Intelligent coal mining subsidence area monitoring system and method

Abstract

The invention relates to the technical field of coal mining subsidence area monitoring, and particularly discloses an intelligent coal mining subsidence area monitoring system and method, and the system comprises an InSAR system which is used for periodically obtaining deformation phase information of each area; the plurality of groups of micro-seismic sensors are used for monitoring underground vibration data in real time; the plurality of groups of pressure sensors are used for monitoring underground stress change data in real time; and the monitoring center is used for acquiring a collapse risk value of each area in real time according to the underground vibration data and the stress change data, judging an early warning value of each area according to the collapse risk value of each area and the historical deformation phase information, and determining an early warning strategy according to the early warning value. Results of multiple monitoring modes are integrated, the problem that the accuracy of the monitoring result is low can be solved, meanwhile, timeliness of the monitoring process is achieved, judgment can be conducted in time when the coal mining collapse risk occurs, and measures are taken to reduce casualties and property losses.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mining subsidence monitoring, and in particular to an intelligent coal mining subsidence monitoring system and method. Background Art

[0002] Coal mining collapse monitoring is an important part of coal mine safety production and mining area environmental protection. The main purpose is to use technical means to monitor surface subsidence, cracks, collapse and other geological disasters in real time, and to issue early warnings and take measures to reduce casualties and property losses. With the development of Internet of Things technology and satellite technology, the monitoring process of coal mining collapse is becoming more and more intelligent. Through the intelligent monitoring process of coal mining collapse risk areas, the timeliness and accuracy of coal collapse monitoring can be improved.

[0003] Existing monitoring methods for coal mining collapse areas mainly include remote sensing monitoring, ground monitoring and underground monitoring technologies. Among them, synthetic aperture radar interferometry (InSAR) uses satellite radar images to monitor surface deformation with an accuracy of up to millimeter level. It has a wide coverage area and does not require ground equipment. Traditional leveling also has high monitoring accuracy and can meet the needs of high-precision monitoring. Underground monitoring technologies include microseismic monitoring systems and strain sensors. The microseismic monitoring system captures microseismic signals generated by rock fractures by deploying an underground sensor network to analyze the stability of the goaf. The strain sensor monitors the stress changes of the surrounding rock in the goaf and predicts the risk of collapse.

[0004] Among the existing monitoring methods, different methods have corresponding advantages, but also corresponding shortcomings. Among them, the update cycle of synthetic aperture radar interferometry (InSAR) data depends on the satellite revisit time, so it cannot meet the needs of real-time monitoring; leveling measurement efficiency is low and is only suitable for small-scale, high-precision monitoring points; although microseismic monitoring systems and strain sensors can realize large-scale real-time monitoring processes, the monitoring accuracy is low when directly judging based on their results. Therefore, how to integrate various monitoring methods to improve the accuracy of coal mining collapse area monitoring is the fundamental problem to be solved by the present invention. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent coal mining subsidence monitoring system and method to solve the following technical problems:

[0006] How to integrate various monitoring methods to improve the accuracy of monitoring coal mining collapse areas.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] An intelligent coal mining subsidence monitoring system, comprising:

[0009] InSAR system, used to periodically obtain deformation phase information of each area;

[0010] Microseismic sensors are provided in several groups for real-time monitoring of underground vibration data;

[0011] Pressure sensors are provided in several groups for real-time monitoring of underground stress change data;

[0012] The monitoring center is used to obtain the collapse risk value of each area in real time based on the underground vibration data and stress change data, determine the warning value of each area based on the collapse risk value and historical deformation phase information of each area, and determine the warning strategy based on the warning value.

[0013] Furthermore, the process of obtaining the collapse risk value includes:

[0014] Perform noise reduction on the vibration data and stress change data from t0 to t, where t is the current time point, t0 is the historical time point, t-t0=△t, and △t is the preset time period;

[0015] Obtain the earthquake source location based on the arrival time difference positioning method and several sets of vibration data;

[0016] Extract features from the processed vibration data and stress change data, and normalize the extracted features;

[0017] The normalized features are combined into a multidimensional vector and input into the risk assessment model to obtain the collapse risk value;

[0018] The risk assessment model is obtained through multi-dimensional vector training based on historical data extraction.

[0019] Furthermore, the features extracted from the vibration data include:

[0020] The maximum vibration energy value mE is obtained through the following process:

[0021] By formula Calculate the maximum vibration energy value E, where i is the order of vibration events in the period t0 to t, A i (t) is the waveform amplitude of the ith vibration event, ti1 is the start time of the ith vibration event, and ti2 is the end time of the ith vibration event;

[0022] The instability coefficient R is obtained by:

[0023] By formula Calculate the instability coefficient R, where N is the number of vibration events during the period t0 to t, i = 1, 2, ..., N, tg i is the time difference between the time point of the ith vibration event and the current time point, and t0 is the preset fixed time difference;

[0024] The spatial aggregation coefficient Y is obtained by:

[0025] Get the coordinates of the location where the vibration event occurred during the period t0 to t, using the formula The spatial aggregation coefficient Y is calculated, where is the number of lines connecting the coordinates of all position points. d j is the length of the jth line, λ is the adjustment coefficient, when there is no risk fault zone, λ is zero, l i is the minimum distance between the coordinates of the location of the ith earthquake event and the nearest risk fault zone;

[0026] The risk fault zone is obtained based on historical data.

[0027] Furthermore, the process of obtaining the risk fault zone includes:

[0028] The location points of all vibration events in the historical data are obtained, and the DBSCAN algorithm is used to determine whether there are clusters. If so, a risk fault zone is fitted based on the clusters. If not, it is determined that there is no risk fault zone.

[0029] Furthermore, the features extracted from the stress data include:

[0030] The maximum stress change rate mF, the acquisition process includes:

[0031] Through the formula mF=m k {m{σ′ k (t)}}Calculate the maximum stress change rate mF, where k is the order of the pressure sensor in the area, σ′ k (t) is the pressure change rate curve of the kth group, m{σ′ k (t)} is the maximum value of the pressure change rate of the kth group, m k {m{σ′ k (t)}}m{σ′ k (t)};

[0032] The process of obtaining the stress accumulation △F includes:

[0033] By formula The stress accumulation △F is calculated; where Q is the number of pressure sensor groups in the area, k = 1, 2, ..., Q, m is the number of time points collected at uniform time intervals, x = 1, 2, ..., m, σ x,k is the stress value corresponding to the xth time point of the kth pressure sensor, σ 0,k is the initial stress value of the kth pressure sensor;

[0034] The stress fluctuation coefficient s, the acquisition process includes:

[0035] By formula The stress fluctuation coefficient s is calculated; where, is the mean stress value corresponding to the kth pressure sensor time point.

[0036] Through the above technical solution,

[0037] Furthermore, the process of obtaining the multidimensional vector includes:

[0038] Normalize the acquired features based on Min-Max normalization to obtain a multidimensional vector [mE, R, Y, mF, ΔF, s] T .

[0039] Furthermore, the process of obtaining the warning value includes:

[0040] The settlement coefficient of each area is obtained based on the historical deformation phase information of the area, and the settlement coefficient is weighted and summed with the collapse risk value to obtain the warning value of each area;

[0041] The calculation process of the sedimentation coefficient includes:

[0042] By the formula u=Δh / h0+τ*f(K T -K T-1 ) is used to calculate the settlement coefficient u; where Δh is the settlement detected in the previous cycle, h0 is the settlement error reference value, f(x) is the definition function, when x>0, f(x)=x, otherwise, f(x)=0, τ is a fixed coefficient, K T K is the overall slope of the settlement change curve that adopts the deformation phase information of the previous cycle, T It is the overall slope of the settlement change curve without adopting the deformation phase information of the previous cycle.

[0043] Furthermore, the process of determining the early warning strategy includes:

[0044] Compare the warning value with the corresponding threshold interval, and determine the corresponding prediction strategy according to the threshold interval where the warning value is located;

[0045] The results corresponding to the threshold interval include high risk, medium risk and low risk;

[0046] The early warning strategy corresponding to the high risk includes issuing an order to transfer personnel from the area;

[0047] The early warning strategy corresponding to the medium risk includes performing collapse risk detection on the area based on leveling;

[0048] The early warning strategy corresponding to the low risk includes maintaining the current monitoring strategy.

[0049] Intelligent coal mining subsidence monitoring method, including:

[0050] The deformation phase information of each area is periodically acquired through the InSAR system;

[0051] Real-time monitoring of underground vibration data through several groups of microseismic sensors;

[0052] Real-time monitoring of underground stress change data through several groups of pressure sensors;

[0053] The monitoring center obtains the collapse risk value of each area in real time based on the underground vibration data and stress change data, determines the warning value of each area based on the collapse risk value and historical deformation phase information of each area, and determines the warning strategy based on the warning value.

[0054] Beneficial effects of the present invention:

[0055] (1) The present invention uses a combination of microseismic sensors and pressure sensors to determine the collapse risk of each area, and simultaneously obtains an early warning value by integrating periodically acquired deformation phase information. The early warning strategy is determined by the size of the early warning value. Compared with the method of monitoring coal mining subsidence areas in a single manner, the results of integrating multiple monitoring methods can make up for the problem of low accuracy of monitoring results, while achieving the timeliness of the monitoring process. When the risk of coal mining collapse occurs, it can make timely judgments, issue early warnings, and take measures to reduce casualties and property losses. By extracting the characteristics of vibration data and stress change data, the risk of collected collapse can be judged based on the numerical value size, data change state, consistency of different data, and other characteristics in the data, thereby improving the accuracy of the judgment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further described below with reference to the accompanying drawings.

[0057] Figure 1 It is a logic block diagram of the intelligent coal mining subsidence monitoring system of the present invention.

[0058] Figure 2 It is a flow chart of the steps of the intelligent coal mining subsidence monitoring method of the present invention. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0060] In one embodiment, an intelligent coal mining subsidence monitoring system is provided. Figure 1 As shown, the system includes an InSAR system, several groups of microseismic sensors, several groups of pressure sensors and a monitoring center, wherein the InSAR system can periodically obtain deformation phase information of each area; the microseismic sensors are used to monitor underground vibration data in real time; the pressure sensors are used to monitor underground stress change data in real time; the monitoring center then obtains the collapse risk value of each area in real time based on the underground vibration data and stress change data, determines the warning value of each area based on the collapse risk value and historical deformation phase information of each area, and determines the warning strategy based on the warning value; in the above scheme, the collapse risk of each area is judged by integrating the microseismic sensors and pressure sensors, and the warning value is obtained by integrating the periodically obtained deformation phase information, and the warning strategy is determined based on the size of the warning value. Compared with the method of monitoring coal mining subsidence areas through a single method, the results of integrating multiple monitoring methods can compensate for the problem of low accuracy of monitoring results, and at the same time achieve timeliness of the monitoring process, so that timely judgment can be made when coal mining collapse risks occur, and timely warning can be taken to reduce casualties and property losses.

[0061] In one embodiment, a process for obtaining a collapse risk value is provided, which specifically includes: performing noise reduction processing on vibration data and stress change data in the time period t0 to t, wherein t is the current time point, t0 is the historical time point, t-t0=△t, △t is a preset time period, and the length of the preset time period is set according to the current regional geological state. When there is a certain collapse risk in the area, the preset time period data is lengthened to obtain more data for judgment. For vibration data noise reduction, wavelet transform and bandpass filtering are used, and the noise reduction processing of stress change data is performed by moving average filtering. The noise reduction processing can reduce interference; then, the earthquake source position of each vibration time is obtained based on the time difference positioning method and several groups of vibration data; after processing, the earthquake source position of each vibration time is obtained. The vibration data and stress change data are subjected to feature extraction, and the extracted features are normalized; the normalized features are merged into a multidimensional vector and input into the risk assessment model to obtain a collapse risk value; wherein, the risk assessment model is obtained by multidimensional vector training based on historical data extraction, and its benchmark model can be implemented by a random forest model, by collecting a large amount of coal mining area collapse data for processing, extracting a multidimensional vector, and obtaining it after training with a random forest model. The specific training process will not be repeated in this embodiment. By extracting the features of the vibration data and stress change data, the risk of the collected collapse can be judged according to the features such as the numerical value size, the data change state, and the consistency of different data in the data, thereby improving the accuracy of the judgment result.

[0062] In one embodiment, the features extracted from the vibration data include three items: the maximum vibration energy value mE, the instability coefficient R, and the spatial aggregation coefficient Y. The maximum vibration energy value mE is obtained by the formula Calculated, where i is the order of the vibration events during the period t0 to t, A i (t) is the waveform amplitude of the ith vibration event, ti1 is the start time of the ith vibration event, and ti2 is the end time of the ith vibration event; therefore, the maximum vibration energy value mE reflects the scale of the rock rupture caused by the vibration event during the period t0 to t. The higher the vibration energy value, the greater the risk; the instability coefficient R is obtained through the formula Calculated, where N is the number of vibration events during the period t0 to t, i = 1, 2, ..., N, tg i is the time difference between the time point of the ith vibration event and the current time point, and t0 is a preset fixed time difference. Therefore, when the number of vibration events in the period t0 to t is greater and the time points of the vibration events are more concentrated, the corresponding risk is higher, and then the instability state of the rock formation is judged by the instability coefficient R; In addition, the acquisition process of the spatial aggregation coefficient Y includes: obtaining the coordinates of the location point where the vibration event occurs in the period t0 to t, and using the formula The spatial aggregation coefficient Y is calculated, where is the number of lines connecting the coordinates of all position points. d j is the length of the jth line, λ is the adjustment coefficient, which is obtained by fitting the historical data. When there is no risk fault zone, the value of λ is zero, and l i is the minimum distance between the coordinates of the location point of the i-th vibration event and the nearest risk fault zone; therefore, through the size of the spatial aggregation coefficient Y, the spatial aggregation of the vibration event can be comprehensively judged according to the concentration of the location point coordinates of the vibration event and the coincidence with the risk fault zone. Therefore, through the extraction process of the maximum vibration energy value mE, the instability coefficient R and the spatial aggregation coefficient Y, the above characteristics can be used to make a more accurate judgment on the risk of coal mining subsidence.

[0063] In addition, the risk fault zone is obtained based on historical data. The process of obtaining it includes: first, obtaining the location points of all vibration events in the historical data, and judging whether there is a cluster based on the DBSCAN algorithm: if there is, it means that the location points of the vibration events are relatively concentrated. Obviously, the concentrated area indicates that the risk of fracture is higher. Therefore, the risk fault zone is fitted based on the cluster. The specific fitting process is based on the existing technology and will not be described in detail here. When there is no cluster, it means that the location points of the vibration events are not concentrated. Therefore, it is temporarily judged that there is no risk fault zone. It should be noted that when there is no risk fault zone, the spatial aggregation coefficient Y calculation formula The item is zero. Through the above-mentioned process of obtaining the risk fault zone, the risk judgment of the coordinates of the location point of the vibration event is realized when the risk fault zone exists.

[0064] In one embodiment, the features extracted from the stress data include the maximum stress change rate mF, the stress accumulation ΔF, and the stress fluctuation coefficient s. The maximum stress change rate mF is obtained by the formula mF=m k {m{σ′ k (t)}} is calculated, where k is the order of the pressure sensor in the area, σ′ k (t) is the pressure change rate curve of the kth group, m{σ′ k (t)} is the maximum pressure change rate of the kth group, m k {m{σ′ k (t)}}m{σ′ k (t)}; therefore, the degree of rock mass failure can be judged by the magnitude of the maximum stress change rate mF; the process of obtaining the stress accumulation △F is through the formula Calculated; where Q is the number of pressure sensor groups in the area, k = 1, 2, ..., Q, m is the number of time points collected at uniform time intervals, x = 1, 2, ..., m, σ x,k is the stress value corresponding to the xth time point of the kth pressure sensor, σ 0,k is the initial stress value of the kth pressure sensor; by comprehensively calculating the pressure accumulation data of all pressure sensor data in the region, the stress accumulation △F can reflect the overall cumulative effect of the region, and then judge the potential energy release risk; In addition, the stress fluctuation coefficient s is obtained through the formula Calculated, where is the mean stress value corresponding to the kth pressure sensor time point. The higher the complexity and disorder of the pressure fluctuation, the higher the collapse risk of the rock formation. The instability state of the rock formation is judged by the stress fluctuation coefficient s. Therefore, by obtaining the characteristics of the maximum stress change rate mF, the stress accumulation △F and the stress fluctuation coefficient s, the above characteristics can be used to make a more accurate judgment on the risk of coal mining subsidence.

[0065] In one embodiment, the process of obtaining the multidimensional vector includes: normalizing the obtained features based on Min-Max normalization. The normalization process performed by Min-Max normalization can eliminate the dimensional differences of different features. Then, during the machine training process, the obtained multidimensional vector [mE, R, Y, mF, ΔF, s] T , and then obtain the risk assessment model.

[0066] In one embodiment, the acquisition process of the warning value includes: first, obtaining the sedimentation coefficient of each area based on the historical deformation phase information of each area. Since the deformation phase information is periodically acquired data, the historical deformation phase information refers to the data before the current time point that is closer to the current time point. The sedimentation coefficient is calculated by the formula u=Δh / h0+τ*f(K T -K T-1 ) is calculated; where Δh is the amount of settlement detected in the previous cycle (the cycle closest to the current time point), h0 is the reference value of the settlement error, which is set according to the InSAR monitoring data in the empirical data, f(x) is the definition function, when x>0, f(x)=x, otherwise, f(x)=0, τ is a fixed coefficient, which is set according to the empirical data fitting and is used to adjust the adjustment dimension and weight, K T K is the overall slope of the settlement change curve that adopts the deformation phase information of the previous cycle, T The overall slope of the settlement change curve does not adopt the deformation phase information of the previous cycle. Therefore, the size of the settlement coefficient can be used to judge the settlement change data in the cycle closest to the current time point and the change status relative to the overall data. Therefore, the settlement coefficient and the collapse risk value are weighted and summed, and the corresponding weight is set according to empirical data. Therefore, the warning value of each area is obtained by combining the two results, and then the coal mining collapse area can be accurately monitored by the size of the warning value.

[0067] In one embodiment, the process of determining the early warning strategy includes: comparing the early warning value with the corresponding threshold interval, the threshold interval is set after fitting historical data, and different threshold intervals have corresponding prediction strategies. In this embodiment, the results corresponding to the threshold interval include high risk, medium risk and low risk; therefore, the corresponding prediction strategy is determined according to the threshold interval where the early warning value is located; when it is high risk, the corresponding early warning strategy is to issue an order to transfer personnel in the area, issue a timely warning and take measures to reduce casualties and property losses; when it is medium risk, the corresponding early warning strategy includes collapse risk detection of the area based on leveling. Since leveling has better detection accuracy, it can further judge the collapse risk of the area; when it is low risk, the corresponding early warning strategy includes maintaining the current monitoring strategy and monitoring and judging the collapse risk in real time according to the size of the early warning value.

[0068] In one embodiment, an intelligent coal mining subsidence monitoring method is provided. Figure 2As shown, the system includes: periodically acquiring deformation phase information of each region through an InSAR system; monitoring underground vibration data in real time through several groups of microseismic sensors; monitoring underground stress change data in real time through several groups of pressure sensors; acquiring the collapse risk value of each region in real time based on the underground vibration data and stress change data through a monitoring center, determining the warning value of each region based on the collapse risk value and historical deformation phase information of each region, and determining the warning strategy based on the warning value; this embodiment can compensate for the low accuracy of monitoring results by integrating the results of multiple monitoring methods, while achieving the timeliness of the monitoring process, enabling timely judgment when coal mining collapse risks occur, and enabling timely warning and taking measures to reduce casualties and property losses. The above is a detailed description of one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. Intelligent coal mining subsidence monitoring system, characterized by: The system comprises: InSAR system, used to periodically obtain deformation phase information of each area; Microseismic sensors are provided in several groups for real-time monitoring of underground vibration data; Pressure sensors are provided in several groups for real-time monitoring of underground stress change data; The monitoring center is used to obtain the collapse risk value of each area in real time based on the underground vibration data and stress change data, determine the warning value of each area based on the collapse risk value and historical deformation phase information of each area, and determine the warning strategy based on the warning value.

2. The intelligent coal mining subsidence monitoring system according to claim 1 is characterized in that: The process of obtaining the collapse risk value includes: Perform noise reduction on the vibration data and stress change data from t0 to t, where t is the current time point, t0 is the historical time point, t-t0=△t, and △t is the preset time period; Obtain the earthquake source location based on the arrival time difference positioning method and several sets of vibration data; Extract features from the processed vibration data and stress change data, and normalize the extracted features; The normalized features are combined into a multidimensional vector and input into the risk assessment model to obtain the collapse risk value; The risk assessment model is obtained through multi-dimensional vector training based on historical data extraction.

3. The intelligent coal mining subsidence monitoring system according to claim 2 is characterized in that: Features extracted from vibration data include: The maximum vibration energy value mE is obtained through the following process: By formula Calculate the maximum vibration energy value E, where i is the order of vibration events in the period t0 to t, A i (t) is the waveform amplitude of the ith vibration event, ti1 is the start time of the ith vibration event, and ti2 is the end time of the ith vibration event; The instability coefficient R is obtained by: By formula Calculate the instability coefficient R, where N is the number of vibration events during the period t0 to t, i = 1, 2, ..., N, tg i is the time difference between the time point of the ith vibration event and the current time point, and t0 is the preset fixed time difference; The spatial aggregation coefficient Y is obtained by: Get the coordinates of the location where the vibration event occurred during the period t0 to t, using the formula The spatial aggregation coefficient Y is calculated, where is the number of lines connecting the coordinates of all position points. d j is the length of the jth line, λ is the adjustment coefficient, when there is no risk fault zone, λ is zero, l i is the minimum distance between the coordinates of the location of the ith earthquake event and the nearest risk fault zone; The risk fault zone is obtained based on historical data.

4. The intelligent coal mining subsidence monitoring system according to claim 3 is characterized in that: The process of obtaining the risk fault zone includes: The location points of all vibration events in the historical data are obtained, and the DBSCAN algorithm is used to determine whether there are clusters. If so, a risk fault zone is fitted based on the clusters. If not, it is determined that there is no risk fault zone.

5. The intelligent coal mining subsidence monitoring system according to claim 3 is characterized in that: Features extracted from stress data include: The maximum stress change rate mF, the acquisition process includes: Through the formula mF=m k {m{σ′ k (t)}}Calculate the maximum stress change rate mF, where k is the order of the pressure sensor in the area, σ′ k (t) is the pressure change rate curve of the kth group, m{σ′ k (t)} is the maximum value of the pressure change rate of the kth group, m k {m{σ′ k (t)}}m{σ′ k (t)}; The process of obtaining the stress accumulation △F includes: By formula The stress accumulation △F is calculated; where Q is the number of pressure sensor groups in the area, k = 1, 2, ..., Q, m is the number of time points collected at uniform time intervals, x = 1, 2, ..., m, σ x,k is the stress value corresponding to the xth time point of the kth pressure sensor, σ 0,k is the initial stress value of the kth pressure sensor; The stress fluctuation coefficient s, the acquisition process includes: By formula The stress fluctuation coefficient s is calculated; where, is the mean stress value corresponding to the kth pressure sensor time point.

6. The intelligent coal mining subsidence monitoring system according to claim 5 is characterized in that: The process of obtaining a multidimensional vector includes: Normalize the acquired features based on Min-Max normalization to obtain a multidimensional vector [mE, R, Y, mF, ΔF, s] T .

7. The intelligent coal mining subsidence monitoring system according to claim 2 is characterized in that: The process of obtaining the warning value includes: The settlement coefficient of each area is obtained based on the historical deformation phase information of the area, and the settlement coefficient is weighted and summed with the collapse risk value to obtain the warning value of each area; The calculation process of the sedimentation coefficient includes: By the formula u=Δh / h0+τ*f(K T -K T-1 ) is used to calculate the settlement coefficient u; where Δh is the settlement detected in the previous cycle, h0 is the settlement error reference value, f(x) is the definition function, when x>0, f(x)=x, otherwise, f(x)=0, τ is a fixed coefficient, K T K is the overall slope of the settlement change curve that adopts the deformation phase information of the previous cycle, T It is the overall slope of the settlement change curve without adopting the deformation phase information of the previous cycle.

8. The intelligent coal mining subsidence monitoring system according to claim 7 is characterized in that: The process of determining the early warning strategy includes: Compare the warning value with the corresponding threshold interval, and determine the corresponding prediction strategy according to the threshold interval where the warning value is located; The results corresponding to the threshold interval include high risk, medium risk and low risk; The early warning strategy corresponding to the high risk includes issuing an order to transfer personnel from the area; The early warning strategy corresponding to the medium risk includes performing collapse risk detection on the area based on leveling; The early warning strategy corresponding to the low risk includes maintaining the current monitoring strategy.

9. Intelligent coal mining subsidence monitoring method, characterized in that: The method adopts the intelligent coal mining subsidence monitoring system according to any one of claims 1 to 8, comprising: The deformation phase information of each area is periodically acquired through the InSAR system; Real-time monitoring of underground vibration data through several groups of microseismic sensors; Real-time monitoring of underground stress change data through several groups of pressure sensors; The monitoring center obtains the collapse risk value of each area in real time based on the underground vibration data and stress change data, determines the warning value of each area based on the collapse risk value and historical deformation phase information of each area, and determines the warning strategy based on the warning value.

Images