Dangerous rock multi-parameter correlation early warning method and system based on air-space-ground comprehensive perception

Abstract

The invention provides a dangerous rock multi-parameter correlation early warning method and system based on air-sky-ground comprehensive perception, and belongs to the technical field of dangerous rock mass monitoring. An air-sky-ground-in integrated monitoring system is arranged in a dangerous rock mass monitoring area, various and multi-scale dangerous rock mass monitoring data are integrated, and dangerous rock mass instability precursor information is utilized to realize early warning of dangerous rock mass instability. A multi-parameter early warning system with data complementation, data fusion and data association is constructed, the problem that data are one-sided when single factor or single scale information is used for prediction is solved, data complementation and comprehensive early warning are achieved, meanwhile, a training database for monitoring the dangerous rock mass is generated based on multiple dangerous rock mass instability cases and correction data streams, and the early warning accuracy is improved. According to the method, the dangerous rock mass data flow is monitored, and the early warning function model containing multi-parameter information is constructed, so that the development trend of the monitored dangerous rock mass data flow and the stable state of the dangerous rock mass are evaluated and pre-judged, early warning indication is made according to the early warning level of the monitored dangerous rock mass, and the early warning accuracy and timeliness of dangerous rock mass collapse disasters are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of dangerous rock mass monitoring, and particularly to a multi-parameter correlation early warning method and system for dangerous rocks based on comprehensive sensing of space-air-ground-indoor Background Art

[0002] Activities such as crustal movement and open-pit mining are extremely likely to form high-steep rock masses. Under the combined action of various factors such as vibration, rain erosion, and external forces, adverse phenomena such as tensile cracks, local rock mass shedding, and rock layer structure damage will occur on the high-steep rock masses, thus forming dangerous rock masses, and ultimately triggering collapse disasters, posing a serious threat to human activities and production safety.

[0003] Due to the characteristics of complexity, randomness, and uncertainty in the deformation and failure of dangerous rock masses, it is extremely difficult to accurately predict them. The key to accurately predicting the deformation and evolution process of dangerous rock mass disasters and making accurate forecasts before disasters occur is to establish an intelligent early warning system for dangerous rock masses. However, most of the existing technologies rely on single dangerous rock mass data parameters, and most of them have problems of independent operation and inability to complement data. The setting of early warning thresholds often lacks scientific basis, resulting in the inability to make full use of monitoring data and limited early warning effects. Summary of the Invention

[0004] To solve the above problems in the prior art, the present invention provides a multi-parameter correlation early warning method and system for dangerous rocks based on comprehensive sensing of space-air-ground-indoor. The invention integrates a variety of multi-scale dangerous rock mass monitoring data by deploying a "space-air-ground-indoor" integrated monitoring system in the dangerous rock mass monitoring area, and uses the precursor information of the instability of the dangerous rock mass to construct a multi-parameter early warning system with data complementarity, data fusion, and data correlation, solving the problem of one-sided data when predicting by a single factor or single-scale information, realizing data complementarity and comprehensive early warning. At the same time, a training database for monitoring dangerous rock masses is generated based on multiple dangerous rock mass instability cases and corrected data streams, and an early warning function model containing multi-parameter information is constructed, so as to evaluate and predict the development trend of the monitored dangerous rock mass data stream and the stable state of the dangerous rock mass, and make an early warning indication according to the early warning level of the monitored dangerous rock mass, improving the early warning accuracy and timeliness of dangerous rock mass collapse disasters. To achieve the above object, the technical solution is as follows:

[0005] On the one hand, the present invention provides a multi-parameter correlation early warning method for dangerous rocks based on comprehensive sensing of space-air-ground-indoor, the method comprising:

[0006] S1. By sorting out and analyzing the precursor information of dangerous rock mass instability cases, and using the orthogonal design method, obtain the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters;

[0007] S2. Based on the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters, train through a machine learning algorithm to obtain a training database associated with multiple parameters.

[0008] S3. Obtain the image data of the dangerous rock mass in the monitoring area through a drone to obtain a three-dimensional model of the monitored dangerous rock mass.

[0009] S4. According to the three-dimensional model of the monitored dangerous rock mass, numerically simulate the instability process of the dangerous rock mass and extract the data information of the relevant parameters during the instability process to obtain a corrected data stream.

[0010] S5. According to the corrected data stream and the training database associated with multiple parameters, train and update again to obtain a training database for monitoring the dangerous rock mass.

[0011] S6. Install a monitoring system on the dangerous rock mass in the monitoring area to obtain the monitoring data information of the monitored dangerous rock mass.

[0012] S7. According to the monitoring data information of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, perform normalization and weighting processing to obtain a monitoring data stream of the monitored dangerous rock mass.

[0013] S8. According to the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the development trend and warning level of the monitored dangerous rock mass and give a warning indication.

[0014] Optionally, in S1, by sorting out and analyzing the precursor information of the dangerous rock mass instability cases and using the orthogonal design method, obtain the weight factors of the parameters related to the dangerous rock mass instability and the correlation coefficients between the parameters, including:

[0015] S11. According to the precursor information of the dangerous rock mass instability cases, extract the parameters related to the dangerous rock mass instability and perform normalization processing to obtain dimensionless parameters of various precursor information. The parameters related to the dangerous rock mass instability include: the change amount of the three-dimensional coordinates of the characteristic points, displacement, dip angle, acceleration, microseismic energy, and the sliding amount of the structural plane.

[0016] S12. According to the dimensionless parameters of various precursor information, use the orthogonal design method to obtain the weight factors of the parameters related to the dangerous rock mass instability.

[0017] S13. According to the dimensionless parameters and the weight factors of the parameters related to the dangerous rock mass instability, perform correlation analysis using the orthogonal design method to obtain the correlation coefficients between the parameters.

[0018] Optionally, in S3, obtaining the three-dimensional model of the monitored dangerous rock mass by obtaining the image data of the dangerous rock mass in the monitoring area through a drone includes:

[0019] S31. Deploy multiple image control points on the dangerous rock mass in the monitoring area, and use RTK positioning technology to obtain the three-dimensional coordinates of multiple such image control points;

[0020] S32. Use a drone to collect multi-angle images of the dangerous rock mass in the monitoring area to obtain two-dimensional image data of the monitored dangerous rock mass;

[0021] S33. According to the two-dimensional image data of the monitored dangerous rock mass and the three-dimensional coordinates of multiple such image control points, use SFM three-dimensional image reconstruction technology to obtain a three-dimensional model of the monitored dangerous rock mass.

[0022] Optionally, in S4, according to the three-dimensional model of the monitored dangerous rock mass, by numerically simulating the instability process of the dangerous rock mass and extracting the data information of relevant parameters during the instability process, a correction data stream is obtained, including:

[0023] S41. According to the three-dimensional model of the monitored dangerous rock mass, obtain a numerical model for finite element simulation of the monitored dangerous rock mass;

[0024] S42. According to the numerical model for finite element simulation of the monitored dangerous rock mass, use numerical simulation software to simulate the instability process of the monitored dangerous rock mass to obtain correction parameter information related to the instability of the monitored dangerous rock mass. The correction parameter information related to the instability of the monitored dangerous rock mass includes: the change in the three-dimensional coordinates of characteristic points related to the instability of the monitored dangerous rock mass, the displacement related to the instability of the monitored dangerous rock mass, the inclination angle related to the instability of the monitored dangerous rock mass, the acceleration related to the instability of the monitored dangerous rock mass, the microseismic energy related to the instability of the monitored dangerous rock mass, and the sliding amount of the structural plane related to the instability of the monitored dangerous rock mass;

[0025] S43. By sorting and analyzing the correction parameter information related to the instability of the monitored dangerous rock mass, use the orthogonal design method to obtain the weight factors of the correction parameters related to the instability of the monitored dangerous rock mass and the correlation coefficients between the correction parameters;

[0026] S44. According to the weight factors of the correction parameters related to the instability of the monitored dangerous rock mass and the correlation coefficients between the correction parameters, obtain a correction data stream.

[0027] Optionally, the monitoring system includes:

[0028] GNSS positioning equipment, used to realize the three-dimensional coordinate positioning of the characteristic points of the dangerous rock mass in the monitoring area at the airspace scale;

[0029] Drone and takeoff and landing platform, used to realize the construction of a three-dimensional model of the dangerous rock mass in the monitoring area and local photography at the celestial scale;

[0030] Surface monitoring instruments, used to realize the acquisition of mesoscopic information of the dangerous rock mass in the monitoring area at the regional scale, including displacement, inclination angle, acceleration, and microseismicity on the surface of the dangerous rock mass in the monitoring area;

[0031] A large flexible deformeter in the hole is used to realize the sliding perception of the internal structural plane of the dangerous rock mass in the monitoring area at the internal domain scale;

[0032] A data acquisition unit is used to collect data information.

[0033] Optionally, in S7, according to the monitoring data information of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, normalization and weighting processing are adopted to obtain the monitoring data stream of the monitored dangerous rock mass, including:

[0034] S71. According to the monitoring data information of the monitored dangerous rock mass, a set of monitoring parameters related to the instability of the monitored dangerous rock mass is obtained;

[0035] S72. According to the set of monitoring parameters related to the instability of the monitored dangerous rock mass, normalization processing is carried out to obtain a dimensionless set of monitoring parameters;

[0036] S73. According to the dimensionless set of monitoring parameters, through Gaussian filtering noise reduction and equalization processing, a processed dimensionless set of monitoring parameters is obtained;

[0037] S74. According to the processed dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, through the correlation coefficients in the training database of the monitored dangerous rock mass, the monitoring parameters that have not been collected and have error information are supplemented to obtain a corrected dimensionless set of monitoring parameters;

[0038] S75. According to the corrected dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, weighting is carried out according to the weight factors in the training database of the monitored dangerous rock mass to obtain the monitoring data stream of the monitored dangerous rock mass.

[0039] Optionally, in S74, according to the processed dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, through the correlation coefficients in the training database of the monitored dangerous rock mass, the monitoring parameters that have not been collected and have error information are supplemented to obtain a corrected dimensionless set of monitoring parameters, including:

[0040] S741. According to the training database of the monitored dangerous rock mass, the correlation coefficients between the various parameters related to the instability of the monitored dangerous rock mass are obtained;

[0041] S742. According to the correlation coefficients between the various parameters related to the instability of the monitored dangerous rock mass and the processed dimensionless set of monitoring parameters, the 3 correlation coefficients with the largest correlation coefficients of the monitoring parameters that have not been collected and have error information are selected to obtain supplementary correlation coefficients;

[0042] S743. According to the supplementary correlation coefficients and the processed dimensionless set of monitoring parameters, calculation and averaging are carried out to obtain supplementary parameters;

[0043] S744. Obtain a corrected dimensionless monitoring parameter set based on the supplementary parameter and the processed dimensionless monitoring parameter set.

[0044] Optionally, in S8, based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the development trend and warning level of the monitored dangerous rock mass, and give a warning indication, including:

[0045] S81. Based on the monitoring data stream of the monitored dangerous rock mass, after cumulative processing, obtain a multi-parameter bandwidth;

[0046] S82. Based on the multi-parameter bandwidth, use filtering processing to separate the interference phases in the multi-parameter bandwidth, and obtain the variation relationship between the multi-parameter bandwidth and time;

[0047] S83. Based on the training database of the monitored dangerous rock mass, obtain a bandwidth threshold;

[0048] S84. Based on the bandwidth threshold and the variation relationship between the multi-parameter bandwidth and time, obtain the development trend of the monitored dangerous rock mass. The development trend of the monitored dangerous rock mass includes: prediction of the imminent trend of the monitored dangerous rock mass, prediction of the short-term trend of the monitored dangerous rock mass, and prediction of the long-term trend of the monitored dangerous rock mass;

[0049] S85. Based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the warning level of the monitored dangerous rock mass, and give a warning indication.

[0050] Optionally, in S85, based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the warning level of the monitored dangerous rock mass, and give a warning indication, including:

[0051] S851. Based on the monitoring data stream of the monitored dangerous rock mass, obtain a three-dimensional coordinate change amount data stream, a displacement data stream, an inclination data stream, an acceleration data stream, a microseismic energy data stream, and a structural plane sliding data stream;

[0052] S852. Based on the training database of the monitored dangerous rock mass, obtain a three-dimensional coordinate change amount weight, a displacement weight, an inclination weight, an acceleration weight, a microseismic energy weight, and a structural plane sliding weight. The relationship between the three-dimensional coordinate change amount weight, the displacement weight, the inclination weight, the acceleration weight, the microseismic energy weight, and the structural plane sliding weight conforms to formula (1),

[0053] (1)

[0054] In the formula: is the three-dimensional coordinate change amount weight at the position of the i-th point, is the displacement weight at the position of the i-th point, is the tilt weight at the position of the i-th point, is the acceleration weight at the position of the i-th point, is the microseismic energy weight at the position of the i-th point, is the structural plane sliding weight at the position of the i-th point;

[0055] S853. According to the three-dimensional coordinate change data stream, the displacement data stream, the tilt data stream, the acceleration data stream, the microseismic energy data stream, the structural plane sliding data stream, the three-dimensional coordinate change weight, the displacement weight, the tilt weight, the acceleration weight, the microseismic energy weight and the structural plane sliding weight, a multi-parameter function model is obtained through formula (2),

[0056] (2)

[0057] In the formula: is the multi-parameter function model, is the three-dimensional coordinate change data stream at the position of the i-th point, , is the displacement data stream at the position of the i-th point, , is the tilt data stream at the position of the i-th point, , is the acceleration data stream at the position of the i-th point, , is the microseismic energy data stream at the position of the i-th point, , is the structural plane sliding data stream at the position of the i-th point, ;

[0058] S854. According to the multi-parameter function model, through the warning response rule, the warning level of the monitored dangerous rock mass is obtained. The warning response rule includes:

[0059] , the danger level is level IV, which is a blue warning,

[0060] , the danger level is level III, which is a yellow warning,

[0061] , the danger level is level II, which is an orange warning,

[0062] , the danger level is level I, which is a red warning;

[0063] When the danger level is level I to III, start the drone to conduct cruise shooting on the positions with the danger level of level I to III.

[0064] On the other hand, the present invention provides a multi-parameter correlation early warning system for dangerous rocks based on integrated sensing within the air, space, and ground. This system is applied to the multi-parameter correlation early warning method for dangerous rocks based on integrated sensing within the air, space, and ground. The system includes:

[0065] A data acquisition module, which is used to obtain the weight factors of parameters related to the instability of dangerous rock masses and the correlation coefficients between the parameters by sorting and analyzing the precursor information of dangerous rock mass instability cases and using the orthogonal design method;

[0066] A training module, which is used to obtain a training database with multi-parameter correlation through machine learning algorithm training according to the weight factors of parameters related to the instability of dangerous rock masses and the correlation coefficients between the parameters;

[0067] A model generation module, which is used to obtain the three-dimensional model of the monitored dangerous rock mass by using an unmanned aerial vehicle to acquire the image data of the dangerous rock mass in the monitoring area;

[0068] A correction module, which is used to obtain a corrected data stream by numerically simulating the instability process of the dangerous rock mass and extracting the data information of relevant parameters during the instability process according to the three-dimensional model of the monitored dangerous rock mass;

[0069] An update module, which is used to retrain and update according to the corrected data stream and the training database with multi-parameter correlation to obtain the training database of the monitored dangerous rock mass;

[0070] A monitoring module, which is used to deploy a monitoring system for the dangerous rock mass in the monitoring area to obtain the monitoring data information of the monitored dangerous rock mass;

[0071] A processing module, which is used to obtain the monitoring data stream of the monitored dangerous rock mass by using normalization and weighting processing according to the monitoring data information of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass;

[0072] An early warning module, which is used to obtain the development trend and early warning level of the monitored dangerous rock mass according to the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, and issue an early warning indication.

[0073] Compared with the prior art, the technical solution of the invention has at least the following beneficial effects:

[0074] On the one hand, the above solution integrates various multi-scale monitoring data of dangerous rock masses by deploying an "air-space-ground-internal" integrated monitoring system in the monitoring area of dangerous rock masses, and uses the precursor information of the instability of dangerous rock masses to construct a multi-parameter early warning system with data complementarity, data fusion, and data correlation, which solves the problem of one-sided data when predicting with single-factor or single-scale information, and realizes data complementarity and comprehensive early warning. On the other hand, a training database for monitoring dangerous rock masses is generated based on multiple cases of dangerous rock mass instability and corrective data streams, and an early warning function model containing multi-parameter information is constructed, so as to evaluate and predict the development trend of the monitoring dangerous rock mass data stream and the stability state of the dangerous rock mass, and give an early warning indication according to the early warning level of the monitored dangerous rock mass, improving the early warning accuracy and timeliness of dangerous rock mass collapse disasters. Description of the Drawings

[0075] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0076] Figure 1 It is a flowchart of an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0077] Figure 2 It is a flowchart of obtaining the weight factors of parameters related to the instability of dangerous rock masses and the correlation coefficients between parameters in an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0078] Figure 3 It is a flowchart of obtaining the three-dimensional model of the monitored dangerous rock mass in an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0079] Figure 4 It is a flowchart of obtaining the corrective data stream in an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0080] Figure 5 It is a flowchart of obtaining the monitoring data stream of the monitored dangerous rock mass in an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0081] Figure 6 It is a flowchart of obtaining the corrected dimensionless monitoring parameter set in an embodiment of the method for multi-parameter correlation early warning of dangerous rocks based on comprehensive air-space-ground-internal perception of the present invention;

[0082] Figure 7It is a flowchart for obtaining the development trend of a monitored dangerous rock mass in an embodiment of the method for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground of the present invention;

[0083] Figure 8 It is a flowchart for obtaining the early warning level of a monitored dangerous rock mass in an embodiment of the method for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground of the present invention;

[0084] Figure 9 It is a schematic diagram of the multi-parameter bandwidth and time variation obtained in an embodiment of the method for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground of the present invention;

[0085] Figure 10 It is a system block diagram of an embodiment of the system for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground of the present invention. Detailed implementation manners

[0086] The following describes the technical solutions in the present invention with reference to the accompanying drawings.

[0087] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two can be selected.

[0088] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0089] As Figure 1 shown in the flowchart of the embodiment of the method for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground of the present invention, the present invention provides a method for early warning of multi-parameter correlation of dangerous rocks, which is implemented by a system for early warning of multi-parameter correlation of dangerous rocks based on comprehensive perception within the air, space, and ground. The method includes:

[0090] S1. By sorting out and analyzing the precursor information of dangerous rock mass instability cases and using the orthogonal design method, obtain the weight factors of parameters related to dangerous rock mass instability and the correlation coefficients between the parameters;

[0091] Specifically, as Figure 2Flowchart of obtaining the weight factors of parameters related to the instability of dangerous rock masses and the correlation coefficients between parameters in the embodiment of the multi-parameter correlation early warning method for dangerous rock masses based on integrated space-air-ground perception of the present invention shown. In S1, by sorting out and analyzing the precursor information of dangerous rock mass instability cases and using the orthogonal design method, the weight factors of parameters related to dangerous rock mass instability and the correlation coefficients between parameters are obtained, including:

[0092] S11. According to the precursor information of dangerous rock mass instability cases, extract the parameters related to dangerous rock mass instability and perform normalization processing to obtain dimensionless parameters of various precursor information. The parameters related to dangerous rock mass instability include: the change amount of the three-dimensional coordinates of characteristic points, displacement, dip angle, acceleration, microseismic energy, and the sliding amount of structural planes;

[0093] S12. According to the dimensionless parameters of the various precursor information, use the orthogonal design method to obtain the weight factors of the parameters related to dangerous rock mass instability;

[0094] S13. According to the dimensionless parameters and the weight factors of the parameters related to dangerous rock mass instability, perform correlation analysis using the orthogonal design method to obtain the correlation coefficients between parameters.

[0095] Furthermore, the correlation coefficient calculation method is as shown in formula (3),

[0096] (3)

[0097] In the formula, is the correlation coefficient between parameter x and parameter y of the dangerous rock mass numbered m at time t, where t is a certain moment, is the data of parameter x at time t, is the data of parameter y at time t, is the average value of the data of parameter x of the dangerous rock mass numbered m at different times, is the average value of the data of parameter y of the dangerous rock mass numbered m at different times, where m is the number of the dangerous rock mass, is the number of time points of the dangerous rock mass.

[0098] S2. According to the weight factors of the parameters related to dangerous rock mass instability and the correlation coefficients between parameters, train through a machine learning algorithm to obtain a training database of multi-parameter correlation;

[0099] S3. Obtain the image data of the dangerous rock mass in the monitoring area through an unmanned aerial vehicle to obtain a three-dimensional model of the monitored dangerous rock mass;

[0100] Specifically, as Figure 3The flowchart of obtaining the 3D model of the monitored dangerous rock mass in the embodiment of the dangerous rock multi-parameter correlation early warning method based on integrated sensing within the air, space, and ground of the present invention as shown. In S3, the image data of the dangerous rock mass in the monitoring area is obtained by an unmanned aerial vehicle, and the 3D model of the monitored dangerous rock mass is obtained, including:

[0101] S31. Arrange multiple image control points on the dangerous rock mass in the monitoring area, and use RTK positioning technology to obtain the 3D coordinates of multiple such image control points;

[0102] S32. Collect multi-angle images of the dangerous rock mass in the monitoring area by an unmanned aerial vehicle to obtain the 2D image data of the monitored dangerous rock mass;

[0103] S33. According to the 2D image data of the monitored dangerous rock mass and the 3D coordinates of multiple such image control points, use SFM 3D image reconstruction technology to obtain the 3D model of the monitored dangerous rock mass.

[0104] S4. According to the 3D model of the monitored dangerous rock mass, numerically simulate the instability process of the dangerous rock mass and extract the data information of relevant parameters during the instability process to obtain the corrected data stream;

[0105] Specifically, as Figure 4 The flowchart of obtaining the corrected data stream in the embodiment of the dangerous rock multi-parameter correlation early warning method based on integrated sensing within the air, space, and ground of the present invention as shown. In S4, according to the 3D model of the monitored dangerous rock mass, numerically simulate the instability process of the dangerous rock mass and extract the data information of relevant parameters during the instability process to obtain the corrected data stream, including:

[0106] S41. According to the 3D model of the monitored dangerous rock mass, obtain the numerical model of the finite element simulation of the monitored dangerous rock mass;

[0107] S42. According to the numerical model of the finite element simulation of the monitored dangerous rock mass, use numerical simulation software to simulate the instability process of the monitored dangerous rock mass to obtain the corrected parameter information related to the instability of the monitored dangerous rock mass. The corrected parameter information related to the instability of the monitored dangerous rock mass includes: the change amount of the 3D coordinates of the characteristic points related to the instability of the monitored dangerous rock mass, the displacement related to the instability of the monitored dangerous rock mass, the inclination angle related to the instability of the monitored dangerous rock mass, the acceleration related to the instability of the monitored dangerous rock mass, the microseismic energy related to the instability of the monitored dangerous rock mass, and the sliding amount of the structural plane related to the instability of the monitored dangerous rock mass;

[0108] S43. By sorting and analyzing the corrected parameter information related to the instability of the monitored dangerous rock mass, use the orthogonal design method to obtain the weight factors of the corrected parameters related to the instability of the monitored dangerous rock mass and the correlation coefficients between the corrected parameters;

[0109] S44. According to the weight factors of the corrected parameters related to the instability of the monitored dangerous rock mass and the correlation coefficients between the corrected parameters, obtain the corrected data stream.

[0110] S5. Re-train and update according to the corrected data stream and the training database associated with multiple parameters to obtain a training database for monitoring dangerous rock masses.

[0111] S6. Install a monitoring system on the dangerous rock masses in the monitoring area to obtain monitoring data information of the dangerous rock masses.

[0112] Specifically, the monitoring system includes:

[0113] A GNSS positioning device for realizing three-dimensional coordinate positioning of characteristic points of dangerous rock masses in the monitoring area at the airspace scale.

[0114] An unmanned aerial vehicle and a take-off and landing platform for realizing three-dimensional model construction and local photography of dangerous rock masses in the monitoring area at the sky domain scale.

[0115] Surface monitoring instruments for obtaining mesoscopic information of dangerous rock masses in the monitoring area at the regional scale, including displacement, inclination, acceleration, and microseismicity on the surface of dangerous rock masses in the monitoring area.

[0116] A large flexible deformeter in the borehole for realizing the sliding perception of the internal structural plane of dangerous rock masses in the monitoring area at the inner domain scale.

[0117] A data acquisition unit for collecting data information.

[0118] Furthermore, select an open position of the dangerous rock mass in the monitoring area to install the GNSS positioning device, select an open position of the dangerous rock mass in the monitoring area to install the take-off and landing platform of the unmanned aerial vehicle, and deploy an unmanned aerial vehicle equipped with an RTK positioning module and a vision sensing device. Select the position with cracks in the dangerous rock mass in the monitoring area to install displacement meters, inclinometers, accelerometers, and microseismic monitors. Select the area with upper and lower inclined structural planes in the dangerous rock mass in the monitoring area to drill and install large flexible deformeters. Select a stable and open area of the dangerous rock mass in the monitoring area to install a data acquisition module, and the data acquisition module is connected to each monitoring instrument through a wireless transmission method to form a monitoring system.

[0119] S7. According to the monitoring data information of the monitored dangerous rock masses and the training database of the monitored dangerous rock masses, perform normalization and weighting processing to obtain a monitoring data stream of the monitored dangerous rock masses.

[0120] Specifically, as Figure 5 shown in the flowchart of obtaining the monitoring data stream of the dangerous rock mass in the embodiment of the multi-parameter correlation early warning method for dangerous rocks based on integrated air, space, ground, and internal perception of the present invention, in S7, according to the monitoring data information of the monitored dangerous rock masses and the training database of the monitored dangerous rock masses, perform normalization and weighting processing to obtain a monitoring data stream of the monitored dangerous rock masses, including:

[0121] S71. Obtain a set of monitoring parameters related to the instability of the monitored dangerous rock mass based on the monitoring data information of the monitored dangerous rock mass;

[0122] S72. Perform normalization processing on the set of monitoring parameters related to the instability of the monitored dangerous rock mass to obtain a dimensionless set of monitoring parameters;

[0123] Further, the data form of the parameters in the dimensionless set of monitoring parameters ranges from [0, 1].

[0124] S73. Through Gaussian filtering noise reduction and equalization processing on the dimensionless set of monitoring parameters, obtain a processed dimensionless set of monitoring parameters;

[0125] S74. According to the processed dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, supplement the monitoring parameters that have not been collected and have incorrect information through the correlation coefficients in the training database of the monitored dangerous rock mass to obtain a corrected dimensionless set of monitoring parameters;

[0126] S75. According to the corrected dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, perform weighting according to the weight factors in the training database of the monitored dangerous rock mass to obtain the monitoring data stream of the monitored dangerous rock mass.

[0127] Further, as Figure 6 shown in the flowchart of the embodiment of the method for early warning of multi-parameter correlation of dangerous rocks based on integrated perception in the air, space and ground of the present invention, in S74, according to the processed dimensionless set of monitoring parameters and the training database of the monitored dangerous rock mass, supplement the monitoring parameters that have not been collected and have incorrect information through the correlation coefficients in the training database of the monitored dangerous rock mass to obtain a corrected dimensionless set of monitoring parameters, including:

[0128] S741. Obtain the correlation coefficients between the parameters related to the instability of the monitored dangerous rock mass according to the training database of the monitored dangerous rock mass;

[0129] S742. According to the correlation coefficients between the parameters related to the instability of the monitored dangerous rock mass and the processed dimensionless set of monitoring parameters, select the 3 correlation coefficients with the largest correlation coefficients of the monitoring parameters that have not been collected and have incorrect information to obtain supplementary correlation coefficients;

[0130] S743. Calculate and average according to the supplementary correlation coefficients and the processed dimensionless set of monitoring parameters to obtain supplementary parameters;

[0131] S744. According to the supplementary parameters and the processed dimensionless set of monitoring parameters, obtain a corrected dimensionless set of monitoring parameters.

[0132] S8. Based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the development trend and warning level of the monitored dangerous rock mass, and issue a warning indication.

[0133] Specifically, as Figure 7 shown in the flowchart of obtaining the development trend of the monitored dangerous rock mass in the embodiment of the method for associated warning of multiple parameters of dangerous rock based on comprehensive perception within the air, space, and ground of the present invention, in S8, based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the development trend and warning level of the monitored dangerous rock mass, and issue a warning indication, including:

[0134] S81. Based on the monitoring data stream of the monitored dangerous rock mass, after cumulative processing, obtain the multi-parameter bandwidth;

[0135] S82. Based on the multi-parameter bandwidth, use filtering processing to separate the interference phase in the multi-parameter bandwidth, and obtain the variation relationship between the multi-parameter bandwidth and time, as Figure 9 shown in the schematic diagram of the variation of the multi-parameter bandwidth with time in the embodiment of the method for associated warning of multiple parameters of dangerous rock based on comprehensive perception within the air, space, and ground of the present invention;

[0136] S83. Based on the training database of the monitored dangerous rock mass, obtain the bandwidth threshold;

[0137] S84. Based on the bandwidth threshold and the variation relationship between the multi-parameter bandwidth and time, obtain the development trend of the monitored dangerous rock mass, and the development trend of the monitored dangerous rock mass includes: prediction of the imminent trend of the monitored dangerous rock mass, prediction of the short-term trend of the monitored dangerous rock mass, and prediction of the long-term trend of the monitored dangerous rock mass;

[0138] S85. Based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the warning level of the monitored dangerous rock mass, and issue a warning indication.

[0139] Furthermore, as Figure 8 shown in the flowchart of obtaining the warning level of the monitored dangerous rock mass in the embodiment of the method for associated warning of multiple parameters of dangerous rock based on comprehensive perception within the air, space, and ground of the present invention, in S85, based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, obtain the warning level of the monitored dangerous rock mass, and issue a warning indication, including:

[0140] S851. Based on the monitoring data stream of the monitored dangerous rock mass, obtain the three-dimensional coordinate change amount data stream, displacement data stream, tilt data stream, acceleration data stream, microseismic energy data stream, and structural plane sliding data stream;

[0141] S852. Based on the training database for monitoring dangerous rock masses, obtain the weights of the three-dimensional coordinate change amount, displacement, inclination, acceleration, microseismic energy, and structural plane sliding. The relationship between the weights of the three-dimensional coordinate change amount, displacement, inclination, acceleration, microseismic energy, and structural plane sliding conforms to formula (1).

[0142] (1)

[0143] In the formula: is the weight of the three-dimensional coordinate change amount at the position of the i-th point, is the displacement weight at the position of the i-th point, is the inclination weight at the position of the i-th point, is the acceleration weight at the position of the i-th point, is the microseismic energy weight at the position of the i-th point, is the structural plane sliding weight at the position of the i-th point;

[0144] S853. According to the three-dimensional coordinate change amount data stream, displacement data stream, inclination data stream, acceleration data stream, microseismic energy data stream, structural plane sliding data stream, the weight of the three-dimensional coordinate change amount, displacement weight, inclination weight, acceleration weight, microseismic energy weight, and structural plane sliding weight, through formula (2), obtain a multi-parameter function model.

[0145] (2)

[0146] In the formula: is the multi-parameter function model, , is the three-dimensional coordinate change amount data stream at the position of the i-th point, , is the displacement data stream at the position of the i-th point, , is the inclination data stream at the position of the i-th point, , is the acceleration data stream at the position of the i-th point, , is the microseismic energy data stream at the position of the i-th point, , is the structural plane sliding data stream at the position of the i-th point, ;

[0147] S854. According to the multi-parameter function model, through the early warning response rule, obtain the early warning level for monitoring dangerous rock masses. The early warning response rule includes:

[0148] ,the danger level is level Ⅳ, which is a blue early warning.

[0149] , the danger level is level III, a yellow warning,

[0150] , the danger level is level II, an orange warning,

[0151] , the danger level is level I, a red warning;

[0152] When the danger level is level I to III, the drone is activated to cruise and photograph the location of the abnormal point, obtain the image data of the abnormal point, compare and analyze the obtained image data with the data in the 3D model, calculate the rock mass displacement and deformation speed in the image area, and conduct further detailed analysis and prediction.

[0153] As Figure 10 shown in the system block diagram of the embodiment of the dangerous rock multi-parameter correlation warning system based on comprehensive perception in the air, on the ground and in the underground of the present invention, the present invention provides a dangerous rock multi-parameter correlation warning system based on comprehensive perception in the air, on the ground and in the underground. This system is applied to the dangerous rock multi-parameter correlation warning method based on comprehensive perception in the air, on the ground and in the underground. The system includes a data acquisition module, a training module, a model generation module, a correction module, an update module, a monitoring module, a processing module and a warning module. Specifically,

[0154] The data acquisition module is used to obtain the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters by sorting and analyzing the precursor information of the instability cases of the dangerous rock mass and using the orthogonal design method;

[0155] The training module is used to obtain a training database with multi-parameter correlation through machine learning algorithms according to the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters;

[0156] The model generation module is used to obtain the image data of the dangerous rock mass in the monitoring area by the drone and obtain the 3D model of the monitored dangerous rock mass;

[0157] The correction module is used to obtain the correction data stream by numerically simulating the instability process of the dangerous rock mass according to the 3D model of the monitored dangerous rock mass and extracting the data information of the relevant parameters during the instability process;

[0158] The update module is used to retrain and update according to the correction data stream and the training database with multi-parameter correlation to obtain the training database of the monitored dangerous rock mass;

[0159] The monitoring module is used to deploy a monitoring system for the dangerous rock mass in the monitoring area to obtain the monitoring data information of the monitored dangerous rock mass;

[0160] A processing module, configured to perform normalization and weighting processing based on the monitoring data information of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, so as to obtain the monitoring data stream of the monitored dangerous rock mass;

[0161] An early warning module, configured to obtain the development trend and early warning level of the monitored dangerous rock mass based on the monitoring data stream of the monitored dangerous rock mass and the training database of the monitored dangerous rock mass, and issue an early warning indication.

[0162] The present invention provides a multi-parameter correlation early warning method and system for dangerous rocks based on integrated air-space-ground-inner perception. The invention integrates various and multi-scale monitoring data of dangerous rock masses by deploying an "air-space-ground-inner" integrated monitoring system in the monitoring area of dangerous rock masses, and uses the precursor information of the instability of dangerous rock masses to construct a multi-parameter early warning system with data complementarity, data fusion and data correlation, which solves the problem of one-sided data in prediction using single-factor or single-scale information, realizes data complementarity and comprehensive early warning. At the same time, a training database of the monitored dangerous rock mass is generated based on multiple dangerous rock mass instability cases and corrected data streams, and an early warning function model containing multi-parameter information is constructed, so as to evaluate and predict the development trend of the monitoring data stream of the monitored dangerous rock mass and the stable state of the dangerous rock mass, and issue an early warning indication according to the early warning level of the monitored dangerous rock mass, improving the early warning accuracy and timeliness of dangerous rock mass collapse disasters.

[0163] It can be understood that the present invention is described through the above embodiments, and should not be construed as a limitation on the implementation manner and scope of the present invention. Those skilled in the art know that without departing from the spirit and scope of the present invention, these features and embodiments can be variously changed or equivalently replaced. In addition, under the teaching of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application belong to the scope protected by the present invention.

Claims

1. A multi-parameter correlation early warning method for dangerous rocks based on comprehensive perception within the air, space, and ground, characterized in that The method includes: S1. By sorting out and analyzing the precursor information of unstable rock mass cases, using the orthogonal design method, obtaining the weight factors of parameters related to the instability of the unstable rock mass and the correlation coefficients between the parameters; S2. According to the weight factors of the parameters related to the instability of the unstable rock mass and the correlation coefficients between the parameters, through machine learning algorithm training, obtaining a training database with multi-parameter association; S3. Obtaining the image data of the unstable rock mass in the monitoring area by using an unmanned aerial vehicle, and obtaining the three-dimensional model of the monitored unstable rock mass; S4. According to the three-dimensional model of the monitored unstable rock mass, through numerical simulation of the instability process of the unstable rock mass and extracting the data information of relevant parameters during the instability process, obtaining a corrected data stream; S5. According to the corrected data stream and the training database with multi-parameter association, training and updating again to obtain the training database of the monitored unstable rock mass; S6. Installing a monitoring system on the unstable rock mass in the monitoring area to obtain the monitoring data information of the monitored unstable rock mass; S7. According to the monitoring data information of the monitored unstable rock mass and the training database of the monitored unstable rock mass, using normalization and weighting processing to obtain the monitoring data stream of the monitored unstable rock mass; S8. According to the monitoring data stream of the monitored unstable rock mass and the training database of the monitored unstable rock mass, obtaining the development trend and warning level of the monitored unstable rock mass, and giving a warning indication.

2. The multi-parameter correlation early warning method for dangerous rocks based on integrated perception within the air, space and ground according to claim 1, wherein In the above S1, by sorting out and analyzing the precursor information of unstable rock mass cases, using the orthogonal design method, obtaining the weight factors of parameters related to the instability of the unstable rock mass and the correlation coefficients between the parameters, including: S11. According to the precursor information of unstable rock mass cases, extracting the parameters related to the instability of the unstable rock mass and performing normalization processing to obtain the dimensionless parameters of various precursor information. The parameters related to the instability of the unstable rock mass include: the change amount of the three-dimensional coordinates of characteristic points, displacement, dip angle, acceleration, microseismic energy, and the sliding amount of structural planes; S12. According to the dimensionless parameters of various precursor information, using the orthogonal design method to obtain the weight factors of parameters related to the instability of the unstable rock mass; S13. According to the dimensionless parameters and the weight factors of the parameters related to the instability of the unstable rock mass, performing correlation analysis by using the orthogonal design method to obtain the correlation coefficients between the parameters.

3. The multi-parameter correlation early warning method for dangerous rocks based on integrated sensing in the air, on the ground and underground according to claim 1, wherein, In the above S3, obtaining the image data of the unstable rock mass in the monitoring area by using an unmanned aerial vehicle, and obtaining the three-dimensional model of the monitored unstable rock mass, including: S31. Installing multiple image control points on the unstable rock mass in the monitoring area, using RTK positioning technology to obtain the three-dimensional coordinates of the multiple image control points; S32. Using an unmanned aerial vehicle to collect multi-angle images of the unstable rock mass in the monitoring area to obtain the two-dimensional image data of the monitored unstable rock mass; S33. According to the two-dimensional image data of the monitored unstable rock mass and the three-dimensional coordinates of the multiple image control points, using the SFM three-dimensional image reconstruction technology to obtain the three-dimensional model of the monitored unstable rock mass.

4. The multi-parameter correlation early warning method for dangerous rocks based on integrated sensing in the air, on the ground and underground according to claim 1, wherein, In the above S4, according to the three-dimensional model of the monitored unstable rock mass, through numerical simulation of the instability process of the unstable rock mass and extracting the data information of relevant parameters during the instability process, obtaining a corrected data stream, including: S41. According to the three-dimensional model of the monitored unstable rock mass, obtaining the numerical model of the finite element simulation of the monitored unstable rock mass; S42. According to the numerical model for finite element simulation of monitoring unstable rock masses, use numerical simulation software to simulate the instability process of the monitored unstable rock masses, and obtain the correction parameter information related to the instability of the monitored unstable rock masses. The correction parameter information related to the instability of the monitored unstable rock masses includes: the change in the three-dimensional coordinates of the characteristic points related to the instability of the monitored unstable rock masses, the displacement related to the instability of the monitored unstable rock masses, the inclination angle related to the instability of the monitored unstable rock masses, the acceleration related to the instability of the monitored unstable rock masses, the microseismic energy related to the instability of the monitored unstable rock masses, and the sliding amount of the structural plane related to the instability of the monitored unstable rock masses. S43. By sorting out and analyzing the correction parameter information related to the instability of the monitored unstable rock masses, use the orthogonal design method to obtain the weight factors of the correction parameters related to the instability of the monitored unstable rock masses and the correlation coefficients between the correction parameters. S44. According to the weight factors of the correction parameters related to the instability of the monitored unstable rock masses and the correlation coefficients between the correction parameters, obtain the correction data stream.

5. The multi-parameter correlation early warning method for dangerous rocks based on integrated air, space and ground perception according to claim 1, characterized in that The monitoring system includes: GNSS positioning equipment, which is used to realize the three-dimensional coordinate positioning of the characteristic points of the unstable rock masses in the monitored area at the airspace scale. An unmanned aerial vehicle and a takeoff and landing platform, which are used to realize the construction of the three-dimensional model of the unstable rock masses in the monitored area and local photography at the celestial scale. Surface monitoring instruments, which are used to obtain the mesoscopic information of the unstable rock masses in the monitored area at the regional scale, including the displacement, inclination angle, acceleration, and microseismicity on the surface of the unstable rock masses in the monitored area. Large flexible deformation gauges in boreholes, which are used to realize the sliding perception of the internal structural planes of the unstable rock masses in the monitored area at the inner domain scale. A data acquisition unit, which is used to collect data information.

6. The multi-parameter correlation early warning method for dangerous rocks based on integrated sensing in the air, on the ground and in the underground according to claim 1, wherein In step S7, according to the monitoring data information of the monitored unstable rock masses and the training database of the monitored unstable rock masses, perform normalization and weighting processing to obtain the monitoring data stream of the monitored unstable rock masses, including: S71. According to the monitoring data information of the monitored unstable rock masses, obtain a set of monitoring parameters related to the instability of the monitored unstable rock masses. S72. According to the set of monitoring parameters related to the instability of the monitored unstable rock masses, perform normalization processing to obtain a dimensionless set of monitoring parameters. S73. According to the dimensionless set of monitoring parameters, through Gaussian filtering for noise reduction and equalization processing, obtain the processed dimensionless set of monitoring parameters. S74. According to the processed dimensionless set of monitoring parameters and the training database of the monitored unstable rock masses, through the correlation coefficients in the training database of the monitored unstable rock masses, supplement the monitoring parameters that have not been collected and have incorrect information to obtain a corrected dimensionless set of monitoring parameters. S75. According to the corrected dimensionless set of monitoring parameters and the training database of the monitored unstable rock masses, perform weighting according to the weight factors in the training database of the monitored unstable rock masses to obtain the monitoring data stream of the monitored unstable rock masses.

7. The multi-parameter correlation early warning method for dangerous rocks based on integrated sensing in the air, on the ground and underground according to claim 6, characterized in that In step S74, according to the processed dimensionless set of monitoring parameters and the training database of the monitored unstable rock masses, through the correlation coefficients in the training database of the monitored unstable rock masses, supplement the monitoring parameters that have not been collected and have incorrect information to obtain a corrected dimensionless set of monitoring parameters, including: S741. Obtain the correlation coefficients between various parameters related to the instability of the monitored dangerous rock mass according to the training database for monitoring the dangerous rock mass; S742. Select the 3 correlation coefficients with the largest correlation coefficients related to the monitoring parameters that have not been collected and have incorrect information according to the correlation coefficients between various parameters related to the instability of the monitored dangerous rock mass and the processed dimensionless monitoring parameter set, and obtain the supplementary correlation coefficients; S743. Calculate and average according to the supplementary correlation coefficients and the processed dimensionless monitoring parameter set to obtain supplementary parameters; S744. Obtain the corrected dimensionless monitoring parameter set according to the supplementary parameters and the processed dimensionless monitoring parameter set.

8. The multi-parameter correlation early warning method for dangerous rocks based on integrated sensing in the air, on the ground and underground according to claim 1, characterized in that In step S8, obtain the development trend and warning level of the monitored dangerous rock mass according to the monitoring data stream of the monitored dangerous rock mass and the training database for the monitored dangerous rock mass, and give a warning indication, including: S81. Obtain the multi-parameter bandwidth through cumulative processing according to the monitoring data stream of the monitored dangerous rock mass; S82. Separate the interference phase in the multi-parameter bandwidth by filtering processing according to the multi-parameter bandwidth to obtain the variation relationship between the multi-parameter bandwidth and time; S83. Obtain the bandwidth threshold according to the training database for the monitored dangerous rock mass; S84. Obtain the development trend of the monitored dangerous rock mass according to the bandwidth threshold and the variation relationship between the multi-parameter bandwidth and time. The development trend of the monitored dangerous rock mass includes: prediction of the imminent trend of the monitored dangerous rock mass, prediction of the short-term trend of the monitored dangerous rock mass, and prediction of the long-term trend of the monitored dangerous rock mass; S85. Obtain the warning level of the monitored dangerous rock mass according to the monitoring data stream of the monitored dangerous rock mass and the training database for the monitored dangerous rock mass, and give a warning indication.

9. The multi-parameter correlation early warning method for dangerous rocks based on integrated perception in the air, on the ground and underground according to claim 1, characterized in that In step S85, obtain the warning level of the monitored dangerous rock mass according to the monitoring data stream of the monitored dangerous rock mass and the training database for the monitored dangerous rock mass, and give a warning indication, including: S851. Obtain the three-dimensional coordinate change amount data stream, displacement data stream, tilt data stream, acceleration data stream, microseismic energy data stream, and structural plane sliding data stream according to the monitoring data stream of the monitored dangerous rock mass; S852. Obtain the three-dimensional coordinate change amount weight, displacement weight, tilt weight, acceleration weight, microseismic energy weight, and structural plane sliding weight according to the training database for the monitored dangerous rock mass. The relationship between the three-dimensional coordinate change amount weight, the displacement weight, the tilt weight, the acceleration weight, the microseismic energy weight, and the structural plane sliding weight conforms to formula (1); （1） Where: is the weight of the three-dimensional coordinate change at the position of the i-th point, is the displacement weight at the position of the i-th point, is the tilt weight at the position of the i-th point, is the acceleration weight at the position of the i-th point, is the microseismic energy weight at the position of the i-th point, is the weight of the structural plane sliding at the position of the i-th point; S853. Obtain the multi-parameter function model through formula (2) according to the three-dimensional coordinate change amount data stream, the displacement data stream, the tilt data stream, the acceleration data stream, the microseismic energy data stream, the structural plane sliding data stream, the three-dimensional coordinate change amount weight, the displacement weight, the tilt weight, the acceleration weight, the microseismic energy weight, and the structural plane sliding weight. （2） In the formula: is a multi-parameter function model, is the data stream of the three-dimensional coordinate change amount at the position of the i-th point, , is the displacement data stream at the position of the i-th point, , is the tilt data stream at the position of the i-th point, , is the acceleration data stream at the position of the i-th point, , is the microseismic energy data stream at the position of the i-th point, , is the structural plane sliding data stream at the position of the i-th point, ; S854. According to the multi-parameter function model, through the warning response rule, obtain the warning level for monitoring the dangerous rock mass, and the warning response rule includes: , the danger level is level IV, a blue warning, , the danger level is level III, a yellow warning, , the danger level is level II, an orange warning, , the risk level is Class I, which is a red warning; When the danger level is from level I to level III, start the drone to conduct cruise shooting on the positions with the danger level from level I to level III.

10. A multi-parameter correlation early warning system for dangerous rocks based on comprehensive sensing in the air, space and on the ground, which is used to implement the multi-parameter correlation early warning method for dangerous rocks based on comprehensive sensing in the air, space and on the ground as described in any one of claims 1-9, characterized in that, The system includes: A data acquisition module, which is used to obtain the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters by sorting and analyzing the precursor information of the instability cases of the dangerous rock mass and adopting the orthogonal design method; A training module, which is used to obtain a training database associated with multiple parameters through machine learning algorithm training according to the weight factors of the parameters related to the instability of the dangerous rock mass and the correlation coefficients between the parameters; A model generation module, which is used to obtain the three-dimensional model of the monitored dangerous rock mass by using the drone to acquire the image data of the dangerous rock mass in the monitoring area; A correction module, which is used to obtain the corrected data stream by numerically simulating the instability process of the dangerous rock mass and extracting the data information of the relevant parameters during the instability process according to the three-dimensional model of the monitored dangerous rock mass; An update module, which is used to retrain and update according to the corrected data stream and the training database associated with multiple parameters to obtain the training database for monitoring the dangerous rock mass; A monitoring module, which is used to deploy a monitoring system for the dangerous rock mass in the monitoring area to obtain the monitoring data information of the dangerous rock mass; A processing module, which is used to obtain the monitoring data stream of the dangerous rock mass by adopting normalization and weighting processing according to the monitoring data information of the dangerous rock mass and the training database of the dangerous rock mass; An early warning module, which is used to obtain the development trend and early warning level of the monitored dangerous rock mass according to the monitoring data stream of the dangerous rock mass and the training database of the dangerous rock mass, and issue an early warning indication.

Images