CN-122024441-A - Geological disaster dynamic comprehensive early warning method and system based on three-dimensional monitoring

CN122024441ACN 122024441 ACN122024441 ACN 122024441ACN-122024441-A

Abstract

The invention relates to the technical field of intelligent early warning, in particular to a geological disaster dynamic comprehensive early warning method and system based on three-dimensional monitoring, wherein the method comprises the steps of acquiring multi-dimensional three-dimensional monitoring data streams of a geological disaster hidden danger area in real time, and carrying out time sequence alignment and space registration on the multi-dimensional three-dimensional monitoring data streams to obtain a multi-source fusion data set; the method comprises the steps of identifying key coupling characteristic factors in a multisource fusion data set, outputting a three-dimensional dynamic risk field model of a geological disaster hidden danger area in a mode of a risk probability cloud graph in a dynamic response process under the action of multisource coupling, extracting dynamic characteristic vectors of risk evolution rate and risk space aggregation degree in the three-dimensional dynamic risk field model, carrying out early warning decision on the geological disaster hidden danger area to obtain early warning grade instructions, reversely optimizing parameters in the dynamic response process, and using the optimized parameters for construction and update of the three-dimensional dynamic risk field model of the next round.

Inventors

SHAO WEI
LI MEIJUN
YU WENJUN
LIN QIGEN
YE XIAO
ZHANG GAOMING

Assignees

南京信息工程大学

Dates

Publication Date: 20260512
Application Date: 20260131

Claims (10)

1. The utility model provides a geological disaster dynamic comprehensive early warning method based on three-dimensional monitoring which is characterized in that the method comprises the following steps: The method comprises the steps of acquiring a multi-dimensional three-dimensional monitoring data stream of a geological disaster hidden danger area in real time through a multi-source sensing equipment network deployed in the geological disaster hidden danger area, and carrying out time sequence alignment and space registration on the multi-dimensional three-dimensional monitoring data stream to obtain a multi-source fusion data set of the geological disaster hidden danger area; Based on a dynamic Bayesian network, carrying out causal association mining and dynamic contribution weight quantification on the multi-source fusion data set in a geological disaster inoculation process to obtain key coupling characteristic factors in the multi-source fusion data set; Simulating a dynamic response process of the geologic body under the action of multi-field coupling according to the key coupling characteristic factors, and outputting a three-dimensional dynamic risk field model of the geological disaster hidden danger area in the form of a risk probability cloud picture; extracting dynamic feature vectors of risk evolution rate and risk space aggregation degree in the three-dimensional dynamic risk field model; Performing early warning decision on the geological disaster hidden danger area according to the dynamic feature vector and a preset early warning decision rule to obtain an early warning grade instruction matched with the current risk evolution stage in the geological disaster hidden danger area, wherein the preset early warning decision rule is used for defining quantization indexes of the risk evolution rate in the dynamic feature vector, different numerical ranges of risk space aggregation degree indexes and corresponding relations between combinations and early warning grades, and the risk evolution stages correspond to the early warning grades; And reversely optimizing parameters in the dynamic response process according to the early warning grade instruction and the morphological change characteristics of the three-dimensional dynamic risk field model, and using the optimized parameters for constructing and updating the three-dimensional dynamic risk field model of the next round.
2. The method for dynamically and comprehensively early warning a geological disaster based on three-dimensional monitoring according to claim 1, wherein the steps of obtaining the multi-dimensional three-dimensional monitoring data stream of the geological disaster hidden danger area in real time through a multi-source sensing equipment network deployed in the geological disaster hidden danger area, and performing time sequence alignment and space registration on the multi-dimensional three-dimensional monitoring data stream to obtain a multi-source fusion data set of the geological disaster hidden danger area comprise the following steps: The method comprises the steps of laying and activating sensing equipment deployed in the earth surface and internal drilling of a geological disaster hidden danger body to form a multi-source sensing equipment network of the geological disaster hidden danger area; Receiving earth surface displacement data streams, soil moisture content data streams, underground rock-soil mass stress data streams and microseismic signal data streams from the multi-source sensing equipment network, and adding a time stamp to each data stream to obtain multi-source asynchronous data streams of the geological disaster hidden danger area; Performing interpolation synchronization on the multi-source asynchronous data stream based on the time stamp to obtain a synchronization monitoring data stream with completely aligned time sequences; Mapping each monitoring data point in the synchronized monitoring data stream to a unified three-dimensional space coordinate system based on a preset geographic information system coordinate reference to obtain a registration data stream of the geological disaster hidden danger area; And carrying out structural recombination on the registration data stream according to the time-space dimension to obtain the multisource fusion data set of the geological disaster hidden danger area.
3. The method for dynamically and comprehensively early warning geological disasters based on three-dimensional monitoring according to claim 1, wherein the step of carrying out causal association mining and dynamic contribution weight quantification on the multi-source fusion dataset based on a dynamic Bayesian network to obtain key coupling characteristic factors in the multi-source fusion dataset comprises the following steps: Performing abnormal data cleaning and standardized normalization processing on the multi-source fusion data set to obtain multi-source monitoring time sequence data of the geological disaster hidden danger area; Extracting the association relation between the earth surface displacement change rate and the underground rock-soil body stress change trend from the multi-source monitoring time sequence data to obtain the displacement-stress covariant characteristic of the geological disaster hidden danger area; Extracting the association relation between the space-time change of the water content of the soil and the occurrence sequence of the microseismic signal event from the multi-source monitoring time sequence data to obtain seepage-microseismic association characteristics of the geological disaster hidden danger area; according to the priori knowledge of a preset geological disaster mechanics model, the contribution degree of the displacement-stress covariant characteristic and the seepage-microseism association characteristic to a geological disaster inoculation stage is estimated; And screening out the characteristic with the contribution higher than a set standard from the displacement-stress covariant characteristic and the seepage-microseism correlation characteristic based on the contribution, and obtaining a key coupling characteristic factor of the geological disaster hidden danger area.
4. The method for dynamically and comprehensively early warning geological disasters based on three-dimensional monitoring as set forth in claim 3, wherein the calculation formula of the contribution degree of the displacement-stress covariant characteristic to the inoculation stage of the geological disasters is as follows: ; In the formula, For the contribution of the displacement-stress covariate feature, To normalize the displacement-stress covariate characteristics for self-varying sensitivity, In order to normalize the seepage-microseismic correlation characteristics to obtain self-variation sensitivity, For the dynamic coupling of the displacement-stress covariant feature and the percolation-microseismic-related feature, In order to couple the enhancement coefficients of the light source, Is a cross adjustment coefficient; The calculation formula of the contribution degree of the seepage-microseismic associated features to the geological disaster inoculation stage is as follows: ; In the formula, For the contribution of the seepage-microseismic correlation features, To normalize the displacement-stress covariate characteristics for self-varying sensitivity, In order to normalize the seepage-microseismic correlation characteristics to obtain self-variation sensitivity, For the dynamic coupling of the displacement-stress covariant feature and the percolation-microseismic-related feature, In order to couple the enhancement coefficients of the light source, Is a cross-adjustment factor.
5. The method for dynamically and comprehensively early warning geological disasters based on three-dimensional monitoring according to claim 4, wherein the simulating of the dynamic response process of the geological body under the action of multi-field coupling according to the key coupling characteristic factors comprises the following steps: packaging constitutive relation sets of interaction relations of stress fields, seepage fields and deformation fields into a coupling analysis engine; The key coupling characteristic factors are used as boundary conditions and driving parameters and are configured into corresponding input interfaces of the coupling analysis engine; setting a time step and a total simulation duration for simulating a dynamic response process in the coupling analysis engine; Driving the coupling analysis engine to run iteratively according to the time step within the total simulation time length; and after the coupling analysis engine finishes iterative operation, extracting and outputting the simulation state parameters updated by each iterative step to obtain the time-space evolution sequence of the geologic body.
6. The method for dynamically and comprehensively early-warning geological disasters based on three-dimensional monitoring according to claim 5, wherein the driving the coupling analysis engine to operate iteratively in the total simulation duration according to the time step comprises: Initializing an initial state vector of a geologic body based on the key coupling characteristic factors and preset initial geological parameters in the coupling analysis engine; starting an iteration loop aiming at the total simulation time length by taking the initial state vector as a start; In each iteration, according to the key coupling characteristic factors corresponding to the current iteration step and the geologic body state vector corresponding to the current iteration step, invoking the constitutive relation set to generate the variable quantity of the geologic body state under the current multi-field coupling effect; updating the current geologic body state vector according to the variable quantity to obtain a geologic body state vector corresponding to the next iteration step; taking the updated geologic body state vector as a new current geologic body state vector of the next iteration period; And when the iteration times reach the iteration steps corresponding to the total simulation time length, terminating the iteration loop and outputting the updated geologic body state vector in all iterations.
7. The method for dynamically and comprehensively early-warning the geological disaster based on the three-dimensional monitoring as set forth in claim 6, wherein the outputting the three-dimensional dynamic risk field model of the geological disaster hidden danger zone in the form of a risk probability cloud graph comprises the following steps: A state parameter value corresponding to the current simulation moment in the time-space evolution sequence is given to each grid unit on the three-dimensional space grid model of the geological disaster hidden danger zone; predicting a risk probability value of the grid unit subjected to instability damage in a next simulation period based on a preset instability criterion and a risk evaluation rule; smoothing the risk probability value to obtain risk probability field data of the geological disaster hidden danger area; and characterizing different risk probability grades and spatial distribution in the risk probability field data by using different colors and transparency to obtain the three-dimensional dynamic risk field model of the geological disaster hidden danger area.
8. The method for dynamically and comprehensively early-warning geological disasters based on three-dimensional monitoring according to claim 7, wherein the extracting the dynamic feature vector of the risk evolution rate and the risk space aggregation degree in the three-dimensional dynamic risk field model comprises: extracting a risk probability time sequence of each space unit in a preset time window from the three-dimensional dynamic risk field model; Taking the time-dependent change rate of the risk probability of each space unit in the risk probability time sequence as a quantification index of the risk evolution rate in the geological disaster hidden danger area; Identifying a space aggregation degree index of a high-risk probability area in the three-dimensional dynamic risk field model; and combining the quantitative index of the risk evolution rate with the space aggregation degree index to obtain a dynamic feature vector of the risk evolution rate and the risk space aggregation degree in the three-dimensional dynamic risk field model.
9. The method for dynamically and comprehensively early-warning a geological disaster based on three-dimensional monitoring according to claim 8, wherein the reversely optimizing parameters in the dynamic response process according to the early-warning level instruction and morphological change characteristics of the three-dimensional dynamic risk field model comprises: Generating an optimization target for improving the precision of the next simulation period according to the level of the early warning level instruction and the morphological change characteristic of the three-dimensional dynamic risk field model; Identifying a parameter set to be adjusted associated with the optimization objective from a constitutive relation set of the coupling analysis engine based on the optimization objective; and fine-tuning the parameter set to be adjusted according to a preset adjustment strategy to generate optimized parameters.
10. The utility model provides a geological disaster dynamic comprehensive early warning system based on three-dimensional monitoring, which is characterized in that the system is used for realizing the geological disaster dynamic comprehensive early warning method based on three-dimensional monitoring, and comprises the following steps: The multi-source data acquisition and fusion module is used for laying and activating a multi-source sensing equipment network deployed in a geological disaster hidden danger area, acquiring a multi-dimensional three-dimensional monitoring data stream of the geological disaster hidden danger area in real time, and carrying out time sequence alignment and space registration on the multi-dimensional three-dimensional monitoring data stream to obtain a multi-source fusion data set of the geological disaster hidden danger area; The key feature factor identification module is used for carrying out causal association mining and dynamic contribution weight quantification on the multi-source fusion data set in the geological disaster inoculation process based on a dynamic Bayesian network to obtain key coupling feature factors in the multi-source fusion data set; The dynamic risk field simulation and visualization module is used for simulating the dynamic response process of the geologic body under the action of multi-field coupling according to the key coupling characteristic factors and outputting a three-dimensional dynamic risk field model of the geological disaster hidden danger area in the form of a risk probability cloud chart; The risk evolution feature extraction module is used for extracting dynamic feature vectors of risk evolution rate and risk space aggregation degree in the three-dimensional dynamic risk field model; the intelligent early warning decision module is used for carrying out early warning decision on the geological disaster hidden danger area according to the dynamic feature vector to obtain an early warning grade instruction matched with the current risk evolution stage in the geological disaster hidden danger area; and the model parameter self-adaptive optimization module is used for reversely optimizing parameters in the dynamic response process according to the early warning grade instruction and the morphological change characteristics of the three-dimensional dynamic risk field model, and using the optimized parameters for constructing and updating the three-dimensional dynamic risk field model of the next round.

Description

Geological disaster dynamic comprehensive early warning method and system based on three-dimensional monitoring Technical Field The invention relates to the technical field of intelligent early warning, in particular to a geological disaster dynamic comprehensive early warning method and system based on three-dimensional monitoring. Background The existing geological disaster early warning technology has obvious limitation in the data acquisition and processing links, lacks the three-dimensional monitoring capability of geological disaster hidden danger areas, depends on single-dimension or small-quantity monitoring data, and does not perform effective time sequence alignment and spatial registration processing on multi-source heterogeneous data. The problem that monitoring data from different sources are asynchronous in time sequence and nonuniform in space reference is caused, so that data fusion quality is low, cooperative change characteristics of the ground surface and the inside of the geologic body cannot be accurately captured, and comprehensive perception and judgment of a disaster inoculation process are further affected. Meanwhile, the prior art fails to accurately identify key coupling characteristic factors in geological disaster evolution, the dynamic response simulation of the geologic body under the action of multi-field coupling lacks pertinence and accuracy, the risk field model construction is static or semi-static, and real-time evolution rules of risks are difficult to reflect. In addition, the existing early warning method lacks a dynamic parameter optimization mechanism, model parameters are fixed and unchanged, self-adaptive adjustment cannot be carried out according to early warning results and risk field form changes, deviation exists between early warning levels and actual risk evolution stages, early warning timeliness and reliability are insufficient, and actual requirements of accurate early warning of geological disasters are difficult to meet. Disclosure of Invention The invention provides a geological disaster dynamic comprehensive early warning method and system based on three-dimensional monitoring, which are used for solving the problems in the background technology. In order to achieve the above purpose, the invention provides a geological disaster dynamic comprehensive early warning method based on three-dimensional monitoring, which comprises the following steps: S1, laying and activating a multi-source sensing equipment network deployed in a geological disaster hidden danger area, acquiring a multi-dimensional three-dimensional monitoring data stream of the geological disaster hidden danger area in real time, and performing time sequence alignment and space registration on the multi-dimensional three-dimensional monitoring data stream to obtain a multi-source fusion data set of the geological disaster hidden danger area; S2, carrying out causal association mining and dynamic contribution weight quantification on the multisource fusion data set in a geological disaster inoculation process based on a dynamic Bayesian network to obtain key coupling characteristic factors in the multisource fusion data set; S3, simulating a dynamic response process of the geologic body under the action of multi-field coupling according to the key coupling characteristic factors, and outputting a three-dimensional dynamic risk field model of the geologic hazard hidden danger area in the form of a risk probability cloud picture; s4, extracting dynamic feature vectors of risk evolution rate and risk space aggregation degree in the three-dimensional dynamic risk field model; s5, carrying out early warning decision on the geological disaster hidden danger area according to the dynamic feature vector to obtain an early warning grade instruction matched with the current risk evolution stage in the geological disaster hidden danger area; And S6, reversely optimizing parameters in the dynamic response process according to the early warning grade instruction and the morphological change characteristics of the three-dimensional dynamic risk field model, and using the optimized parameters for constructing and updating the three-dimensional dynamic risk field model of the next round. In a preferred embodiment, the laying and activating a multi-source sensing device network deployed in a geological disaster hidden danger area, acquiring a multi-dimensional stereoscopic monitoring data stream of the geological disaster hidden danger area in real time, and performing time sequence alignment and spatial registration on the multi-dimensional stereoscopic monitoring data stream to obtain a multi-source fusion data set of the geological disaster hidden danger area, including: The method comprises the steps of laying and activating sensing equipment deployed in the earth surface and internal drilling of a geological disaster hidden danger body to form a multi-source sensing equipment network of the geological