Method, device and electronic equipment for monitoring support action of hydraulic support

Abstract

The present disclosure relates to the technical field of intelligent mining equipment, and in particular to a method, device and electronic device for monitoring the support action of a hydraulic support. The method comprises: collecting multiple support data of a hydraulic support, and determining the correlation between the multiple support data; using the multiple support data as network nodes and the correlation as connecting edges to construct a complex network; using the complex network to perform dynamic trend analysis on the hydraulic support to obtain the action trend of the hydraulic support. The present disclosure using the above scheme can predict the action trend of the hydraulic support and reduce the probability of accident risks.

Description

Technical Field

[0001] The present disclosure relates to the technical field of intelligent mining equipment, and in particular to a method, device and electronic equipment for monitoring the supporting action of a hydraulic support. Background Art

[0002] Hydraulic supports are widely used support equipment in underground projects such as coal mines. Real-time monitoring of the support action of hydraulic supports is crucial to ensure the safety and efficiency of the working face. Traditional support support action monitoring methods mainly rely on manual inspections and simple sensors. However, manual inspections are inefficient and it is difficult to detect abnormal situations in a timely manner. Data collection and analysis methods are backward, and only abnormal analysis can be performed based on current data. The current status can only be controlled in real time. It cannot avoid risks in advance, and it is difficult to conduct a comprehensive analysis and prediction of the support action. In an emergency, only emergency repairs and maintenance can be carried out, which makes the risk of accidents on site extremely high. Summary of the invention

[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first purpose of the present disclosure is to propose a method for monitoring the support action of a hydraulic support so as to predict the action trend of the hydraulic support and reduce the probability of accident risks.

[0005] The second objective of the present disclosure is to provide a support action monitoring device for a hydraulic support.

[0006] A third objective of the present disclosure is to provide an electronic device.

[0007] A fourth objective of the present disclosure is to provide a computer-readable storage medium.

[0008] A fifth object of the present disclosure is to provide a computer program product.

[0009] To achieve the above-mentioned purpose, the first embodiment of the present disclosure proposes a method for monitoring the supporting action of a hydraulic support, comprising:

[0010] Collect various support data of hydraulic supports and determine the correlation between the various support data;

[0011] Use various scaffold data as network nodes and association relationships as connecting edges to build a complex network;

[0012] The complex network is used to analyze the dynamic trend of the hydraulic support and obtain the action trend of the hydraulic support.

[0013] Optionally, the support data includes at least one of the following:

[0014] Bracket vibration frequency;

[0015] The vibration amplitude of the bracket;

[0016] Friction between the support and the coal seam;

[0017] The amount of electrostatic charge on the surface of the bracket;

[0018] Bracket load;

[0019] Bracket displacement;

[0020] Bracket inclination angle.

[0021] Optionally, determining the association relationship between multiple scaffold data includes:

[0022] Classify and label various scaffold data according to data types to obtain labeled scaffold data;

[0023] Segmenting the marked stent data according to a time window of a preset length to obtain multiple data segments;

[0024] Extracting features from each of the multiple data segments, and performing dimensionality reduction processing on the extracted features to obtain dimensionality reduction features corresponding to each data segment;

[0025] Different types of dimensionality reduction features are fused and analyzed to obtain the correlation between various scaffold data.

[0026] Optionally, the method further comprises:

[0027] Analyze different types of dimensionality reduction features to obtain analysis results corresponding to each type of dimensionality reduction features, and perform status assessment on the hydraulic support based on the analysis results to obtain status assessment results;

[0028] If the analysis result meets the optimization requirements, the action of the hydraulic support is controlled to optimize the support data corresponding to the analysis result.

[0029] Optionally, after collecting various support data of the hydraulic support, the method further includes:

[0030] Perform outlier detection on various stent data;

[0031] If there are abnormal values ​​in the data of multiple stents, the abnormal values ​​will be corrected.

[0032] Optionally, when performing outlier detection on multiple stent data, at least one of the following detection methods is used:

[0033] Outlier detection based on statistical models;

[0034] Outlier detection based on cluster analysis;

[0035] Outlier detection based on machine learning models;

[0036] Outlier detection based on time series trend analysis;

[0037] Outlier detection based on cross-validation of multi-source data.

[0038] Optionally, a complex network is used to perform dynamic trend analysis on the hydraulic support, including:

[0039] Determine the risk indicators corresponding to complex networks;

[0040] Predict the risk value of each node in a complex network based on risk indicators and correlation relationships;

[0041] The risk propagation simulation is carried out according to the complex network and risk value to obtain the action trend of the hydraulic support.

[0042] Optionally, after obtaining the action trend of the hydraulic support, the method further includes:

[0043] Analyze the action trend and obtain risk action information;

[0044] If the risk action information meets the early warning requirements, a warning signal will be issued.

[0045] To achieve the above-mentioned purpose, a second aspect of the present disclosure provides a support action monitoring device for a hydraulic support, comprising:

[0046] A data acquisition unit, used to acquire various support data of the hydraulic support and determine the correlation relationship between the various support data;

[0047] A network construction unit is used to construct a complex network by taking various scaffold data as network nodes and the association relationships as connecting edges;

[0048] The action prediction unit is used to perform dynamic trend analysis on the hydraulic support by using a complex network to obtain the action trend of the hydraulic support.

[0049] To achieve the above-mentioned purpose, a third aspect of the present disclosure provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0050] Memory stores computer-executable instructions;

[0051] The processor executes the computer-executable instructions stored in the memory to implement the method shown in any one of the first aspects above.

[0052] To achieve the above-mentioned objectives, an embodiment of the fourth aspect of the present disclosure proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method shown in any one of the above-mentioned first aspects.

[0053] To achieve the above-mentioned objective, a fifth aspect of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the method shown in any one of the above-mentioned first aspects.

[0054] In summary, the method, device, equipment and storage medium provided by the present disclosure collect various support data of hydraulic supports and determine the correlation between various support data; use various support data as network nodes and the correlation as connecting edges to construct a complex network; use the complex network to perform dynamic trend analysis on the hydraulic supports to obtain the action trend of the hydraulic supports. Therefore, the action of the support can be predicted by combining existing data and data analysis, and a more accurate and real-time support action prediction can be provided, thereby improving support efficiency, optimizing support maintenance and enhancing support safety, and reducing the probability of accident risks.

[0055] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0057] Figure 1 A schematic diagram of a flow chart of a method for monitoring a supporting action of a hydraulic support provided in an embodiment of the present disclosure;

[0058] Figure 2 A schematic flow chart of a method for monitoring a supporting action of a hydraulic support provided by another embodiment of the present disclosure;

[0059] Figure 3 A schematic diagram of the structure of a supporting action monitoring device for a hydraulic support provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0061] The present disclosure is described in detail below with reference to specific embodiments.

[0062] In the first embodiment, if Figure 1 As shown, Figure 1 The present invention provides a flow chart of a method for monitoring the supporting action of a hydraulic support provided in an embodiment of the present invention. The method can be implemented by a computer program and can be run on a device for monitoring the supporting action of a hydraulic support. The computer program can be integrated into an application or run as an independent tool application.

[0063] Among them, the supporting action monitoring device of the hydraulic support can be an electronic device with the supporting action monitoring function of the hydraulic support.

[0064] Among them, the supporting action monitoring method of the hydraulic support can be executed by an electronic device.

[0065] For example, the method for monitoring the supporting action of the hydraulic support includes the following steps:

[0066] S101, collecting various support data of the hydraulic support, and determining the correlation relationship between the various support data;

[0067] According to some embodiments, the support data refers to data related to a hydraulic support.

[0068] It is easy to understand that when the electronic device performs the support action monitoring of the hydraulic support, the electronic device can collect various support data of the hydraulic support and determine the correlation relationship between the various support data.

[0069] S102, using various scaffold data as network nodes and association relationships as connecting edges to construct a complex network;

[0070] It is easy to understand that when an electronic device obtains the association relationship between multiple types of bracket data, the electronic device can use the multiple types of bracket data as network nodes and the association relationship as a connecting edge to build a complex network.

[0071] S103, using a complex network to perform dynamic trend analysis on the hydraulic support to obtain the action trend of the hydraulic support.

[0072] It is easy to understand that when the electronic device obtains the complex network, the electronic device can use the complex network to perform dynamic trend analysis on the hydraulic support to obtain the action trend of the hydraulic support.

[0073] In summary, the method provided in this embodiment collects multiple support data of the hydraulic support and determines the correlation between the multiple support data; uses the multiple support data as network nodes and the correlation as the connecting edge to construct a complex network; uses the complex network to perform dynamic trend analysis on the hydraulic support to obtain the action trend of the hydraulic support. Therefore, the action of the support can be predicted by combining existing data and data analysis, and a more accurate and real-time support action prediction can be provided, thereby improving support efficiency, optimizing support maintenance and enhancing support safety, and reducing the probability of accident risks.

[0074] This embodiment also provides another method for monitoring the supporting action of a hydraulic support, which can be executed by an electronic device.

[0075] For example, Figure 2 As shown, the supporting action monitoring method of the hydraulic support may include the following steps:

[0076] S201, collecting various support data of the hydraulic support;

[0077] According to some embodiments, the collected stent data may include at least one of the following:

[0078] Bracket vibration frequency;

[0079] The vibration amplitude of the bracket;

[0080] Friction between the support and the coal seam;

[0081] The amount of electrostatic charge on the surface of the bracket;

[0082] Bracket load;

[0083] Bracket displacement;

[0084] Bracket inclination angle.

[0085] In some embodiments, the bracket data may be collected using devices such as acceleration sensors, friction sensors, electrostatic sensors, pressure sensors, telescopic displacement sensors, and inclination sensors.

[0086] Among them, the acceleration sensor can be installed on the top beam and base of the hydraulic support to accurately sense the overall vibration;

[0087] Among them, the friction sensor can be installed on the supporting surface where the support contacts the coal seam, and multiple friction sensors are evenly distributed to obtain comprehensive friction data;

[0088] Among them, electrostatic sensors can be installed in places where static electricity is easily accumulated, such as metal contact surfaces, movable joints, etc.

[0089] Among them, the pressure sensor can be installed at the bottom of the hydraulic support to monitor the support load by measuring the pressure of the hydraulic oil;

[0090] Among them, the telescopic displacement sensor can be installed on the top of the hydraulic support, using laser or ultrasonic technology to measure the telescopic and displacement of the support;

[0091] Among them, the inclination sensor can be installed at the top center, top edge, and top connection of the hydraulic support to measure the inclination of the support.

[0092] S202, classifying and labeling the various types of stent data according to data types to obtain labeled stent data;

[0093] According to some embodiments, various types of supported data collected, including but not limited to supported vibration frequency, supported vibration amplitude, friction between the supported and coal seam, electrostatic charge on the supported surface, supported load, supported displacement, supported inclination and other data, can be classified according to data type, and specific tags can be added to each type of data for subsequent processing and identification.

[0094] It should be noted that the labeled stent data can be standardized and normalized to make data of different types and magnitudes comparable. For example, both pressure data and displacement data can be mapped to a range of 0 to 1.

[0095] In some embodiments, the data after standardization and normalization can also be verified to check the integrity, accuracy and consistency of the data. The effect of data processing can be evaluated by calculating data quality indicators such as accuracy and recall, and the processing process can be adjusted and optimized based on the evaluation results.

[0096] According to some embodiments, after obtaining the labeled bracket data, the labeled bracket data may be compressed to reduce the overhead of data storage and transmission. For example, a lossless compression algorithm such as Huffman coding may be used.

[0097] It should be noted that, in order to improve the accuracy of subsequent monitoring, outlier detection may be performed on multiple stent data; if outliers exist in multiple stent data, the outliers are corrected.

[0098] In some embodiments, when performing outlier detection on multiple stent data, at least one of the following detection methods may be used:

[0099] Outlier detection based on statistical models;

[0100] Outlier detection based on cluster analysis;

[0101] Outlier detection based on machine learning models;

[0102] Outlier detection based on time series trend analysis;

[0103] Outlier detection based on cross-validation of multi-source data.

[0104] Among them, in the process of outlier detection based on statistical models, each type of support data, such as the mean, median, standard deviation and interquartile range of the support vibration frequency, support vibration amplitude, and the friction between the support and the coal seam, can be calculated; using the box plot, the data points outside the upper and lower whisker ranges of the box plot are marked as preliminary outliers. At the same time, the data points whose distance from the mean exceeds 3 times the standard deviation and whose skewness or kurtosis obviously deviates from the normal distribution characteristics are also marked as preliminary outliers;

[0105] In the process of outlier detection based on cluster analysis, the hierarchical clustering algorithm combined with the K-Means clustering algorithm can be used to cluster the data. First, hierarchical clustering can be used to determine the approximate number of cluster categories, and then K-Means can be used for precise clustering; for data points that cannot be attributed to any cluster center, the Mahalanobis distance to each cluster center is calculated, and the data points with a Mahalanobis distance greater than the threshold are regarded as potential outliers; in addition, the density distribution of the clusters is analyzed, and the data points in the low-density area are focused on to determine whether they are potential outliers.

[0106] Among them, in the process of outlier detection based on machine learning models, a deep learning model based on Long Short-Term Memory (LSTM) or Convolutional Neural Network (CNN) can be constructed, and a large amount of normal data and some manually labeled abnormal data are used for training. The data to be detected is input into the model. The model not only outputs the judgment of whether the data point is an outlier, but also gives a probability estimate of the degree of abnormality. For data points with uncertain model judgment results, further analysis can be carried out in combination with their distribution characteristics in the data set and their correlation with other data;

[0107] In the process of outlier detection based on time series trend analysis, for time series data, such as the change of vibration frequency over time, a trend line can be fitted, such as a linear, polynomial or exponential trend line. Data points that deviate greatly from the trend line can be considered outliers;

[0108] In the process of outlier detection based on cross-validation of multi-source data, the intrinsic relationship between different types of data can be combined. For example, vibration frequency and pressure usually have a certain correlation. When a data point is considered an anomaly in its own data sequence and its relationship with related data does not meet expectations, it can be confirmed as an outlier.

[0109] In some embodiments, when correcting outliers, the anomaly type of the outliers can be determined. If the outlier is a single-point anomaly caused by random noise, the average value of adjacent data points can be used for correction; if it is multiple consecutive outliers, linear interpolation or polynomial interpolation methods based on time series can be used for correction; for outliers that cannot be corrected by the above methods, they can be marked and specially processed in subsequent analysis, and relevant information of the abnormal situation can be recorded to further investigate the cause.

[0110] Take a scenario as an example. Suppose 1,000 data points of bracket vibration frequency are collected and outlier detection is performed using the following steps:

[0111] First, the corresponding mean is calculated to be 50Hz, the median is 48Hz, the standard deviation is 10Hz, the skewness is 0.1, and the kurtosis is 0.2. The upper whisker of the box plot is 70Hz, and the lower whisker is 30Hz. One of these data points is 90Hz, which is more than 3 times the standard deviation from the mean (3×10=30Hz) and is obviously beyond the range of the box plot, so this data point can be marked as a preliminary outlier.

[0112] Next, a hierarchical clustering algorithm was used to preliminarily determine that the data could be divided into five cluster categories, and then the K-Means algorithm was used for precise clustering. Assume that the cluster centers are 40Hz, 45Hz, 50Hz, 55Hz, and 60Hz respectively. The Mahalanobis distance from each data point to the cluster center was calculated. Among these data points, the Mahalanobis distance from one data point to its nearest cluster center is 25, and the set threshold is 20, so this data point is considered a potential outlier. At the same time, it was found that the density of one cluster was significantly lower than that of the other clusters, and the data points in this low-density cluster were focused on. After further analysis, some of these data points were confirmed as potential outliers.

[0113] After that, the LSTM model was trained with 5,000 normal vibration frequency data and 500 manually labeled abnormal data. These 1,000 data points were input into the trained LSTM model. If a new data point was input, the model outputted a probability of 0.85 for it to be an outlier, indicating that the data point was more likely to be an outlier.

[0114] Finally, for the uncertain data points with an output probability of about 0.5 in the LSTM model, further analysis is performed in combination with their distribution characteristics in the data set, such as the distance to other data points and the correlation with other data, such as pressure data.

[0115] S203, segmenting the marked stent data according to a time window of a preset length to obtain a plurality of data segments;

[0116] Taking a scenario as an example, you can use 5 seconds as a window to divide long-term data into multiple short-term data segments.

[0117] S204, extracting features from each of the multiple data segments, and performing dimensionality reduction processing on the extracted features to obtain dimensionality reduction features corresponding to each data segment;

[0118] According to some embodiments, when performing feature extraction, the extracted features include but are not limited to mean, variance, peak value, valley value, etc.

[0119] In some embodiments, principal components analysis (PCA) or other dimensionality reduction algorithms may be used to perform dimensionality reduction processing on the extracted features, thereby reducing the dimension of the data while retaining the main information.

[0120] S205, performing fusion analysis on different types of dimensionality reduction features to obtain correlation relationships between multiple scaffold data;

[0121] For example, during the fusion analysis, the correlation between the bracket vibration frequency and bracket vibration amplitude and the bracket load, or the correlation pattern between the bracket displacement and friction force can be studied.

[0122] It should be noted that different types of dimensionality reduction features can also be analyzed to obtain analysis results corresponding to each type of dimensionality reduction feature, and the state of the hydraulic support can be evaluated based on the analysis results to obtain state evaluation results; if the analysis results meet the optimization requirements, the action of the hydraulic support is controlled to optimize the support data corresponding to the analysis results.

[0123] According to some embodiments, in the process of performing status assessment on the hydraulic support according to the analysis results, the assessment objectives include but are not limited to the health status of the support, the remaining service life of the support, etc.

[0124] In some embodiments, the process of optimizing the support data corresponding to the analysis results includes but is not limited to the extension and retraction speed, load and inclination angle of the support, thereby improving support efficiency and safety.

[0125] For example, if the analysis results indicate that the extension and retraction speed of the bracket is abnormal and needs to be optimized, the extension and retraction speed of the bracket can be slowed down to prevent further damage; if the analysis results indicate that the load of the bracket is abnormal and needs to be optimized, the bracket pressure can be increased to improve the stability of the bracket; if the analysis results indicate that the inclination angle of the bracket is abnormal and needs to be optimized, the inclination angle of the bracket can be adjusted to ensure the stability of the bracket.

[0126] In some embodiments, when any type of dimensionality reduction feature is greater than a predefined threshold, the analysis result may indicate that the type of dimensionality reduction feature is abnormal. Alternatively, when any type of dimensionality reduction feature matches a known abnormal pattern, the analysis result may indicate that the type of dimensionality reduction feature is abnormal.

[0127] In some embodiments, if the analysis result meets the optimization requirements, an alarm signal may be issued during the optimization to indicate abnormal behavior of the support. The alarm signal is used to indicate that an event requiring the operator's attention has been detected.

[0128] It should be noted that if any type of dimensionality reduction feature is greater than the emergency response threshold, the hydraulic support can be controlled to stop working to prevent casualties or equipment damage. At the same time, an emergency alarm signal can be triggered to notify the operator to take immediate action. Secondly, if the analysis results indicate that there is an electrical fault in the liquid support system, the power supply can also be cut off to prevent fire or explosion.

[0129] In some embodiments, the emergency alert signal indicates that a serious abnormal situation has been detected and immediate action is required.

[0130] S206, using various scaffold data as network nodes and association relationships as connecting edges to construct a complex network;

[0131] According to some embodiments, processed bracket-related data, such as vibration frequency, amplitude, friction, load, displacement, inclination, etc., can be regarded as nodes in a network. Based on the correlation between the data, for example, vibration frequency and pressure may be correlated in some cases, connecting edges between the nodes are established to obtain a complex network.

[0132] For example, node 1 represents vibration frequency data, node 2 represents pressure data, node 3 represents displacement data, etc. A connection edge is established based on the association between the data. For example, vibration frequency and pressure may be associated in some cases, so a connection edge can be established between node 1 and node 2.

[0133] S207, determining risk indicators corresponding to the complex network;

[0134] According to some embodiments, a series of indicators that can indicate risk actions of stents can be determined in combination with domain knowledge and actual application scenarios. These indicators can be comprehensively determined based on statistical characteristics of data and outlier detection results.

[0135] For example, when the vibration frequency exceeds the normal range of [specific value], or the pressure suddenly rises to more than [specific value], it is defined as having a certain risk.

[0136] S208, predicting the risk value of each node in the complex network according to the risk indicators and the correlation relationship;

[0137] According to some embodiments, for each node, that is, a certain data, its risk value can be evaluated based on its historical data and current data; for example, if the vibration frequency at a certain moment exceeds the normal range and continues to increase, or the pressure increases abnormally, it may indicate that there is a higher risk in the aspect corresponding to the node.

[0138] In some embodiments, a graph convolutional neural network (GCN) may be used as a basic model to evaluate the risk value of each node in a complex network.

[0139] Among them, the input of the GCN model is a matrix representing node features and connection edge features. The node features include various processed bracket data, such as vibration frequency, amplitude, friction, etc., as well as corresponding risk assessment indicators; the connection edge features represent the association relationship between nodes, including connection relationship and strength.

[0140] In some embodiments, the following node risk assessment formula is used in the GCN model:

[0141]

[0142] in, Risk nodei Representation Node i risk value. x i Representation Node i The feature vector includes historical data and current data. θ Represents the model parameters of the GCN model. f Represents a risk assessment function, which can be a function based on a neural network.

[0143] Taking the vibration frequency node as an example, assuming that at a certain moment, the value of the vibration frequency is [specific frequency value]. By comparing it with the normal range, it is found that it exceeds the normal range and has shown a continuous upward trend over the past period of time. Then the following risk assessment function can be used f Calculate its risk value Risk node1 :

[0144]

[0145] in, α and β is the weight coefficient, CurrentFrequency is the current vibration frequency value, NormalFrequency is the middle value of the normal vibration frequency range, TrendFactor It is a numerical value that represents a trend. For example, a continuous rise is 1, a stable state is 0, and a decline is -1.

[0146] S209, performing risk propagation simulation according to the complex network and the risk value to obtain the action trend of the hydraulic support;

[0147] According to some embodiments, risk propagation simulation refers to simulating the risk propagation process in a complex network. When the risk level of a node exceeds a threshold, the risk is propagated to other nodes connected to it according to preset propagation rules; the propagation rules can take into account factors such as the connection strength between nodes and the correlation between data.

[0148] In some embodiments, GCN may be used as a basic model to predict the risk value of each node in a complex network.

[0149] In some embodiments, the GCN model can predict the risk value according to the following formula:

[0150]

[0151] in, PredictedRisk nodei Representation Node i The predicted risk value of NetworkStruture is the structural information of the entire complex network. g is the prediction function, are the parameters of the prediction function.

[0152] According to some embodiments, the prediction function g In the risk propagation simulation, the following risk propagation formula can be used:

[0153]

[0154] in, Is a node j exist t +1 moment risk value, Is a node j exist t The risk value at the moment, Neighbors ( j ) is a node j The set of neighbor nodes of w ij Is a node i To Node j The edge weights of Transfer ij Is a node i To Node j risk transmission coefficient.

[0155] It should be noted that the key nodes and paths in the network that have an important impact on risk propagation are found. These key nodes are the places where risks are prone to occur or spread quickly, and the key paths are the main ways for risk propagation. The corresponding parameters are w ij and Transfer ij , which can be obtained by analyzing the association relationship.

[0156] Take a scenario as an example. Assume that the risk value of the vibration frequency node (node ​​1) is high. There is an edge connecting it to the pressure node (node ​​2), and the edge weight is w 12 , the risk transmission coefficient is Transfer 12 .

[0157] Then the predicted risk value of the pressure node is:

[0158]

[0159] It should be noted that the model used in this embodiment can be regularly updated through the continuous collection of new data. Model updating can improve prediction accuracy and adapt to changes in stent motion over time.

[0160] S210, analyzing the action trend to obtain risk action information;

[0161] According to some embodiments, risks may be simulated multiple times and the spread of risks in the network may be observed to analyze the trend of risk actions of the support.

[0162] For example, are there certain nodes or connections that are prone to cause the spread of risks? Do the patterns and speeds of risk transmission change under different working conditions or time periods?

[0163] Taking a scenario as an example, risk action information may include that the risk of vibration frequency nodes is often propagated to pressure nodes and displacement nodes, and the propagation speed is faster under certain working conditions.

[0164] S211: If the risk action information meets the warning requirements, a warning signal is issued.

[0165] According to some embodiments, based on the analysis of risk propagation trends, the probability and time range of possible risk actions of the stent in the future can be predicted. When the risk propagates to a certain extent or the risk indicator of a key node reaches a warning threshold, a corresponding warning signal is issued so that timely maintenance or adjustment measures can be taken.

[0166] For example, suppose that after a period of observation and simulation, it is found that when the risk value of the vibration frequency node exceeds the threshold value, it is likely to cause some risky actions of the bracket, such as component damage. Then in actual monitoring, when the risk value of the vibration frequency node approaches or reaches the threshold, an early warning signal can be issued so that corresponding maintenance measures can be taken.

[0167] In some embodiments, the early warning signal may, for example, indicate that a potential abnormal situation has been detected and requires close attention.

[0168] It should be noted that the early warning signal, alarm signal and emergency alarm signal used in this embodiment can be used, for example, by using an audible and visual alarm device and an application (Application, app) to warn the operator.

[0169] In some embodiments, the stent data collected in this embodiment and the subsequent analysis results, early warning signals, alarm signals and emergency alarm signals, event records and response measures data can all be stored in a database.

[0170] In summary, the method provided in this embodiment, first, collects various support data of hydraulic supports; classifies and labels various support data according to data types to obtain labeled support data; divides the labeled support data according to a time window of preset length to obtain multiple data segments; extracts features for each data segment in the multiple data segments, and performs dimensionality reduction processing on the extracted features to obtain the dimensionality reduction features corresponding to each data segment; performs fusion analysis on different types of dimensionality reduction features to obtain the association relationship between various support data; therefore, the accuracy of determining the association relationship can be improved. Then, various support data are used as network nodes, and the association relationship is used as a connecting edge to construct a complex network; determine the risk index corresponding to the complex network; predict the risk value of each node in the complex network based on the risk index and the association relationship; perform risk propagation simulation based on the complex network and the risk value to obtain the action trend of the hydraulic support; therefore, the action of the support can be predicted by combining existing data and data analysis, and a more accurate and real-time support action prediction can be provided, thereby improving support efficiency, optimizing support maintenance and enhancing support safety, and reducing the probability of accident risks. Finally, by analyzing the movement trend, risk movement information is obtained; if the risk movement information meets the warning requirements, a warning signal is issued; therefore, the risk movement of the bracket can be predicted, thereby reminding the staff to conduct inspections and maintenance in the corresponding places in advance, so as to avoid risks in advance.

[0171] In order to implement the above embodiment, the present disclosure further proposes a support action monitoring system for a hydraulic support, comprising:

[0172] Data acquisition module: used to collect various support data of hydraulic supports;

[0173] Data processing module: used to process the collected data of various stents;

[0174] Communication module: used to transmit processed data to the control module;

[0175] Control module: used to receive, store and analyze processed data;

[0176] Alarm module: alarm the operator through sound and light alarm device and app;

[0177] Early warning and emergency response module: When the monitoring data exceeds the preset threshold, an alarm is issued and emergency procedures are initiated;

[0178] Human-computer interaction module: displays real-time data, alarm information, historical records and analysis reports;

[0179] Power module: Ensure that all modules have a stable power supply.

[0180] It should be noted that the above explanation of the embodiment of the supporting action monitoring method of the hydraulic support is also applicable to the supporting action monitoring system of the hydraulic support of this embodiment, and will not be repeated here.

[0181] In summary, the system provided by the embodiments of the present disclosure can combine existing data and data analysis to predict the movement of the support, and can provide more accurate and real-time support movement prediction, thereby improving support efficiency, optimizing support maintenance and enhancing support safety, and reducing the probability of accident risks.

[0182] In order to implement the above-mentioned embodiment, the present disclosure also proposes a supporting action monitoring device of a hydraulic support.

[0183] like Figure 3 As shown, the support action monitoring device 300 of the hydraulic support includes:

[0184] The data collection unit 301 is used to collect various support data of the hydraulic support and determine the correlation relationship between the various support data;

[0185] A network construction unit 302 is used to construct a complex network by using various scaffold data as network nodes and association relationships as connecting edges;

[0186] The action prediction unit 303 is used to perform dynamic trend analysis on the hydraulic support by using a complex network to obtain the action trend of the hydraulic support.

[0187] Optionally, the support data includes at least one of the following:

[0188] Bracket vibration frequency;

[0189] The vibration amplitude of the bracket;

[0190] Friction between the support and the coal seam;

[0191] The amount of electrostatic charge on the surface of the bracket;

[0192] Bracket load;

[0193] Bracket displacement;

[0194] Bracket inclination angle.

[0195] Optionally, when the data acquisition unit 301 is used to determine the association relationship between multiple types of bracket data, it is specifically used to:

[0196] Classify and label various scaffold data according to data types to obtain labeled scaffold data;

[0197] Segmenting the marked stent data according to a time window of a preset length to obtain multiple data segments;

[0198] Extracting features from each of the multiple data segments, and performing dimensionality reduction processing on the extracted features to obtain dimensionality reduction features corresponding to each data segment;

[0199] Different types of dimensionality reduction features are fused and analyzed to obtain the correlation between various scaffold data.

[0200] Optionally, the data acquisition unit 301 is further used for:

[0201] Analyze different types of dimensionality reduction features to obtain analysis results corresponding to each type of dimensionality reduction features, and perform status assessment on the hydraulic support based on the analysis results to obtain status assessment results;

[0202] If the analysis result meets the optimization requirements, the action of the hydraulic support is controlled to optimize the support data corresponding to the analysis result.

[0203] Optionally, after collecting various support data of the hydraulic support, the data collection unit 301 is further used to:

[0204] Perform outlier detection on various stent data;

[0205] If there are abnormal values ​​in the data of multiple stents, the abnormal values ​​will be corrected.

[0206] Optionally, the data acquisition unit 301 may use at least one of the following detection methods when performing outlier detection on multiple stent data:

[0207] Outlier detection based on statistical models;

[0208] Outlier detection based on cluster analysis;

[0209] Outlier detection based on machine learning models;

[0210] Outlier detection based on time series trend analysis;

[0211] Outlier detection based on cross-validation of multi-source data.

[0212] Optionally, when the action prediction unit 303 is used to perform dynamic trend analysis on the hydraulic support using a complex network, it is specifically used to:

[0213] Determine the risk indicators corresponding to complex networks;

[0214] Predict the risk value of each node in a complex network based on risk indicators and correlation relationships;

[0215] The risk propagation simulation is carried out according to the complex network and risk value to obtain the action trend of the hydraulic support.

[0216] Optionally, after obtaining the action trend of the hydraulic support, the action prediction unit 303 is further used to:

[0217] Analyze the action trend and obtain risk action information;

[0218] If the risk action information meets the early warning requirements, a warning signal will be issued.

[0219] It should be noted that the above explanation of the embodiment of the method for monitoring the supporting action of a hydraulic support is also applicable to the device for monitoring the supporting action of a hydraulic support of this embodiment, and will not be repeated here.

[0220] In summary, the device provided by the embodiment of the present disclosure can combine existing data and data analysis to predict the movement of the bracket, and can provide more accurate and real-time bracket movement prediction, thereby improving support efficiency, optimizing bracket maintenance and enhancing support safety, and reducing the probability of accident risks.

[0221] In order to implement the above embodiments, the present disclosure also proposes an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0222] In order to implement the above embodiments, the present disclosure further proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0223] In order to implement the above embodiments, the present disclosure also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0224] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this disclosure shall comply with the relevant laws and regulations and shall not violate public order and good morals.

[0225] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0226] The present disclosure anticipates providing implementation schemes for users to selectively block the use or access of personal information data. That is, the present disclosure anticipates providing hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, risks can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0227] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they contradict each other.

[0228] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0229] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.

[0230] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0231] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0232] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0233] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0234] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.

Claims

1. A method for monitoring the supporting action of a hydraulic support, characterized in that: include: Collect various support data of hydraulic supports, and determine the correlation between the various support data, wherein the various support data are classified and labeled according to data types to obtain labeled support data; the labeled support data are segmented according to a time window of a preset length to obtain multiple data segments; feature extraction is performed on each data segment of the multiple data segments, and the extracted features are subjected to dimensionality reduction processing to obtain dimensionality reduction features corresponding to each data segment; different types of dimensionality reduction features are fused and analyzed to obtain the correlation between the various support data; wherein the support data includes at least one of support vibration frequency, support vibration amplitude, friction between the support and the coal seam, electrostatic charge on the support surface, support load, support displacement, and support inclination; Using the multiple scaffold data as network nodes and the association relationships as connecting edges to construct a complex network; The complex network is used to perform dynamic trend analysis on the hydraulic support to obtain the action trend of the hydraulic support, wherein the risk index corresponding to the complex network is determined; the risk value of each node in the complex network is predicted according to the risk index and the association relationship; a risk propagation simulation is performed according to the complex network and the risk value to obtain the action trend of the hydraulic support; Analyzing the action trend to obtain risk action information; If the risk action information meets the warning requirements, a warning signal is issued.

2. The method according to claim 1, characterized in that The method further comprises: Analyzing the different types of dimensionality reduction features to obtain analysis results corresponding to each type of dimensionality reduction features, and performing a state assessment on the hydraulic support according to the analysis results to obtain a state assessment result; If the analysis result meets the optimization requirement, the action of the hydraulic support is controlled to optimize the support data corresponding to the analysis result.

3. The method according to claim 1, characterized in that After collecting the various support data of the hydraulic support, the method further includes: Performing outlier detection on the multiple stent data; If there are abnormal values ​​in the multiple types of bracket data, the abnormal values ​​are corrected.

4. The method according to claim 3, characterized in that When performing outlier detection on the multiple stent data, at least one of the following detection methods is used: Outlier detection based on statistical models; Outlier detection based on cluster analysis; Outlier detection based on machine learning models; Outlier detection based on time series trend analysis; Outlier detection based on cross-validation of multi-source data.

5. A support action monitoring device for a hydraulic support, characterized in that: include: A data acquisition unit is used to collect multiple types of support data of a hydraulic support and determine the correlation between the multiple types of support data, specifically used to classify and mark the multiple types of support data according to data types to obtain marked support data; segment the marked support data according to a time window of a preset length to obtain multiple data segments; perform feature extraction on each of the multiple data segments, and perform dimensionality reduction processing on the extracted features to obtain dimensionality reduction features corresponding to each data segment; perform fusion analysis on different types of dimensionality reduction features to obtain the correlation between the multiple types of support data; wherein the support data includes at least one of support vibration frequency, support vibration amplitude, friction between the support and the coal seam, electrostatic charge on the support surface, support load, support displacement, and support inclination; A network construction unit, used to construct a complex network by using the multiple scaffold data as network nodes and the association relationships as connecting edges; An action prediction unit is used to perform a dynamic trend analysis on the hydraulic support using the complex network to obtain the action trend of the hydraulic support, and is specifically used to determine the risk index corresponding to the complex network; predict the risk value of each node in the complex network according to the risk index and the association relationship; perform a risk propagation simulation according to the complex network and the risk value to obtain the action trend of the hydraulic support; The action prediction unit is further used to analyze the action trend to obtain risk action information; if the risk action information meets the warning requirements, a warning signal is issued.

6. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 4.

Images