A mine operation abnormality early warning system and method based on data analysis

Abstract

The present invention discloses a mine operation abnormality warning system and method based on data analysis, which belongs to the field of data monitoring technology. The present invention uses 5G / Wi‑Fi wireless transmission network to associate data with vital sign monitoring equipment and build an associated node network, and monitors and collects the vital sign status data of coal mine workers in real time; performs noise reduction and cleaning processing on the collected data, compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database, and builds a mapping curve analysis model; analyzes the real-time operation vital sign status data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time status data according to the analysis results; performs comprehensive warning on the operation area scene according to the judgment results; feedback outputs the real-time status data of each node in the scene, and performs operation behavior risk warning in combination with the comprehensive abnormal data of the scene.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and in particular to a mine operation abnormality early warning system and method based on data analysis. Background Art

[0002] The mine operation abnormality warning system is an advanced monitoring technology that can timely discover potential safety hazards and operation deviations by collecting and analyzing mine environment data, equipment data and operation behavior data in real time, thereby issuing early warning signals and guiding miners to take preventive measures to ensure the safety and efficiency of mine operations;

[0003] At present, the status monitoring of workers in mining environments is still in a passive monitoring state, which only provides emergency warnings through manual static observation of the workers' physiological signals or makes judgments by setting simple monitoring thresholds; however, this mode is often unable to conduct in-depth analysis and judgment on the real-time vital signs data of the workers, or there are rough judgments and high misjudgment rates, and there are delayed warnings, which makes it impossible to prevent accidents in advance. Summary of the invention

[0004] The purpose of the present invention is to provide a mine operation abnormality warning system and method based on data analysis to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A method for early warning of abnormal mine operations based on data analysis, the method comprising the following steps:

[0007] S100. Use 5G / Wi-Fi wireless transmission network to associate data of vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of coal mine workers in real time;

[0008] S200, performing noise reduction and cleaning processing on the collected data, comparing and analyzing the processed data with the corresponding conventional vital sign threshold data set in the database, and constructing a mapping curve analysis model;

[0009] S300, analyzing the real-time operation vital sign status data of each node according to the curve analysis model, and performing abnormal analysis and judgment on the real-time status data according to the analysis results; and performing comprehensive early warning on the operation area scene according to the judgment results;

[0010] S400: Feedback and output the real-time status data of each node in the scene, and conduct risk warning of operation behavior in combination with the comprehensive abnormal data of the scene.

[0011] The specific steps of the S100 using the 5G / Wi-Fi wireless transmission network to associate data with vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of coal mine workers in real time are as follows:

[0012] S101. By arranging a local supervision network in the mine operation area, the vital sign sensing equipment of the mine workers is connected through the local network to build an equipment data interconnection network; the vital sign sensing equipment is used to monitor the vital sign data of the workers when they are working in the mine; the scene back-end monitoring point is used as the Internet data integration point, and the vital sign equipment carried by each worker is used as the Internet node; the local supervision network is a 5G / Wi-Fi wireless transmission network;

[0013] S102, using the equipment data Internet network to monitor and collect real-time vital sign data of the operator, and integrating and transmitting the collected data to the back-end monitoring point via the Internet; the vital sign data includes blood pressure data, heart rate data and other physiological signal data.

[0014] The S200 performs noise reduction and cleaning processing on the collected data, and compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database. The specific steps of constructing the mapping curve analysis model are as follows:

[0015] S201, the back-end monitoring point performs corresponding node classification processing on the collected data, assigns numbers to the corresponding nodes through the Internet, and classifies the corresponding data according to the node numbers, obtains the collected data of the corresponding nodes, and constructs the corresponding information set Among them, M n is the data set corresponding to each node number, n∈[1,i], i is the number of nodes; is the 1st, 2nd, 3rd...ath type data of the corresponding node; a is the number of data types; the noise reduction and cleaning processing is performed on the data of each node after classification, and the processing formula is:

[0016]

[0017] Among them, X(f) is the frequency domain signal data obtained by Fourier transform of the original collected time domain signal data x(t); Y(f) is the frequency domain signal data output after noise reduction processing; h(f) is the response function; f1 and f2 are the filter frequency intervals; Y(t) is the time domain signal data output after noise reduction processing; F(.) and F -1 (.) are Fourier transform and inverse Fourier transform calculations respectively; X(f)·h(f) is the dot product calculation;

[0018] S202, compare and classify the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the historical conventional data of each node in the database, and comparing and integrating each type of data with the collected data, map the corresponding conventional data curve and the collected data curve to the same plane to construct a same-plane comparison hyperbola model, and obtain the comparison hyperbola model between the corresponding type of collected data and the conventional data.

[0019] The S300 analyzes the real-time operation vital sign status data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time status data according to the analysis results; the specific steps of performing comprehensive early warning on the operation area scene according to the judgment results are as follows:

[0020] S301, analyze the operation status of each node according to the curve model of each node; call the hyperbolic model constructed by each node corresponding to each type of data, and perform difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbolic model. The calculation formula is:

[0021]

[0022] Among them, Sr a is the synchronization rate between the control hyperbolas in the hyperbolic model corresponding to each type of data; T is the window period; p is the time delay; is the value of point t on the curve corresponding to each type of collected data; It is the average value within the corresponding curve period of each type of collected data; is the value of point t+p on the corresponding curve for each type of conventional data; The average value within the corresponding curve period of each type of conventional data; according to the synchronization rate analysis results of the hyperbola corresponding to each type of data, by changing the value of the time delay p, the corresponding curve of the collected data is translated in the hyperbola model, and the translation results of the two curves when the synchronization rate is the highest are output; by translating the hyperbola between each type of data with different time delays, the relative position of the hyperbola when the synchronization rate is the highest is obtained, and the difference between the two curves is obtained more intuitively; the synchronization rate refers to the overlap rate of the two curves obtained by changing the time delay;

[0023] S302, perform a difference area search on the hyperbola model after translation output with the highest synchronization rate; mark the corresponding curve part of the non-overlapping collected data on the hyperbola, and extract the marked curve result and the corresponding time period data; obtain the corresponding derivative curve vector of the extracted curve part through vector transformation and fusion; according to the corresponding time data, point the corresponding curve point of the previous time point between adjacent time points on the collected data curve to the next time point to construct the curve vector between adjacent time points on the curve; fuse the curve vectors between the corresponding non-overlapping curve parts in the continuous time corresponding to each adjacent time point; the vector fusion process is the summation calculation of adjacent vectors in the corresponding time period; analyze the degree of divergence of the hyperbola according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data, and the calculation formula is: Where Dk a It is the degree of divergence of the non-overlapping curve part and the corresponding time period data on the hyperbola corresponding to each type of hyperbola model; The curve vectors of the collected data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; Curve vectors corresponding to the conventional data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; is the time period corresponding to the non-overlapping curve part on the hyperbola model of each type of data; z is the number of time periods corresponding to the non-overlapping curve part on the hyperbola model of each type of data; according to the analysis results of the degree of separation of the two curves in the hyperbola model corresponding to each type of data, the abnormality analysis of the hyperbola of each type of data is performed, and the calculation formula is: Among them, Hm a To calculate the abnormality of the hyperbola corresponding to each type of data; is the value of the collected data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; is the value of the regular data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; if there is Hm a >Hm a (th), then it is judged that the data collected by the current corresponding node is abnormal; otherwise, the data collected by the corresponding node is normal; where Hm a (th) is the preset threshold of the system; wherein, when calculating the divergence of the hyperbola, the corresponding vector is obtained by deriving the non-coincident curve part, and the cosine value between the two vectors is analyzed and calculated, and the difference is analyzed by citing the principle of cosine similarity;

[0024] S303: Comprehensively analyze the data collected from abnormal nodes in the operation scene, and the calculation formula is: Where EN is the comprehensive abnormality rate of the operation scene; J[Hma >Hm a (th)] is the number of abnormal nodes in the scene; J all is the total number of nodes in the scene; if EN>EN(th), it is judged that the current operation scene has an environmental anomaly; otherwise, the current scene environment is normal; EN(th) is the system preset threshold.

[0025] The specific steps of S400 for feeding back and outputting the real-time status data of each node in the scene and performing risk warning of operation behavior in combination with the comprehensive abnormal data of the scene are as follows:

[0026] S401, providing data feedback for the data collected by each node in the scene, issuing warnings for abnormal data and outputting specific abnormal data;

[0027] S402: Output the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issue an early warning of operational behavior risks.

[0028] A mine operation abnormality warning system based on data analysis, the system comprises a state data acquisition module, a model building module, an abnormal state analysis module and a comprehensive feedback module;

[0029] The status data acquisition module uses the 5G / Wi-Fi wireless transmission network to associate the data of the vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of underground coal mine workers in real time; the model construction module performs noise reduction and cleaning processing on the collected data, compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database, and constructs a mapping curve analysis model; the abnormal state analysis module analyzes the real-time operation vital sign status data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time status data according to the analysis results; and performs comprehensive warning on the operation area scene according to the judgment results; the comprehensive feedback module feeds back and outputs the real-time status data of each node in the scene, and performs operation behavior risk warning in combination with the comprehensive abnormal data of the scene.

[0030] The status data acquisition module includes an equipment network construction unit and a vital sign data acquisition unit; the equipment network construction unit arranges a local supervision network in the mine operation area, connects the vital sign sensing equipment of the mine workers through the local network, and builds an equipment data Internet network; the scene back-end monitoring point is used as the Internet data integration point, and the vital sign equipment carried by each worker is used as the Internet node; the local supervision network is a 5G / Wi-Fi wireless transmission network; the vital sign data acquisition unit uses the equipment data Internet network to monitor the real-time vital sign data of the workers and collect data, and integrates and transmits the collected data to the back-end monitoring point via the Internet.

[0031] The model building module includes a data noise reduction processing unit and a hyperbolic model building unit; the back-end monitoring point of the data noise reduction processing unit performs corresponding node classification processing on the collected data, assigns numbers to the corresponding nodes through the Internet, and performs corresponding data classification for each node number, obtains the collected data of the corresponding node, and constructs a corresponding information set; performs noise reduction and cleaning processing on the classified data of each node; the hyperbolic model building unit compares and classifies the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the historical conventional data of each node in the database, and comparing and integrating each type of data with the collected data, the corresponding conventional data curves and the collected data curves are mapped to the same plane to construct a same-plane comparison hyperbolic model, and the comparison hyperbolic model between the corresponding types of collected data and conventional data is obtained.

[0032] The abnormal state analysis module includes a curve synchronization rate analysis unit, an abnormal state analysis unit and a scene comprehensive analysis unit; the curve synchronization rate analysis unit performs operation state analysis on each node according to the curve model of each node; retrieves the hyperbola model constructed by each node corresponding to each type of data, and performs difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbola model; outputs the translation result of the two curves when the synchronization rate is the highest according to the synchronization rate analysis result of the hyperbola corresponding to each type of data; the abnormal state analysis unit performs difference area search on the hyperbola translation curve model output with the highest synchronization rate; marks the curve part corresponding to the non-overlapping collected data on the hyperbola, and extracts the marked curve result and the corresponding time period data; extracts the extracted curve The line part obtains the corresponding derivative curve vector through vector transformation and fusion; according to the corresponding time data, the curve vector between adjacent time points on the collected data curve is constructed by pointing the corresponding curve point of the previous time point between adjacent time points on the curve to the next time point; the curve vectors between the corresponding adjacent time points of the non-overlapping curve part in the continuous time are fused; the degree of divergence of the hyperbola is analyzed according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data; according to the analysis results of the degree of divergence of the two curves in the hyperbola model corresponding to each type of data, the abnormality of the hyperbola of each type of data is analyzed; according to the analysis results of the degree of divergence of the two curves in the hyperbola model corresponding to each type of data, the abnormality of the hyperbola of each type of data is judged; the data collected at the current corresponding node is judged to be abnormal according to the analysis results; the scene comprehensive analysis unit performs comprehensive analysis on the data collected at the abnormal node in the operation scene.

[0033] The comprehensive feedback module includes an abnormal sign feedback unit and a scene comprehensive feedback unit; the abnormal sign feedback unit provides data feedback on the data collected by each node in the scene, warns of the abnormal data and outputs the specific abnormal data; the scene comprehensive feedback unit outputs the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issues an early warning of the risk of operational behavior.

[0034] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention utilizes front-end network data acquisition and back-end data analysis to construct a relevant curve analysis model to realize status monitoring and abnormal warning of the actual mine working environment and operators; by constructing an equipment association network in the scene, the real-time status of the operators is monitored, and quantitative analysis is performed by noise reduction processing and curve model conversion of the monitoring data; by performing curve comparison analysis on the real-time status data and conventional data, the synchronization rate, degree of separation and abnormality of the comparison curve are analyzed respectively to perform abnormal analysis on the operators; by analyzing the conditions of abnormal operators in the scene, the scene environment is analyzed abnormally; the present invention realizes correlation analysis and deep mining of the real-time status data of the operators in the actual scene, realizes accurate positioning and timely warning of abnormal personnel, improves the accuracy and efficiency of abnormal monitoring in mines, and reduces the probability of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0036] Figure 1 It is a structural schematic diagram of a mine operation abnormality early warning system based on data analysis of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0038] See also Figure 1 , the present invention provides a technical solution:

[0039] A method for early warning of abnormal mine operations based on data analysis, the method comprising the following steps:

[0040] S100. Use 5G / Wi-Fi wireless transmission network to associate data of vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of coal mine workers in real time;

[0041] S200, performing noise reduction and cleaning processing on the collected data, comparing and analyzing the processed data with the corresponding conventional vital sign threshold data set in the database, and constructing a mapping curve analysis model;

[0042] S300, analyzing the real-time operation vital sign status data of each node according to the curve analysis model, and performing abnormal analysis and judgment on the real-time status data according to the analysis results; and performing comprehensive early warning on the operation area scene according to the judgment results;

[0043] S400: Feedback and output the real-time status data of each node in the scene, and conduct risk warning of operation behavior in combination with the comprehensive abnormal data of the scene.

[0044] The specific steps of the S100 using the 5G / Wi-Fi wireless transmission network to associate data with vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of coal mine workers in real time are as follows:

[0045] S101. By arranging a local supervision network in the mine operation area, the vital sign sensing devices of the mine workers are connected through the local network to build an equipment data Internet network; the scene back-end monitoring point is used as the Internet network data integration point, and the vital sign devices carried by each worker are used as Internet network nodes; the local supervision network is a 5G / Wi-Fi wireless transmission network;

[0046] S102, using the equipment data Internet network to monitor and collect real-time vital sign data of operators, and integrating and transmitting the collected data to the back-end monitoring point via the Internet.

[0047] The S200 performs noise reduction and cleaning processing on the collected data, and compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database. The specific steps of constructing the mapping curve analysis model are as follows:

[0048] S201, the back-end monitoring point performs corresponding node classification processing on the collected data, assigns numbers to the corresponding nodes through the Internet, and classifies the corresponding data according to the node numbers, obtains the collected data of the corresponding nodes, and constructs the corresponding information set Among them, M n is the data set corresponding to each node number, n∈[1,i], i is the number of nodes; is the 1st, 2nd, 3rd...ath type data of the corresponding node; a is the number of data types; the noise reduction and cleaning processing is performed on the data of each node after classification, and the processing formula is:

[0049]

[0050] Among them, X(f) is the frequency domain signal data obtained by Fourier transform of the original collected time domain signal data x(t); Y(f) is the frequency domain signal data output after noise reduction processing; h(f) is the response function; f1 and f2 are the filter frequency intervals; Y(t) is the time domain signal data output after noise reduction processing; F(.) and F -1(.) are Fourier transform and inverse Fourier transform calculations respectively; X(f)·h(f) is the dot product calculation;

[0051] S202, compare and classify the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the historical conventional data of each node in the database, and comparing and integrating each type of data with the collected data, map the corresponding conventional data curve and the collected data curve to the same plane to construct a same-plane comparison hyperbola model, and obtain the comparison hyperbola model between the corresponding type of collected data and the conventional data.

[0052] The S300 analyzes the real-time operation vital sign status data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time status data according to the analysis results; the specific steps of performing comprehensive early warning on the operation area scene according to the judgment results are as follows:

[0053] S301, analyze the operation status of each node according to the curve model of each node; call the hyperbolic model constructed by each node corresponding to each type of data, and perform difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbolic model. The calculation formula is:

[0054]

[0055] Among them, Sr a is the synchronization rate between the control hyperbolas in the hyperbolic model corresponding to each type of data; T is the window period; p is the time delay; is the value of point t on the curve corresponding to each type of collected data; It is the average value within the corresponding curve period of each type of collected data; is the value of the t+p point on the corresponding curve for each type of conventional data; The average value within the period of the corresponding curve of each type of conventional data; according to the synchronization rate analysis results of the hyperbola corresponding to each type of data, by changing the value of the time delay p, the corresponding curve of the collected data is shifted in the hyperbola model, and the shift results of the two curves when the synchronization rate is the highest are output;

[0056] S302, perform a difference area search on the hyperbola model after translation output with the highest synchronization rate; mark the corresponding curve part of the non-overlapping collected data on the hyperbola, and extract the marked curve result and the corresponding time period data; obtain the corresponding derivative curve vector of the extracted curve part through vector transformation and fusion; according to the corresponding time data, point the corresponding curve point of the previous time point between adjacent time points on the collected data curve to the next time point to construct the curve vector between adjacent time points on the curve; fuse the curve vectors between the corresponding non-overlapping curve parts in the continuous time corresponding to each adjacent time point; analyze the degree of divergence of the hyperbola according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data, and the calculation formula is: Where Dk a It is the degree of divergence of the non-overlapping curve part and the corresponding time period data on the hyperbola corresponding to each type of hyperbola model; The curve vectors of the collected data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; Curve vectors corresponding to the conventional data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; is the time period corresponding to the non-overlapping curve part on the hyperbola model of each type of data; z is the number of time periods corresponding to the non-overlapping curve part on the hyperbola model of each type of data; according to the analysis results of the degree of separation of the two curves in the hyperbola model corresponding to each type of data, the abnormality analysis of the hyperbola of each type of data is performed, and the calculation formula is:

[0057]

[0058] Among them, Hm a To calculate the abnormality of the hyperbola corresponding to each type of data; is the value of the collected data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; is the value of the regular data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; if there is Hm a >Hm a (th), then it is judged that the data collected by the current corresponding node is abnormal; otherwise, the data collected by the corresponding node is normal; where Hm a (th) is the system preset threshold;

[0059] S303: Comprehensively analyze the data collected from abnormal nodes in the operation scene, and the calculation formula is: Where EN is the comprehensive abnormality rate of the operation scene; J[Hm a >Hm a (th)] is the number of abnormal nodes in the scene; Jall is the total number of nodes in the scene; if EN>EN(th), it is judged that the current operation scene has an environmental anomaly; otherwise, the current scene environment is normal; EN(th) is the system preset threshold.

[0060] The specific steps of S400 for feeding back and outputting the real-time status data of each node in the scene and performing risk warning of operation behavior in combination with the comprehensive abnormal data of the scene are as follows:

[0061] S401, each node in the scene collects data for data feedback, warns of abnormal data and outputs specific abnormal data;

[0062] S402: Output the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issue an operational behavior risk warning.

[0063] A mine operation abnormality warning system based on data analysis, the system comprises a state data acquisition module, a model building module, an abnormal state analysis module and a comprehensive feedback module;

[0064] The status data acquisition module uses the 5G / Wi-Fi wireless transmission network to associate the data of the vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of underground coal mine workers in real time; the model construction module performs noise reduction and cleaning processing on the collected data, compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database, and constructs a mapping curve analysis model; the abnormal state analysis module analyzes the real-time operation vital sign status data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time status data according to the analysis results; and performs comprehensive warning on the operation area scene according to the judgment results; the comprehensive feedback module feeds back and outputs the real-time status data of each node in the scene, and performs operation behavior risk warning in combination with the comprehensive abnormal data of the scene.

[0065] The status data acquisition module includes an equipment network construction unit and a vital sign data acquisition unit; the equipment network construction unit arranges a local supervision network in the mine operation area, connects the vital sign sensing equipment of the mine workers through the local network, and builds an equipment data Internet network; the scene back-end monitoring point is used as the Internet data integration point, and the vital sign equipment carried by each worker is used as the Internet node; the local supervision network is a 5G / Wi-Fi wireless transmission network; the vital sign data acquisition unit uses the equipment data Internet network to monitor the real-time vital sign data of the workers and collect data, and integrates and transmits the collected data to the back-end monitoring point via the Internet.

[0066] The model building module includes a data noise reduction processing unit and a hyperbolic model building unit; the back-end monitoring point of the data noise reduction processing unit performs corresponding node classification processing on the collected data, assigns numbers to the corresponding nodes through the Internet, and performs corresponding data classification for each node number, obtains the collected data of the corresponding node, and constructs a corresponding information set; performs noise reduction and cleaning processing on the classified data of each node; the hyperbolic model building unit compares and classifies the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the historical conventional data of each node in the database, and comparing and integrating each type of data with the collected data, the corresponding conventional data curves and the collected data curves are mapped to the same plane to construct a same-plane comparison hyperbolic model, and the comparison hyperbolic model between the corresponding types of collected data and conventional data is obtained.

[0067] The abnormal state analysis module includes a curve synchronization rate analysis unit, an abnormal state analysis unit and a scene comprehensive analysis unit; the curve synchronization rate analysis unit performs operation state analysis on each node according to the curve model of each node; retrieves the hyperbola model constructed by each node corresponding to each type of data, and performs difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbola model; outputs the translation result of the two curves when the synchronization rate is the highest according to the synchronization rate analysis result of the hyperbola corresponding to each type of data; the abnormal state analysis unit performs difference area search on the hyperbola translation curve model output with the highest synchronization rate; marks the curve part corresponding to the non-overlapping collected data on the hyperbola, and extracts the marked curve result and the corresponding time period data; extracts the extracted curve The line part obtains the corresponding derivative curve vector through vector transformation and fusion; according to the corresponding time data, the curve vector between adjacent time points on the collected data curve is constructed by pointing the corresponding curve point of the previous time point between adjacent time points on the curve to the next time point; the curve vectors between the corresponding adjacent time points of the non-overlapping curve part in the continuous time are fused; the degree of divergence of the hyperbola is analyzed according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data; according to the analysis results of the degree of divergence of the two curves in the hyperbola model corresponding to each type of data, the abnormality of the hyperbola of each type of data is analyzed; according to the analysis results of the degree of divergence of the two curves in the hyperbola model corresponding to each type of data, the abnormality of the hyperbola of each type of data is judged; the data collected at the current corresponding node is judged to be abnormal according to the analysis results; the scene comprehensive analysis unit performs comprehensive analysis on the data collected at the abnormal node in the operation scene.

[0068] The comprehensive feedback module includes an abnormal sign feedback unit and a scene comprehensive feedback unit; the abnormal sign feedback unit provides data feedback on the data collected by each node in the scene, warns of abnormal data and outputs specific abnormal data; the scene comprehensive feedback unit outputs the results of the comprehensive abnormal analysis data in the scene to the monitoring management port, and issues an early warning of the risk of the operation behavior;

[0069] In the examples:

[0070] A coal mine underground operation site monitoring system is now equipped with the mine operation abnormality warning system based on data analysis of the present invention to monitor the status of the operators in the operation scene; it arranges a 5g local supervision network in the coal mine underground operation area, connects the vital sign sensing equipment of the mine operators through the local network, and builds an equipment data Internet network; the scene back-end monitoring point is used as the Internet data integration point, and the vital sign equipment carried by each operator is used as the Internet node; the real-time vital sign data of the operators is monitored and collected by using the equipment data Internet network, and the collected data is integrated and transmitted to the back-end monitoring point via the Internet;

[0071] The back-end monitoring point classifies the collected data according to the corresponding nodes, assigns numbers to the corresponding nodes through the Internet, classifies the corresponding data according to the node numbers, obtains the collected data of the corresponding nodes, and constructs the corresponding information set. The noise reduction and cleaning process is performed on the data of each node after classification. The processing formula is:

[0072]

[0073] Compare and classify the processed real-time collected data of each node with the corresponding historical conventional data; retrieve the historical conventional data of each node in the database, and compare and integrate each type of data with the collected data, map the corresponding conventional data curve and the collected data curve to the same plane to build a same-plane comparison hyperbola model, and obtain the comparison hyperbola model between the corresponding types of collected data and conventional data;

[0074] According to the curve model of each node, the operation status of each node is analyzed respectively; the hyperbolic model constructed by each node corresponding to each type of data is called, and the difference analysis of the current collected data is performed by analyzing the synchronization rate between the control curves in each hyperbolic model. The calculation formula is:

[0075]

[0076] According to the synchronization rate analysis results of the hyperbola corresponding to each type of data, by changing the value of the time delay p, the corresponding curve of the collected data is translated in the hyperbola model, and the translation results of the two curves when the synchronization rate is the highest are output; the curve model after the hyperbola translation output with the highest synchronization rate is searched for difference areas; the corresponding curve parts of the non-overlapping collected data are marked on the hyperbola, and the marked curve results and the corresponding time period data are extracted; the corresponding derivative curve vectors are obtained by vector transformation and fusion of the extracted curve parts; according to the corresponding time data, the curve vectors between adjacent time points on the collected data curve are constructed by pointing to the next time point with the corresponding curve point of the previous time point between adjacent time points on the curve; the curve vectors between the corresponding non-overlapping curve parts in the continuous time are fused; the degree of separation of the hyperbola is analyzed according to the non-overlapping curve parts on the corresponding hyperbola model and the corresponding time period data, and the calculation formula is: According to the analysis results of the degree of separation between the two curves in the hyperbolic model corresponding to each type of data, the abnormality analysis of the hyperbolic curve of each type of data is performed, and the calculation formula is:

[0077] If there is Hm a >Hm a (th), then it is judged that the current corresponding node collection data is abnormal; otherwise, the corresponding node collection data is normal; a comprehensive analysis is performed on the abnormal node collection data in the operation scene, and its calculation formula is: If EN>EN(th), it is judged that the current operation scene has an abnormal environment; otherwise, the current scene environment is normal;

[0078] Provide data feedback for the data collected by each node in the scene, issue warnings for abnormal data and output specific abnormal data; output the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issue risk warnings for operational behavior.

[0079] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0080] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for warning abnormal mine operations based on data analysis, characterized in that: The method comprises the following steps: S100. Use 5G / Wi-Fi wireless transmission network to associate data of vital sign monitoring equipment and build an associated node network to monitor and collect vital sign status data of coal mine workers in real time; S200, performing noise reduction and cleaning processing on the collected data, comparing and analyzing the processed data with the corresponding conventional vital sign threshold data set in the database, and constructing a mapping curve analysis model; S300, analyzing the real-time operation vital sign status data of each node according to the curve analysis model, and performing abnormal analysis and judgment on the real-time status data according to the analysis results; and performing comprehensive early warning on the operation area scene according to the judgment results; The specific steps of S300 are as follows: S301, analyze the operation status of each node according to the curve model of each node; call the hyperbolic model constructed by each node corresponding to each type of data, and perform difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbolic model. The calculation formula is: Among them, Sr a is the synchronization rate between the control hyperbolas in the hyperbolic model corresponding to each type of data; T is the window period; p is the time delay; is the value of point t on the curve corresponding to each type of collected data; It is the average value within the corresponding curve period of each type of collected data; is the value of the t+p point on the corresponding curve for each type of conventional data; The average value within the corresponding curve period of each type of conventional data; according to the synchronization rate analysis results of the hyperbola corresponding to each type of data, by changing the value of the time delay p, the corresponding curve of the collected data is shifted in the hyperbola model, and the shift results of the two curves when the synchronization rate is the highest are output; S302, perform a difference area search on the hyperbola model after translation output with the highest synchronization rate; mark the corresponding curve part of the non-overlapping collected data on the hyperbola, and extract the marked curve result and the corresponding time period data; obtain the corresponding derivative curve vector of the extracted curve part through vector transformation and fusion; according to the corresponding time data, point the corresponding curve point of the previous time point between adjacent time points on the collected data curve to the next time point to construct the curve vector between adjacent time points on the curve; fuse the curve vectors between the corresponding non-overlapping curve parts in the continuous time corresponding to each adjacent time point; analyze the degree of divergence of the hyperbola according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data, and the calculation formula is: Where Dk a It is the degree of divergence of the non-overlapping curve part and the corresponding time period data on the hyperbola corresponding to each type of hyperbola model; The curve vectors of the collected data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; Curve vectors corresponding to the conventional data curves corresponding to the non-overlapping curve parts in each non-overlapping time period on the hyperbola model of each type of data; is the time period corresponding to the non-overlapping curve part on the hyperbola model of each type of data; z is the number of time periods corresponding to the non-overlapping curve part on the hyperbola model of each type of data; according to the analysis results of the degree of separation of the two curves in the hyperbola model corresponding to each type of data, the abnormality analysis of the hyperbola of each type of data is performed, and the calculation formula is: Among them, Hm a To calculate the abnormality of the hyperbola corresponding to each type of data; is the value of the collected data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; is the value of the regular data curve corresponding to the non-overlapping curve in each non-overlapping time period in the hyperbola model corresponding to each type of data; if there is Hm a >Hm a (th), then it is judged that the data collected by the current corresponding node is abnormal; otherwise, the data collected by the corresponding node is normal; where Hm a (th) is the system preset threshold; S303: Comprehensively analyze the data collected from abnormal nodes in the operation scene, and the calculation formula is: Where EN is the comprehensive abnormality rate of the operation scene; J[Hm a >Hm a (th)] is the number of abnormal nodes in the scene; J all is the total number of nodes in the scene; if EN>EN(th), it is judged that the current operation scene has an abnormal environment; otherwise, the current scene environment is normal; EN(th) is the system preset threshold; S400: Feedback and output the real-time status data of each node in the scene, and conduct risk warning of operation behavior in combination with the comprehensive abnormal data of the scene.

2. The method for warning abnormal mine operations based on data analysis according to claim 1, characterized in that: The specific steps of the S100 using the 5G / Wi-Fi wireless transmission network to associate data with vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of coal mine workers in real time are as follows: S101. By arranging a local supervision network in the mine operation area, the vital sign sensing devices of the mine workers are connected through the local network to build an equipment data Internet network; the scene back-end monitoring point is used as the Internet network data integration point, and the vital sign devices carried by each worker are used as Internet network nodes; the local supervision network is a 5G / Wi-Fi wireless transmission network; S102, using the equipment data Internet network to monitor and collect real-time vital sign data of operators, and integrating and transmitting the collected data to the back-end monitoring point via the Internet.

3. The method for abnormal mine operation warning based on data analysis according to claim 2 is characterized in that: The S200 performs noise reduction and cleaning processing on the collected data, and compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database. The specific steps of constructing the mapping curve analysis model are as follows: S201, the back-end monitoring point performs corresponding node classification processing on the collected data, assigns numbers to the corresponding nodes through the Internet, and classifies the corresponding data according to the node numbers, obtains the collected data of the corresponding nodes, and constructs the corresponding information set Among them, M n is the data set corresponding to each node number, n∈[1,i], i is the number of nodes; is the 1st, 2nd, 3rd...ath type data of the corresponding node; a is the number of data types; the noise reduction and cleaning processing is performed on the data of each node after classification, and the processing formula is: Among them, X(f) is the frequency domain signal data obtained by Fourier transform of the original collected time domain signal data x(t); Y(f) is the frequency domain signal data output after noise reduction processing; h(f) is the response function; f1 and f2 are the filter frequency intervals; Y(t) is the time domain signal data output after noise reduction processing; F(.) and F -1 (.) are Fourier transform and inverse Fourier transform calculations respectively; X(f)·h(f) is the dot product calculation; S202, compare and classify the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the historical conventional data of each node in the database, and comparing and integrating each type of data with the collected data, map the corresponding conventional data curve and the collected data curve to the same plane to construct a same-plane comparison hyperbola model, and obtain the comparison hyperbola model between the corresponding type of collected data and the conventional data.

4. The method for early warning of abnormal mine operation based on data analysis according to claim 3 is characterized by: The specific steps of S400 for feeding back and outputting the real-time status data of each node in the scene and performing risk warning of operation behavior in combination with the comprehensive abnormal data of the scene are as follows: S401, providing data feedback for the data collected by each node in the scene, issuing warnings for abnormal data and outputting specific abnormal data; S402: Output the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issue an early warning of operational behavior risks.

5. A mine operation abnormality warning system based on data analysis, using a mine operation abnormality warning method based on data analysis as claimed in any one of claims 1 to 4, characterized in that: The system includes a state data acquisition module, a model building module, an abnormal state analysis module and a comprehensive feedback module; The status data acquisition module uses the 5G / Wi-Fi wireless transmission network to associate data with the vital sign monitoring equipment and build an associated node network to monitor and collect the vital sign status data of underground coal mine workers in real time; the model construction module performs noise reduction and cleaning processing on the collected data, compares and analyzes the processed data with the corresponding conventional vital sign threshold data set in the database, and builds a mapping curve analysis model; The abnormal state analysis module analyzes the real-time operation vital sign state data of each node according to the curve analysis model, and performs abnormal analysis and judgment on the real-time state data according to the analysis results; and performs comprehensive early warning on the operation area scene according to the judgment results; The comprehensive feedback module outputs feedback on the real-time status data of each node in the scene, and performs risk warning of operation behavior in combination with the comprehensive abnormal data of the scene.

6. The mine operation abnormality warning system based on data analysis according to claim 5 is characterized by: The status data acquisition module includes an equipment network construction unit and a vital sign data acquisition unit; the equipment network construction unit arranges a local supervision network in the mine operation area, connects the vital sign sensing equipment of the mine workers through the local network, and builds an equipment data Internet network; the scene back-end monitoring point is used as the Internet data integration point, and the vital sign equipment carried by each worker is used as the Internet node; the local supervision network is a 5G / Wi-Fi wireless transmission network; the vital sign data acquisition unit uses the equipment data Internet network to monitor the real-time vital sign data of the workers and collect data, and integrates and transmits the collected data to the back-end monitoring point via the Internet.

7. The mine operation abnormality warning system based on data analysis according to claim 6 is characterized by: The model building module includes a data noise reduction processing unit and a hyperbolic model building unit; the back-end monitoring point of the data noise reduction processing unit performs corresponding node classification processing on the collected data, assigns numbers to each node through the Internet, and performs corresponding data classification for each node number, obtains the collected data of the corresponding node, and constructs a corresponding information set; performs noise reduction and cleaning processing on the classified node data; The hyperbolic model construction unit compares and classifies the processed real-time collected data of each node with the corresponding historical conventional data; by retrieving the conventional data of each node in the database, and comparing and integrating each type of data with the collected data, the corresponding historical conventional data curves and the collected data curves are mapped to the same plane to construct a same-plane comparison hyperbolic model, and the comparison hyperbolic model between the corresponding types of collected data and the conventional data is obtained.

8. The mine operation abnormality warning system based on data analysis according to claim 7 is characterized by: The abnormal state analysis module includes a curve synchronization rate analysis unit, an abnormal state analysis unit and a scene comprehensive analysis unit; the curve synchronization rate analysis unit performs operation state analysis on each node according to the curve model of each node; retrieves the hyperbola model constructed by each node corresponding to each type of data, and performs difference analysis on the current collected data by analyzing the synchronization rate between the control curves in each hyperbola model; outputs the translation result of the two curves when the synchronization rate is the highest according to the synchronization rate analysis result of the hyperbola corresponding to each type of data; the abnormal state analysis unit performs difference area search on the curve model after the hyperbola translation output with the highest synchronization rate; marks the curve part corresponding to the non-overlapping collected data on the hyperbola, and compares the marked curve result with the corresponding hyperbola; Extract the data of the corresponding time period; obtain the corresponding derivative curve vector by vector transformation and fusion of the extracted curve part; construct the curve vector between adjacent time points on the curve by pointing the corresponding curve point of the previous time point between adjacent time points on the collected data curve to the next time point according to the corresponding time data; fuse the curve vectors between the corresponding adjacent time points for the corresponding non-overlapping curve part in the continuous time; analyze the degree of separation of the hyperbola according to the non-overlapping curve part on the corresponding hyperbola model and its corresponding time period data; analyze the abnormality of the hyperbola of each type of data according to the analysis results of the degree of separation of the two curves in the hyperbola model corresponding to each type of data; judge the abnormality of the collected data of the current corresponding node according to the analysis results; The scene comprehensive analysis unit performs comprehensive analysis on the data collected from abnormal nodes in the operation scene.

9. The mine operation abnormality warning system based on data analysis according to claim 8 is characterized by: The comprehensive feedback module includes an abnormal sign feedback unit and a scene comprehensive feedback unit; the abnormal sign feedback unit provides data feedback on the data collected by each node in the scene, warns of the abnormal data and outputs the specific abnormal data; the scene comprehensive feedback unit outputs the comprehensive abnormal analysis data results in the scene to the monitoring management port, and issues an early warning of the risk of operational behavior.

Images