight a ij , until the convergence condition is met, and the optimized node feature matrix is ​​obtained Therefore, the optimized node feature matrix is ​​used to optimize and adjust the initial monitoring point model to obtain the target monitoring point model.

[0078] For example, the node features are updated based on the relationship weight matrix A, and the node features are iteratively optimized using a message passing mechanism, as shown in the following formula (8):

[0079]

[0080] in, is the node v in the kth iteration i The eigenvector of (k) is the weight matrix of the kth iteration, σ is the nonlinear activation function; repeat the iteration to update the node feature vector and relationship weight a ij , until the convergence condition is met, and the optimized node feature matrix is ​​obtained The node feature matrix is ​​the generated multi-dimensional data representation; using the optimized node feature matrix H * Update the hyperedge weight w(e); optimize the hypergraph neural network model H through iterative training, and finally generate a matrix H containing node features * and optimize the hyperedge weight W * The hypergraph neural network model of , that is, the target monitoring point model, is shown in the following formula (9):

[0081]

[0082] Among them, H (t+1) represents the optimized hypergraph neural network model in the t+1th iteration, is for node v i and v j The characteristic function for modeling the relationship between j ) represents the hyperedge weight; the first term represents the node feature update aggregated by the hyperedge weight, α1 is the learning rate of the aggregate update; the second term is the gradient update term based on the objective function L, γ is the learning rate, y i For node v i The target value of , α1 is the control parameter for gradient update.

[0083] As an optional implementation manner, the method for monitoring underground coal mine operations further includes: identifying abnormal data from the multi-dimensional data; and outputting the abnormal data in response to the existence of the abnormal data.

[0084] In this embodiment, abnormal data is identified from multi-dimensional data; when abnormal data exists, it indicates that the underground environment is safe abnormal, based on which, the abnormal data is output.

[0085] Optionally, potential safety hazards are identified based on the multi-dimensional data representation of the physical layer, sensor layer and control layer, and real-time warnings are issued for abnormal data, thereby achieving the purpose of monitoring the underground environment.

[0086] It should be noted that the above embodiments can be implemented through a multi-level electronic fence monitoring system based on the underground working environment.

[0087] In this embodiment, a monitoring data set of the downhole operation environment is obtained, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment under multiple monitoring dimensions; based on the monitoring data set, a target monitoring point model is constructed, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; based on the target monitoring point model, a target monitoring strategy is determined, wherein the target monitoring strategy is used to indicate the rules for monitoring the downhole operation; and the downhole operation is monitored according to the target monitoring strategy. That is to say, the embodiment of the present invention processes the downhole operation environment data set and constructs a target monitoring point model, thereby monitoring the downhole environment according to the target monitoring strategy determined by the target monitoring point model, thereby achieving the purpose of monitoring the safety of the downhole operation environment, thereby solving the technical problem of low efficiency of safety monitoring of the downhole operation environment, and achieving the technical effect of improving the efficiency of safety monitoring of the downhole operation environment.

[0088] The technical solution of the embodiment of the present invention is illustrated below in conjunction with preferred implementation modes.

[0089] At present, in the field of coal mining, especially when facing steeply inclined and thick coal seams or complex geological conditions, dynamic disasters such as rock burst and rock burst are important threats to mine safety. These dynamic disasters are often caused by stress concentration and energy accumulation in coal and rock masses, posing a serious threat to the life safety of miners and mine facilities. Therefore, how to effectively relieve pressure to reduce the stress concentration of coal and rock masses has become an urgent problem to be solved in mining.

[0090] In the related art, the safety monitoring system of the underground working environment mainly relies on traditional physical fences, sensors and control systems to monitor the working area. The traditional monitoring system usually adopts a single-layer monitoring method. The physical layer is responsible for monitoring the switch status of the fence, the sensor layer is responsible for collecting environmental parameters, and the control layer triggers alarms or executes relevant control instructions through preset conditions. Although this type of monitoring method can realize real-time monitoring of the working environment to a certain extent, with the complexity of the underground working environment, the traditional single-layer monitoring system has been difficult to meet the actual needs. The existing monitoring system has obvious limitations in processing multi-level and multi-source data. The data of the physical layer, sensor layer and control layer often run independently and act independently. It is impossible to effectively integrate the related information between different levels, resulting in bottlenecks in multi-dimensional data processing and comprehensive analysis, making it difficult for the system to respond to potential dangers or abnormal situations in a timely and accurate manner, and there are certain safety hazards. Therefore, there is a technical problem of low efficiency in the safety monitoring of the underground working environment. In view of the above-mentioned technical problem of low efficiency in the safety monitoring of the underground working environment, no effective solution has been proposed.

[0091] However, an embodiment of the present invention proposes a multi-level electronic fence monitoring method based on an underground operating environment, in which a multi-level monitoring data acquisition system is constructed in the underground operating environment and a multi-level monitoring data set is collected; the multi-level monitoring data sets in multiple monitoring points are transmitted through a local area network, and the local area network covers multiple monitoring points in the underground operating area; a hypergraph neural network model is constructed for the multi-level monitoring data sets collected from multiple monitoring points, and the data relationship between each monitoring point and the layer is modeled based on the hypergraph neural network model, the relationship weight between each node is calculated, and the hypergraph neural network model is optimized through iterative training; based on the optimized hypergraph neural network model, the electronic fence monitoring strategy in the underground operating environment is dynamically adjusted, and potential safety hazards are identified according to the multi-dimensional data representation of the physical layer, the sensor layer and the control layer, and real-time warning is given for abnormal data; using the optimized monitoring strategy, the data linkage of multiple monitoring points is coordinated through the multi-dimensional data representation, so as to realize the safety protection of the underground operation electronic fence, solve the technical problem of low efficiency of safety monitoring of the underground operating environment, and achieve the technical effect of improving the efficiency of safety monitoring of the underground operating environment.

[0092] The embodiments of the present invention are further described below.

[0093] Figure 2 : is a flow chart of a multi-level electronic fence monitoring method based on an underground working environment according to an embodiment of the present invention, the multi-level electronic fence monitoring method comprising the following steps:

[0094] Step S201, collecting multi-level monitoring data sets.

[0095] In this embodiment, a multi-level monitoring data acquisition system is constructed in an underground operation environment and a multi-level monitoring data set is acquired.

[0096] Optionally, a multi-level monitoring data acquisition system is constructed in the underground operation environment, and sensors deployed in the multi-level monitoring data acquisition system are used to collect multi-level monitoring data of the physical layer, the sensing layer and the control layer to construct a multi-level monitoring data set D total , including the fence status data D of the physical layer phy , represents the physical status information of the underground fence, which is collected by the physical fence sensors installed around the underground working environment. The status information S collected by the physical fence sensors phy Contains the opening, closing and damage information of the electronic fence. Real-time environmental data of the sensor layer D sens The sensor network in the underground environment collects environmental parameters and feedback command data from the control layer in real time. ctrl ,Through the command feedback information of the control equipment in the ,downhole operation environment, the control layer feedback data is ,generated by the multi-level monitoring data acquisition system according to the ,abnormal information fed back by the sensors, including the emergency braking and ,alarm activation status.

[0097] Optionally, the data at each level is combined into a multi-level monitoring data set as shown in the above formula (1), which will not be described in detail here.

[0098] Step S202, transmitting the multi-level monitoring data sets in the multiple monitoring points through the local area network.

[0099] In this embodiment, the multi-level monitoring data sets in the multiple monitoring points are transmitted through the local area network, and the local area network covers the multiple monitoring points in the underground operation area.

[0100] Step S203, constructing a hypergraph neural network model for the multi-level monitoring data set collected from multiple monitoring points.

[0101] In this embodiment, a hypergraph neural network model is constructed for a multi-level monitoring data set collected from multiple monitoring points, with each monitoring point and each level of data as a node in the hypergraph, and the correlation between the physical layer, sensor layer and control layer data as a hyperedge.

[0102] Optionally, for a multi-level monitoring data set D collected from multiple monitoring points total The hypergraph neural network model H is constructed, and each monitoring point and its associated data at each level are taken as nodes in the hypergraph. The node set is shown in the above formula (2), which will not be repeated here. i The data corresponding to a monitoring point or a certain level is shown in the above formula (3), which will not be elaborated here.

[0103] Optionally, the initial hypergraph H0 in combination with the downhole operating environment is defined as the aforementioned formula (4), which will not be described in detail here.

[0104] Optionally, the correlation between the fence status data of the physical layer, the real-time environment data of the sensor layer, and the feedback instruction data of the control layer is represented as a hyperedge E, and the hyperedge set is recorded as E = {e1, e2, ..., e m}, each hyperedge e j Connecting a monitoring point and its data at each level, the association strength between hyperedges is expressed as the aforementioned formula (5) according to the distance weighting, which will not be repeated here.

[0105] Optionally, based on the multi-level dataset D of each monitoring point total Generate the initial feature matrix X = {x1, x2, ..., x n}, x i For node v i The feature vector of .

[0106] Step S204, optimizing the hypergraph neural network model.

[0107] In this embodiment, features are extracted from nodes and hyperedges in the hypergraph neural network model to generate a multi-dimensional data representation, the data relationships between monitoring points and levels are modeled based on the hypergraph neural network model, the relationship weights between nodes are calculated, and the hypergraph neural network model is optimized through iterative training.

[0108] Optionally, feature extraction is performed on the node set V and the hyperedge set E in the hypergraph neural network model to extract the feature vector of each node v. i The eigenvector x i , the initial feature vector is composed of the corresponding multi-level monitoring data to form a node feature matrix X; based on the hypergraph structure, using the feature mapping function The node feature vector x i Projected into d-dimensional space, generating a multi-dimensional feature representation h of the node i As mentioned above in formula (6), it will not be repeated here.

[0109] Optionally, the data relationship between nodes is modeled using the association relationship in the hyperedge E, and the relationship weight matrix A between each node is calculated. ij}, relationship weight a ij The calculation of is the aforementioned formula (7), which will not be repeated here.

[0110] Optionally, the node features are updated based on the relationship weight matrix A, and the node features are iteratively optimized using a message passing mechanism as shown in the aforementioned formula (8), which will not be described in detail here.

[0111] Optionally, repeatedly iterate and update the node feature vector and relationship weight a ij , until the convergence condition is met, and the optimized node feature matrix is ​​obtained The node feature matrix is ​​the generated multi-dimensional data representation; using the optimized node feature matrix H * Update the hyperedge weight w(e); optimize the hypergraph neural network model H through iterative training, and finally generate a matrix H containing node features * and optimize the hyperedge weight W * The hypergraph neural network model is as shown in the above formula (9), which will not be repeated here.

[0112] Step S205, optimizing the electronic fence system using multi-dimensional data representation.

[0113] In this embodiment, the electronic fence system is optimized using multi-dimensional data representation, and a small amount of labeled data and a large amount of unlabeled data are used to further train the hypergraph neural network model under a weakly supervised learning framework.

[0114] Optionally, based on the optimized hypergraph neural network model H (t+1) , the electronic fence system is further optimized by using multi-dimensional data representation, and a small number of labeled data sets D are constructed. label and a large number of unlabeled datasets D unlabel .

[0115] Optionally, define the loss function L in a weakly supervised learning framework, including the supervised learning loss term L label The comparative learning loss term L with unlabeled data contrast As shown in the following formula (10):

[0116] L=λ1·L label +λ2·L contrast (10)

[0117] Among them, λ1 and λ2 are adjustment coefficients, which represent the weights of supervised learning and contrastive learning loss terms; supervised learning loss L label Through the labeled data D label The target value y i and the model prediction value The calculation is shown in the following formula (11):

[0118]

[0119] Among them, y i For node v i The true label value of is the value predicted by the hypergraph neural network model; contrastive learning loss L contrast Optimize the unlabeled dataset D by calculating the similarity between unlabeled dataunlabel As shown in the following formula (12):

[0120]

[0121] Among them, h i and h j For node v i and v j The eigenvector of cosθ(h i ,h j ) is the cosine similarity between two nodes, and δ is the threshold parameter for contrastive learning. Based on the loss function L, the hypergraph neural network model H is optimized by back propagation. (t+1) , update the node feature vector h i and the hyperedge weight w(e j ); Repeat the iterative training until the hypergraph neural network model converges, and finally obtain the further optimized hypergraph neural network model H final .

[0122] Step S206, dynamically adjusting the electronic fence monitoring strategy in the underground operation environment based on the optimized hypergraph neural network model.

[0123] In this embodiment, based on the optimized hypergraph neural network model, the electronic fence monitoring strategy in the underground working environment is dynamically adjusted, potential safety hazards are identified based on the multi-dimensional data representation of the physical layer, sensor layer and control layer, and real-time warning is issued for abnormal data.

[0124] Optionally, based on the hypergraph neural network model H final Dynamically adjust the electronic fence monitoring strategy in the underground operation environment, and use multi-dimensional data to represent H * Real-time analysis of data from the physical layer, sensor layer, and control layer; multi-dimensional data representation of H * Identify potential safety hazards in the underground working environment and calculate the risk score R (v i ) is shown in the following formula (13):

[0125]

[0126] Among them, w(e j ) represents the monitoring point v i and node v j The hyperedge weight between them, Risk(h j ) is the node v j The corresponding risk value of multi-dimensional data, R(v i ) is the node v i The risk score of each node is R(v i ) and the preset safety threshold θr For comparison, when R(v i )>θ r When the node has potential safety hazards, it is judged and an abnormal warning signal is issued; the electronic fence monitoring strategy is adjusted in real time, and the updated hypergraph neural network model H is used final Optimize monitoring strategies, dynamically adjust monitoring levels in high-risk areas, increase sensor data collection frequency, and issue relevant safety instructions. ctrl Improve the protection level of the area.

[0127] Optionally, by utilizing the optimized monitoring strategy, data linkage of multiple monitoring points is coordinated through multi-dimensional data representation to achieve safety protection of the electronic fence for underground operations.

[0128] Optionally, the electronic fence monitoring strategy is used to coordinate the electronic fence monitoring data of each monitoring point in the underground operation area based on the hypergraph neural network model, so that the electronic fence monitoring system can operate stably in the local area network. Through the linkage relationship between nodes, the linkage matrix L between the electronic fence monitoring points is constructed. ij}, l ij Indicates the electronic fence monitoring point v i and monitoring point v j The data linkage strength between them is shown in the following formula (14):

[0129]

[0130] Among them, h i and h j For node v i and node v j The multi-dimensional feature vector represents the linkage strength between the electronic fence monitoring points. The larger the value, the stronger the data correlation. The monitoring strategy of each electronic fence monitoring point is dynamically adjusted according to the linkage matrix L, and the monitoring points with larger linkage strength are given priority for data sharing and processing, so as to realize multi-point linkage monitoring of the electronic fence in the underground operation area.

[0131] Figure 3 FIG. 1 is a schematic diagram of a multi-level electronic fence monitoring system based on an underground operation environment according to an embodiment of the present invention. Figure 3 As shown, the multi-level electronic fence monitoring system 300 may include: a multi-level monitoring data acquisition module 301, a data transmission module 302, a hypergraph neural network modeling module 303, a weak supervision learning module 304, a monitoring strategy dynamic adjustment module 305 and a data linkage coordination module 306.

[0132] The multi-level monitoring data acquisition module 301 is used to collect multi-level monitoring data of the physical layer, the sensor layer and the control layer in the underground operation environment to generate a multi-level monitoring data set.

[0133] The data transmission module 302 transmits the multi-level monitoring data sets collected by each monitoring point to the central processing unit through the local area network for centralized analysis and processing.

[0134] The hypergraph neural network modeling module 303 is used to construct a hypergraph neural network model, taking the monitoring points and their associated data at various levels as hypergraph nodes, and generating hyperedges based on the correlation between the physical layer, the sensor layer and the control layer to form a hypergraph structure for multi-dimensional data modeling and feature extraction.

[0135] The weakly supervised learning module 304 optimizes and trains the hypergraph neural network model using a small amount of labeled data and a large amount of unlabeled data.

[0136] The monitoring strategy dynamic adjustment module 305 dynamically adjusts the electronic fence monitoring strategy in the underground operation environment based on the optimized hypergraph neural network model, identifies potential safety hazards and issues real-time warnings for abnormal data.

[0137] The data linkage coordination module 306 uses the optimized monitoring strategy to construct a linkage matrix between the electronic fence monitoring points, coordinate the data linkage between multiple monitoring points, and realize all-round safety protection in the underground operation area.

[0138] The feasibility of the multi-level electronic fence monitoring system of the present invention in an underground working environment is further illustrated below through specific implementation examples.

[0139] In order to verify the feasibility of the multi-level electronic fence monitoring system of the present invention in the underground operation environment, the experimenters conducted an experimental design and used two publicly available mine safety datasets: MineSafety2022 and UndergroundMonitoring2019. The MineSafety2022 dataset contains multi-level monitoring data of underground operations collected from January 2020 to March 2022, including physical fence status, environmental parameters (temperature, humidity and gas concentration) and control layer feedback in the operating area. The UndergroundMonitoring2019 dataset contains accident records and related monitoring data in mine operations over the years, covering from January 2019 to December 2021. These two datasets have a wide range of application scenarios in the field of underground operation safety monitoring, and the data is relatively comprehensive.

[0140] The experimenter divided the two data sets into training set and test set in a ratio of 8:2. In order to evaluate the performance of the system of the present invention in multi-level monitoring data processing and anomaly detection, the monitoring data was first preprocessed. For the fence status data of the physical layer, the experimenter used a simple representation based on Boolean values, and the open and closed states of the fence were represented as 1 and 0 respectively. For the environmental data of the sensor layer, including temperature, humidity, carbon monoxide, and methane gas concentration, the original values ​​were used for recording. The feedback data of the control layer mainly includes abnormal alarms and emergency braking operation instructions. All data were standardized before entering the system.

[0141] The system of the present invention is based on a hypergraph neural network model. The experimenter uses the multi-level data in MineSafety2022 to construct a hypergraph structure, in which the data of each monitoring point is represented as a node in the hypergraph, and the correlation between the physical layer, the sensor layer, and the control layer is modeled through hyperedges. In order to further optimize the system, the experimenter uses a small amount of labeled accident data as the labeled data set, and combines a large amount of normal operating data as unlabeled data, and trains the model through a weakly supervised learning framework. The experimenter uses the Adam optimizer to train the model, with the learning rate set to 0.001, the number of training rounds to 200, and the mini-batch set to 128.

[0142] In order to verify the anomaly detection capability of the system of the present invention, the experimenter selected actual abnormal event records from the UndergroundMonitoring2019 dataset and used these event data to test the detection effect of the system. The experimenter used accuracy, precision, recall and F1 score as evaluation indicators to comprehensively measure the performance of the system when processing complex multi-level monitoring data.

[0143] Table 1 Schematic diagram of comparison results

[0144] method Accuracy (%) Accuracy (%) Recall rate (%) F1 score (%) Traditional monitoring system 85.4 82.3 78.6 80.4 System of the present invention 93.6 91.2 89.7 90.4

[0145] Optionally, Table 1 is a comparison result between the multi-level electronic fence monitoring system of the present invention and the traditional method on the MineSafety2022 dataset. During the experiment, the data in Table 1 show that on the MineSafety2022 dataset, the system of the present invention is superior to the traditional monitoring system in all indicators, especially in the recall rate. The system of the present invention significantly improves the ability to identify potential safety hazards, reaching a recall rate of 89.7%, while the traditional system is only 78.6%.

[0146] Optionally, Table 2 shows the comparison results of the multi-level electronic fence monitoring system of the present invention and the traditional method on the UndergroundMonitoring2019 dataset.

[0147] method Accuracy (%) Accuracy (%) Recall rate (%) F1 score (%) Traditional monitoring system 82.7 80.1 75.5 77.8 System of the present invention 91.9 90.5 87.3 88.9

[0148] On the UndergroundMonitoring2019 dataset, the proposed system significantly outperformed the traditional system in all evaluation indicators, especially in recall rate and F1 score, which reached 87.3% and 88.9% respectively, which were 11.8% and 11.1% higher than the traditional system. These results show that the proposed system can more accurately and comprehensively identify potential abnormal situations in the underground operation environment, significantly improving the safety and monitoring efficiency of the system.

[0149] In order to further verify the performance of the system in the actual underground operation environment, the experimenters deployed the multi-level electronic fence monitoring system of the present invention in a certain mining area and compared it with the existing traditional monitoring system in the mining area. During the one-month test period, a total of 12 abnormal events occurred in the mining area, involving rapid temperature increase, fence damage, and excessive harmful gases.

[0150] In one test scenario, the monitoring point on the south side of the mining area detected a sudden rise in temperature and the carbon monoxide concentration gradually exceeded the standard. In the traditional system, the sensor triggered an alarm, but the operator failed to discover in time that similar situations had occurred in the surrounding areas, resulting in untimely follow-up disposal. In the system of the present invention, the hypergraph neural network quickly associated the data of the monitoring point on the south side and the surrounding monitoring points through the data linkage mechanism, identified the potential abnormal situation in the area, and issued an alarm within 5 seconds to notify the operator to evacuate urgently. In the same situation, the traditional system took about 12 seconds to respond. In the end, the system of the present invention effectively avoided the further expansion of the accident and reduced operational losses.

[0151] Through tests in public data sets and actual mining operating environments, the multi-level electronic fence monitoring system of the present invention has demonstrated significant advantages in processing complex underground monitoring data. Whether in terms of data processing efficiency, anomaly detection accuracy, or multi-point linkage response speed, the system of the present invention is significantly superior to traditional monitoring methods. The results show that the system of the present invention can effectively improve the safety protection level of the underground working environment, reduce the occurrence of accidents, and has broad application prospects.

[0152] The present invention introduces a hypergraph neural network model, takes the data of the physical layer, sensor layer and control layer in the underground operating environment as nodes in the hypergraph, and constructs hyperedges by utilizing the correlation between the state of the physical layer data, the sensor layer environmental data and the control layer feedback data. Through the characteristics of the hypergraph neural network, it effectively processes multi-source and multi-dimensional data, fully captures the complex relationship between data at each level, and then generates a high-dimensional data representation. This not only improves the system's processing efficiency for multi-dimensional data, but also greatly improves the accuracy of identifying potential dangers and abnormal situations, overcoming the problem of difficulty in integrating multi-level data in traditional monitoring systems.

[0153] The present invention utilizes a small amount of labeled data and a large amount of unlabeled data, and trains and optimizes the hypergraph neural network model through a weakly supervised learning framework, thereby reducing the dependence on manually labeled data and significantly improving the model's ability to process unlabeled data. Weakly supervised learning achieves efficient optimization of unlabeled data through the mechanism of comparative learning, and enhances the model's adaptability in complex environments. Compared with the existing system that relies on fixed rules for anomaly detection, the present invention can adjust the electronic fence monitoring strategy in real time according to environmental changes, reduce missed reports and false alarms, and further improve the stability and reliability of the monitoring system.

[0154] The present invention constructs a linkage matrix between monitoring points so that multiple monitoring points underground can achieve data linkage and coordination, ensuring the efficient operation of the monitoring system within the local area network. By calculating the data linkage strength between different monitoring points, the monitoring data of high-risk areas is processed preferentially, and a multi-point data sharing mechanism is used to achieve rapid response and dynamic protection. Compared with the traditional single-point monitoring method, the large-scale multi-point linkage design of the present invention significantly shortens the response time of the system, improves the ability to respond to emergencies in the underground working environment, and comprehensively enhances the overall safety protection effect.

[0155] In this embodiment, a multi-level monitoring data acquisition system is constructed in an underground operating environment and a multi-level monitoring data set is collected; the multi-level monitoring data sets in multiple monitoring points are transmitted through a local area network, and the local area network covers multiple monitoring points in the underground operating area; a hypergraph neural network model is constructed for the multi-level monitoring data sets collected from multiple monitoring points, and the data relationship between each monitoring point and the level is modeled based on the hypergraph neural network model, the relationship weight between each node is calculated, and the hypergraph neural network model is optimized through iterative training; based on the optimized hypergraph neural network model, the electronic fence monitoring strategy in the underground operating environment is dynamically adjusted, and potential safety hazards are identified according to the multi-dimensional data representation of the physical layer, the sensor layer and the control layer, and real-time warning is given for abnormal data; using the optimized monitoring strategy, the data linkage of multiple monitoring points is coordinated through the multi-dimensional data representation, so as to realize the safety protection of the underground operation electronic fence, solve the technical problem of low efficiency of safety monitoring of the underground operating environment, and achieve the technical effect of improving the efficiency of safety monitoring of the underground operating environment.

[0156] According to an embodiment of the present invention, a monitoring device for downhole operations is also provided. It should be noted that the monitoring device for downhole operations can be used to execute the downhole operations monitoring method in the method embodiment.

[0157] Figure 4 Schematic diagram of a monitoring device for underground operations according to an embodiment of the present invention. Figure 4 As shown, the downhole operation monitoring device 400 may include: an acquisition unit 401 , a construction unit 402 , a determination unit 403 and a monitoring unit 404 .

[0158] The acquisition unit 401 is used to acquire a monitoring data set of a downhole operation environment, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored on the downhole operation environment in multiple monitoring dimensions.

[0159] The construction unit 402 is used to construct a target monitoring point model based on the monitoring data set, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation and the data transmission rules between different monitoring points.

[0160] The determination unit 403 is used to determine a target monitoring strategy based on the target monitoring point model, wherein the target monitoring strategy is used to indicate a rule for monitoring downhole operations.

[0161] The monitoring unit 404 is used to monitor the downhole operation according to the target monitoring strategy.

[0162] Optionally, the acquisition unit 401 may include: an acquisition module for acquiring multiple monitoring data of the downhole operating environment, wherein the monitoring data includes at least one of the following: data of each monitoring point or multi-layer data, and the multi-layer data includes at least: control layer data, sensor layer data and physical layer data; an integration module for acquiring a monitoring data set, including: integrating multiple monitoring data to obtain a monitoring data set.

[0163] Optionally, the construction unit 402 may include: an extraction module, used to extract multiple monitoring point data and multi-layer data from the monitoring data set; a first determination module, used to determine each monitoring point data as a model node; a second determination module, used to determine the multi-layer data as a model edge; a construction module, used to construct an initial monitoring point model based on the model nodes and model edges; and an adjustment module, used to adjust the initial monitoring point model to obtain a target monitoring point model.

[0164] Optionally, the adjustment module may include: an extraction submodule, used to extract attribute features from model nodes and model edges, wherein the attribute features are used to indicate features related to the attributes of the model nodes and model edges; a generation submodule, used to generate multi-dimensional data based on the attribute features; and an adjustment submodule, used to adjust the initial monitoring point model based on the multi-dimensional data to obtain a target monitoring point model.

[0165] Optionally, the adjustment submodule can also be used to determine the weights between each model node based on multi-dimensional data, wherein the weight is used to represent the importance of each model node; based on the weight, the initial monitoring point model is adjusted to obtain the target monitoring point model.

[0166] Optionally, the monitoring device 400 for underground operations may further include: an identification unit for identifying abnormal data from the multi-dimensional data; and an output unit for outputting the abnormal data in response to the existence of the abnormal data.

[0167] In this embodiment, a monitoring data set of the downhole operation environment is obtained, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment under multiple monitoring dimensions; based on the monitoring data set, a target monitoring point model is constructed, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; based on the target monitoring point model, a target monitoring strategy is determined, wherein the target monitoring strategy is used to indicate the rules for monitoring the downhole operation; and the downhole operation is monitored according to the target monitoring strategy. That is to say, the embodiment of the present invention processes the downhole operation environment data set and constructs a target monitoring point model, thereby monitoring the downhole environment according to the target monitoring strategy determined by the target monitoring point model, thereby achieving the purpose of monitoring the safety of the downhole operation environment, thereby solving the technical problem of low efficiency of safety monitoring of the downhole operation environment, and achieving the technical effect of improving the efficiency of safety monitoring of the downhole operation environment.

[0168] According to an embodiment of the present invention, there is further provided a computer-readable storage medium, the storage medium comprising a stored program, wherein the program executes the method for monitoring downhole operations in the method embodiment.

[0169] According to an embodiment of the present invention, a processor is further provided. The processor is used to run a program, wherein the program executes the method for monitoring downhole operations in the method embodiment when it is run.

[0170] According to an embodiment of the present invention, a computer program product is further provided. The computer program product includes computer instructions. When the computer instructions are executed by a processor, the method for monitoring downhole operations in the method embodiment is implemented.

[0171] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0172] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0173] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0174] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0175] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0176] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent functional component, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software functional component, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store program codes.

[0177] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for monitoring underground operations, characterized in that: include: Acquire a monitoring data set of a downhole operation environment, wherein the downhole operation environment is used to represent the environment in which the downhole operation is located, and the monitoring data set includes data monitored for the downhole operation environment in multiple monitoring dimensions; Based on the monitoring data set, construct a target monitoring point model, wherein the target monitoring point model is used to indicate the association relationship between different monitoring points in the downhole operation, and the data transmission rules between different monitoring points; Determining a target monitoring strategy based on the target monitoring point model, wherein the target monitoring strategy is used to indicate a rule for monitoring the downhole operation; The downhole operation is monitored according to the target monitoring strategy.

2. The method according to claim 1, characterized in that The method further comprises: Acquire a plurality of monitoring data of the downhole operation, wherein the monitoring data at least includes: data of each monitoring point or multi-layer data, and the multi-layer data at least includes: control layer data, sensor layer data and physical layer data; Acquiring a monitoring data set includes: integrating a plurality of the monitoring data to obtain the monitoring data set.

3. The method according to claim 1, characterized in that Based on the monitoring data set, a target monitoring point model is constructed, including: Extracting multiple monitoring point data and multiple layers of data from the monitoring data set; Determine each of the monitoring point data as a model node, and determine the multi-layer data as a model edge; Based on the model nodes and the model edges, construct an initial monitoring point model; The initial monitoring point model is adjusted to obtain the target monitoring point model.

4. The method according to claim 3, characterized in that: Adjusting the initial monitoring point model to obtain the target monitoring point model includes: Extracting attribute features from the model nodes and the model edges, wherein the attribute features are used to indicate features related to the attributes of the model nodes and the model edges; Based on the attribute characteristics, generate multi-dimensional data; Based on the multi-dimensional data, the initial monitoring point model is adjusted to obtain the target monitoring point model.

5. The method according to claim 4, characterized in that Based on the multi-dimensional data, the initial monitoring point model is adjusted to obtain the target monitoring point model, including: Based on the multi-dimensional data, determining the weights between the model nodes, wherein the weights are used to represent the importance of the model nodes; Based on the weights, the initial monitoring point model is adjusted to obtain the target monitoring point model.

6. The method according to any one of claims 4-5, characterized in that: The method further comprises: Identifying abnormal data from the multi-dimensional data; In response to the abnormal data existing, the abnormal data is output.

7. A monitoring device for underground operations, characterized in that: include: An acquisition unit, used for acquiring a monitoring data set, wherein the monitoring data set is used for indicating various aspects of the downhole operating environment; A construction unit, configured to construct a target monitoring point model based on the monitoring data set, wherein the target monitoring point model is used to indicate the relationship between different monitoring points in the downhole operation and the data transmission mode; A determination unit, configured to determine a target monitoring strategy based on the target monitoring point model, wherein the target monitoring strategy is used to indicate a rule for monitoring the downhole operation; A monitoring unit is used to monitor the downhole operation according to the target monitoring strategy.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 6.

9. A processor, characterized in that: The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 6 when running.

10. A computer program product, characterized in that The computer program product comprises computer instructions, which implement the method according to any one of claims 1 to 6 when executed by a processor.

Images