Big data-based geological disaster risk assessment platform and method

Abstract

The invention discloses a geological disaster risk assessment platform and method based on big data, and relates to the technical field of geological disaster risk assessment of a mining construction area. The platform divides the mining construction area into a plurality of sub-areas, and collects geological data, induced disturbance data and disaster-bearing data in real time; the multi-source data is uploaded to the central platform after being preprocessed; by constructing a convolutional neural network model and utilizing data to train, recognition and joint dynamic risk assessment of regional geological sensitivity, induction factor response and human activity bearing characteristics are realized; calculating a geological sensitivity coefficient, and evaluating geological environment stability; calculating an induction intensity coefficient, and judging disaster induction intensity; calculating a disaster-bearing vulnerability coefficient, and evaluating human activity load and evacuation capability; and aiming at each evaluation result, the platform can formulate a corresponding risk control strategy in time, the safety management level of a mining construction area is improved, and the geological disaster early warning and prevention capability is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geological disaster risk assessment of mining construction area, in particular to a geological disaster risk assessment platform and method based on big data. BACKGROUND

[0002] As one of the important types of natural disasters, geological disasters have the characteristics of strong suddenness, wide influence range and complex causes, and often cause serious threats to field engineering construction, urban planning and infrastructure safety. With the continuous advancement of large-scale projects such as mountainous area transportation, water conservancy and hydropower, and mineral resources development, the probability of geological disasters induced is significantly increased, and the traditional geological disaster risk identification and assessment method based on experience judgment cannot meet the actual needs of modern engineering for dynamic and accurate early warning.

[0003] In the prior art, geological disaster risk assessment mostly relies on field investigation, single data source analysis and static risk classification model, and has problems of insufficient data collection dimension, delayed response and difficulty in dynamic updating. At the same time, in the face of multi-source heterogeneous geological, meteorological, construction and population distribution data, there is a lack of unified data fusion framework and intelligent analysis mechanism, and it is difficult to realize accurate modeling of disaster cause mechanism and real-time prediction of its evolution process.

[0004] In recent years, emerging technologies such as big data, artificial intelligence and remote sensing monitoring have been gradually applied in geological disaster research, providing new ideas for improving the accuracy and efficiency of disaster risk assessment. However, how to build a dynamic risk assessment platform for field construction area, which integrates geological environmental characteristics, external induced disturbance and human activity disaster bearing capacity, still faces technical bottlenecks such as complex data processing, model coupling difficulty and imperfect response mechanism. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a geological disaster risk assessment platform and method based on big data to solve the problems mentioned in the background art.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: a geological disaster risk assessment platform and method based on big data, comprising: a data acquisition module, an AI dynamic risk assessment model establishment module, a geological monitoring module, an induced disturbance monitoring module and a disaster bearing monitoring module; The data acquisition module is used to divide the entire mining construction area into a plurality of sub-regions, to monitor the sub-regions in real time, to collect and establish geological data sets, induced disturbance data sets and disaster bearing data sets, and to pre-process the multi-source data and upload it to the center platform through the gateway; The AI ​​dynamic risk assessment model building module is used to construct a convolutional neural network model. It uses geological, induced disturbance and disaster-bearing datasets as training inputs, extracts intermediate layer feature vectors, identifies regional geological sensitivity, induced factor response and human activity carrying capacity characteristics, and realizes dynamic modeling and joint assessment of geological disaster risks in the construction area. The geological monitoring module is used to monitor the geological data, topographic data and historical annotation data of the construction sub-area in real time, calculate the geological sensitivity coefficient GDZ, and compare it with the first threshold Q1 to determine whether the geological environment stability of the construction sub-area is qualified. If it is not qualified, a strategy is given. The induced disturbance monitoring module is used to monitor the geological dynamic response data and induced factor data of the construction sub-area in real time, calculate and obtain the induced intensity coefficient YFX, and compare and analyze it with the second threshold Q2 to determine whether the intensity of the induced disaster in the construction sub-area is normal. If it is abnormal, a strategy is given. The disaster-bearing monitoring module is used to monitor the surface human activity load, building distribution, population concentration and basic evacuation capacity of the construction sub-area in real time, calculate the disaster vulnerability coefficient CZCx, and compare it with the third threshold Q3 to determine whether the disaster-bearing capacity of the construction sub-area is qualified. If it is not qualified, a strategy is given.

[0007] Preferably, the data acquisition module includes a segmentation unit, a geological data acquisition unit, an induced disturbance data acquisition unit, a disaster-bearing data acquisition unit, and a data processing connection unit; The partitioning unit is used to divide the entire mining construction area into several sub-regions of uniform area. These n sub-regions are then divided in the big data geological hazard model according to... , , ... Perform sequential marking; The geological data acquisition unit is used to collect structural geological data, slope morphology data, surface soil structure data, and surface deformation data by setting up sampling points in several sub-regions, deploying multi-band high-resolution geological radar detection equipment, shallow seismic reflectors, and drilling and geophysical exploration combined sampling technology; and to comprehensively establish a geological dataset. The induced disturbance data acquisition unit is used to collect meteorological and hydrological data, earthquake and microseismic data, and human construction activity disturbance data by deploying automatic rain gauges and connecting to the regional meteorological radar system; and to comprehensively establish an induced disturbance dataset. The disaster-bearing data acquisition unit is used to collect building structure data, population aggregation and flow data, and evacuation and accessibility data through UAV oblique photogrammetry modeling and ground LiDAR point cloud scanning system; and to comprehensively establish a disaster-bearing dataset. The data processing connection unit is used for multi-dimensional sensing equipment and data interface to transmit and cache data through an edge computing gateway EdgeGateway, data standardization and cleaning, space-time coordinate correction and timestamp alignment.

[0008] Preferably, the AI dynamic risk assessment model establishment module is used to construct a convolutional neural network initial model, train and test the convolutional neural network initial model with geological data set data, induced disturbance data set data and disaster-bearing data set data, and use the trained convolutional neural network initial model as the AI dynamic risk assessment model; meanwhile, the intermediate layer output of the multi-source data is used as a feature vector to identify information reflecting regional geological structure sensitivity, external induced factor response characteristics and human activity bearing characteristics, and the AI dynamic risk assessment model is trained and tested through the obtained feature information, and the trained AI dynamic risk assessment model is used as data operation for geological disaster risk assessment, so as to realize joint identification and dynamic assessment of field construction regional geological condition response capability, external disturbance triggering risk and construction pressure bearing situation.

[0009] Preferably, the geological monitoring module comprises a first calculation unit and a first analysis unit. The first calculation unit is used to monitor the geological data, topographic data and historical annotation data of the construction sub-region in real time, analyze the data in combination with a digital elevation model DEM, and calculate the geological sensitivity coefficient GDZ after dimensionless processing.

[0010] Preferably, the first analysis unit is used to preset a first threshold Q1 in advance, compare and analyze the geological sensitivity coefficient GDZ with the first threshold Q1, and obtain a first evaluation result, which includes: When the geological sensitivity coefficient GDZ is less than or equal to the first threshold Q1, it indicates that the geological environment stability of the construction sub-region is qualified, and the construction operation is allowed to be performed, and the monitoring is continued. When the geological sensitivity coefficient GDZ is greater than the first threshold Q1, it indicates that the geological environment stability of the construction sub-region is unqualified, a first early warning instruction is triggered, and a first strategy is generated: an induced disaster factor monitoring mechanism is started.

[0011] Preferably, the induced disturbance monitoring module comprises a second calculation unit and a second analysis unit. The second calculation unit is used to monitor the geological dynamic response data and induced factor data of the construction sub-region in real time when the first early warning instruction is received, analyze the input data through spatial projection in combination with time sequence change information, fuse the response characteristics of the induced scene after dimensionless processing, and finally calculate and obtain the induced intensity coefficient YFX.

[0012] Preferably, the second analysis unit is used to preset a second threshold Q2 in advance, and compare the induced intensity coefficient YFX with the second threshold Q2 to obtain a second evaluation result, which includes: When the induced intensity coefficient YFX is less than the second threshold Q2, it indicates that the intensity of induced disasters in the construction sub-region is normal, and continuous monitoring is performed. When the induced intensity coefficient YFX is greater than or equal to the second threshold Q2, it indicates that the intensity of induced disasters in the construction sub-region is abnormal, a second warning instruction is triggered, and a second strategy is generated: stopping construction activities, generating a disaster evaluation report, and starting a disaster-bearing capacity monitoring mechanism.

[0013] Preferably, the disaster-bearing monitoring module includes a third calculation unit and a third analysis unit. The third calculation unit is used to monitor the ground human activity load, building distribution, population gathering situation and basic evacuation capacity of the construction sub-region in real time when the second warning instruction is received, extract the building volume, accessibility and evacuation potential index in combination with the geographic information system (GIS), and calculate the disaster-bearing vulnerability coefficient CZCx after dimensionless processing.

[0014] Preferably, the third analysis unit is used to preset a third threshold Q3 in advance, and compare the disaster-bearing vulnerability coefficient CZCx with the third threshold Q3 to obtain a third evaluation result, which includes: When the disaster-bearing vulnerability coefficient CZCx is less than or equal to the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-region is qualified, and continuous monitoring is performed. When the disaster-bearing vulnerability coefficient CZCx is greater than the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-region is unqualified, a third warning instruction is triggered, and a third strategy is generated: generating a disaster-bearing data report, starting an emergency command mechanism, adding 50% emergency shelters, and implementing forced evacuation drills and emergency drill plans.

[0015] Preferably, a geological disaster risk assessment method based on big data includes the following steps: Step 1: Divide the entire mining construction area into several sub-regions, monitor the sub-regions in real time, collect and establish geological data sets, induced disturbance data sets and disaster-bearing data sets, and preprocess the multi-source data and upload it to the center platform through the gateway. Step 2: Build a convolutional neural network model, use the geological, induced disturbance and disaster-bearing data sets as training inputs, extract intermediate layer feature vectors, identify regional geological sensitivity, induced factor response and human activity bearing characteristics, and realize dynamic modeling and joint evaluation of the geological disaster risk of the construction area. Step three, real-time monitoring of the geological data, topographic data and historical annotation data of the construction sub-area, calculating the geological sensitivity coefficient GDZ, and comparing and analyzing with the first threshold Q1, judging whether the geological environment stability of the construction sub-area is qualified, if not qualified, giving strategy; Step four, real-time monitoring of the geological dynamic response data and inducing factor data of the construction sub-area, calculating the inducing intensity coefficient YFX, and comparing and analyzing with the second threshold Q2, judging whether the intensity of induced disaster of the construction sub-area is normal, if abnormal, giving strategy; Step five, real-time monitoring of the surface human activity load, building distribution, population gathering situation and basic evacuation capacity of the construction sub-area, calculating the disaster-bearing vulnerability coefficient CZCx, and comparing and analyzing with the third threshold Q3, judging whether the disaster-bearing capacity of the construction sub-area is qualified, if not qualified, giving strategy.

[0016] The application provides a geological disaster risk assessment platform and method based on big data. (1) The geological disaster risk assessment platform and method based on big data, by dividing the entire mining construction area into multiple sub-areas, using high-resolution geological radar, InSAR remote sensing, rainfall station, unmanned aerial vehicle oblique photography, LiDAR point cloud and other sensing means, respectively constructing geological data set, inducing disturbance data set and disaster-bearing data set, and with the help of edge computing gateway for data standardization, cleaning and uploading, significantly improving the spatial resolution and real-time of data acquisition, providing high-quality data support for risk assessment.

[0017] (2) The geological disaster risk assessment platform and method based on big data, using convolutional neural network CNN to establish a dynamic risk assessment model, by introducing three types of intermediate layer feature vectors of geological sensitivity, inducing factor response characteristics and human activity bearing characteristics, realizing the extraction of potential disaster mode from multi-dimensional, multi-period and multi-type data, improving the recognition ability of complex disaster evolution mechanism, and significantly enhancing the generalization performance and dynamic prediction ability of the model.

[0018] (3) The geological disaster risk assessment platform and method based on big data, by calculating the geological sensitivity coefficient GDZ, the inducing intensity coefficient YFX and the disaster-bearing vulnerability coefficient CZCx, and comparing with the first threshold Q1, the second threshold Q2 and the third threshold Q3 respectively, realizing the layer-by-layer judgment and linkage strategy triggering mechanism of "geological stability determination-inducing risk identification-disaster-bearing capacity assessment", which can automatically generate early warning response and emergency strategy according to real-time risk state, improving the decision efficiency and on-site adaptability.

[0019] (4) The geological disaster risk assessment platform and method based on big data can pass through the whole-link response path from the field perception layer, the intelligent calculation layer to the emergency command layer, can generate a risk report, a shutdown suggestion, an emergency drill starting instruction and an emergency drill starting instruction in real time when there is a geological anomaly in the construction area, the inducing intensity is enhanced or the disaster bearing capacity is insufficient, has good engineering adaptability and disaster response capability, and is particularly suitable for high-risk field engineering areas such as mines, traffic and water conservancy. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The geological disaster risk assessment platform based on big data is a block diagram flowchart. Figure 2 The geological disaster risk assessment method based on big data is a step schematic diagram. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0022] Embodiment 1 Please refer to Figure 1 The present application provides a geological disaster risk assessment platform based on big data, comprising a data acquisition module, an AI dynamic risk assessment model establishment module, a geological monitoring module, an inducing disturbance monitoring module and a disaster bearing monitoring module. The data acquisition module is used for dividing the whole mining construction area into a plurality of sub-regions, monitoring the sub-regions in real time, collecting and establishing geological data sets, inducing disturbance data sets and disaster bearing data sets, and pre-processing multi-source data and uploading to the center platform through a gateway. The AI dynamic risk assessment model establishment module is used for constructing a convolutional neural network model, taking the geological, inducing disturbance and disaster bearing data sets as training inputs, extracting intermediate layer feature vectors, identifying regional geological sensitivity, inducing factor response and human activity bearing characteristics, and realizing dynamic modeling and joint evaluation of the geological disaster risk of the construction area. The geological monitoring module is used for monitoring the geological data, topographic data and historical labeled data of the construction sub-region in real time, calculating and obtaining a geological sensitivity coefficient GDZ, and comparing and analyzing the geological sensitivity coefficient GDZ with a first threshold Q1 to judge whether the geological environment stability of the construction sub-region is qualified, and giving a strategy if not qualified. The induced disturbance monitoring module is used for monitoring the geological dynamic response data and induced factor data of the construction sub-region in real time, calculating and obtaining an induced intensity coefficient YFX, and comparing and analyzing the induced intensity coefficient YFX with a second threshold Q2 to determine whether the intensity of the induced disaster of the construction sub-region is normal, and if not, a strategy is given; The disaster-bearing monitoring module is used for monitoring the surface human activity load, building distribution, population gathering situation and basic evacuation capacity of the construction sub-region in real time, calculating and obtaining a disaster-bearing vulnerability coefficient CZCx, and comparing and analyzing the disaster-bearing vulnerability coefficient CZCx with a third threshold Q3 to determine whether the disaster-bearing capacity of the construction sub-region is qualified, and if not, a strategy is given.

[0023] In this embodiment, the geological disaster risk assessment platform and method based on big data provided by the present application can realize fusion perception and intelligent processing of multi-source heterogeneous data, divide the construction area into multiple sub-regions and perform real-time monitoring, dynamically construct three types of key data sets of geology, induced disturbance and disaster-bearing through a data acquisition module, and perform joint modeling analysis of geological sensitivity, induced factor response intensity and human activity disaster-bearing capacity of each sub-region based on an AI dynamic risk assessment model of a convolutional neural network to output a multi-dimensional risk feature vector. Compared with the traditional assessment method mainly based on field investigation and static model, the present application has the following advantages: comprehensive monitoring dimension, intelligent evaluation process, real-time response mechanism and high application scene matching degree.

[0024] Embodiment 2 This embodiment is an explanation and description in embodiment 1. Specifically, the data acquisition module includes a division unit, a geological data acquisition unit, an induced disturbance data acquisition unit, a disaster-bearing data acquisition unit and a data processing connection unit. The division unit is used for dividing the entire mining construction area into a plurality of sub-regions with uniform areas, and the n sub-regions are sequentially marked in the big data geological disaster model according to , , ,..., . The geological data acquisition unit is used for setting sampling points in the plurality of sub-regions, deploying multi-band high-resolution geological radar detection equipment, shallow seismic reflection instrument and drilling geophysical joint sampling technology, acquiring structural geological data including stratum lithology data, fracture zone spatial density distribution information and rock layer thickness joint strike information, acquiring slope shape data including slope distribution information, slope direction information and slope length information, acquiring surface soil structure data including soil surface and middle layer water content profile information and soil type and particle structure characteristic spectrum information, acquiring surface deformation data including InSAR remote sensing deformation sequence and historical disaster event frequency data, and comprehensively establishing a geological data set. The induced disturbance data acquisition unit is used to acquire meteorological and hydrological data including real-time rainfall intensity, 24-hour cumulative rainfall, and 72-hour cumulative rainfall by deploying automatic rain gauges and accessing regional weather radar systems; to acquire earthquake and microseismic data including microseismic wave spectrum data and information on energy release of the seismic source and epicenter distribution; and to acquire human construction activity disturbance data including disturbance intensity generated by construction, operation period, operation frequency, operation type label information, and temporary traffic load and material stacking density map information; and to comprehensively establish an induced disturbance data set. The disaster-affected data acquisition unit is used to acquire building structure data including total building volume, building height, land area, and density distribution map information in a sub-region by using unmanned aerial vehicle oblique photogrammetry modeling and ground LiDAR point cloud scanning systems; to acquire population aggregation flow data including population density per unit area and crowd flow trend map and residence duration matrix; to acquire evacuation accessibility data including accessible evacuation area, average evacuation response time, and information on the number, capacity, and layout of shelters; and to comprehensively establish a disaster-affected data set. The data processing connection unit is used to connect multi-dimensional sensing devices and data interfaces through an edge computing gateway EdgeGateway and a central data fusion hub transmission and caching mechanism, data standardization and cleaning, space-time coordinate correction, and time stamp alignment.

[0025] In this embodiment, by setting the division unit, the geological data acquisition unit, the induced disturbance data acquisition unit, the disaster-affected data acquisition unit, and the data processing connection unit, a multi-dimensional data acquisition system covering comprehensively, being fine in layers, and responding efficiently is constructed, which can realize the automatic acquisition and high-quality modeling of multi-source, multi-scale, and multi-dimensional geological disaster related data in the mining construction area. The system has high sub-region division precision: the entire construction area is refined into n equal-area sub-regions by the division unit, so that the risk assessment has a spatial resolution basis, facilitating differentiated monitoring and positioning management; the geological information acquisition system is systematic: high-resolution radar, multi-channel seismic reflection instrument, and joint sampling technology are used to acquire data such as structural geology, slope morphology, soil structure, and deformation history, improving the accuracy and physical basis completeness of geological information modeling; the induced disturbance monitoring is intelligent: automatic rain gauges, regional weather radars, and microseismic monitoring equipment are integrated to realize comprehensive monitoring of weather, earthquakes, and construction activities, which can effectively identify external driving factors of induced geological disasters and enhance dynamic response capability; the disaster-affected data modeling is multi-dimensional: tilt photography and LiDAR point cloud modeling technology are used to accurately obtain key indicators such as buildings, crowds, and evacuation accessibility, enhancing the fine level of disaster-affected capacity evaluation; the data processing is efficient: the edge computing gateway EdgeGateway is used to realize fast preprocessing of on-site data, unified standard conversion, and coordinate time sequence alignment, ensuring the consistency and fusibility of uploaded data, and providing high-quality data support for subsequent AI modeling and risk analysis.

[0026] Embodiment 3 This embodiment is an explanation and illustration in embodiment 2, specifically, the AI dynamic risk assessment model establishment module is used to construct a convolutional neural network initial model by using a convolutional neural network, and train and test the convolutional neural network initial model with geological data set data, induced disturbance data set data and disaster-bearing data set data, and use the trained convolutional neural network initial model as the AI dynamic risk assessment model; at the same time, use the intermediate layer output of the multi-source data as a feature vector to identify information reflecting the sensitivity of regional geological structure, the response characteristics of external induced factors and the bearing characteristics of human activities, and train and test the AI dynamic risk assessment model through the obtained feature information, and use the trained AI dynamic risk assessment model as the data operation of geological disaster risk assessment, to realize the joint identification and dynamic assessment of the response ability of the field construction area geological conditions, the external disturbance trigger risk and the construction pressure situation.

[0027] In this embodiment, by constructing an AI dynamic risk assessment model based on convolutional neural network CNN, the joint training and testing are carried out by fusing geological data set, induced disturbance data set and disaster-bearing data set, which has the following remarkable beneficial effects: strong model structure adaptability: the convolutional neural network has excellent feature automatic extraction capability, which can adaptively learn the spatial structure information and nonlinear correlation characteristics in the multi-source input data, avoiding the subjectivity of artificial setting weight and rule in traditional risk assessment; multi-dimensional feature fusion precision: by extracting the intermediate layer feature vector of geological structure sensitivity, induced factor response characteristics and human activity bearing characteristics, the model realizes the deep fusion and semantic expression of multi-source heterogeneous data, improves the explanation power and identification accuracy of disaster mechanism; strong dynamic modeling capability: the trained AI model can realize real-time updating and dynamic assessment of the risk state of different sub-regions of the construction area, effectively making up for the problem that the static model cannot respond to the rapid changes of the environment; strong intelligent decision support: the model output can be used to identify the geological condition fragile area, the external induced high response area and the human disaster-bearing high risk area, to provide intelligent support for construction strategy optimization, emergency resource deployment and early warning decision; suitable for complex environment scene: through the deep learning framework, multi-time and space scale data are included in the same evaluation system, which is suitable for disaster risk prediction in complex geological environment and high dynamic construction scene in the field, and has good generalization and engineering adaptability.

[0028] Embodiment 4 This embodiment is an explanation and illustration in embodiment 3, specifically, the geological monitoring module comprises a first calculation unit and a first analysis unit; The first calculation unit is used for monitoring geological data, terrain data and historical annotation data of the construction sub-area in real time, combining with digital elevation model DEM analysis, after dimensionless processing, the geological sensitivity coefficient GDZ is calculated and obtained, and the formula is as follows:

[0029] In the formula, represents the stratum fragmentation index, represents the local slope index, which is obtained by DEM gradient analysis, represents the soil water content index, represents the InSAR deformation variable normalized score, represents the historical disaster event frequency, w1, w2, w3, w4 and w5 represent weight coefficients, , , , , , and ; ; In the formula, represents the stratum fracture zone density, represents the stratum lithology type number, represents the total stratum lithology number, w6 and w7 represent weight coefficients, , , and ; ; In the formula, represents the real-time soil water content, represents the reference soil water content, represents the soil saturated water content; ; In the formula, Vd represents the vertical or horizontal deformation velocity of the construction sub-area surface, represents the minimum value of the historical statistics of the vertical or horizontal deformation velocity of the construction sub-area surface, represents the maximum value of the historical statistics of the vertical or horizontal deformation velocity of the construction sub-area surface.

[0030] In this embodiment, by setting up a first calculation unit and a first analysis unit in the geological monitoring module, and combining the digital elevation model (DEM) with multi-source geological data, a multi-factor calculation model for the geological sensitivity coefficient (GDZ) is constructed. This model has the following significant advantages: Comprehensive multi-source factor fusion: The GDZ coefficient simultaneously integrates five core geological risk factors: stratigraphic fracturing degree, slope characteristics, soil moisture changes, surface deformation, and frequency of historical disaster events. It comprehensively reflects the geological environmental sensitivity of the construction sub-area through weighted superposition, achieving systematic and complete assessment dimensions; Scientific and quantifiable calculation logic: Each sub-factor, such as the stratigraphic fracturing index, local slope index, soil moisture content index, and InSAR deformation score, is based on... Standardized processing and clear formula derivation avoid the subjectivity problems of traditional experience-based scoring methods, improving the objectivity and repeatability of model evaluation results; responsive to real-time geological conditions: by real-time access to geological sensors and remote sensing data such as InSAR deformation sequences in the construction area, the GDZ coefficient value can be dynamically updated, achieving rapid response and accurate reflection of geological environment changes, providing real-time decision support for construction scheduling and risk early warning; enhanced geological environment stability assessment capabilities: the established GDZ calculation system can be used to determine whether the geological stability of the construction sub-area is within the early warning threshold range, assisting in determining whether it is necessary to adjust the construction intensity, deploy support, or implement preventive engineering control measures, thereby improving the level of on-site geological disaster prevention and control.

[0031] Example 5 This embodiment is an explanation of Embodiment 4. Specifically, the first analysis unit is used to preset a first threshold Q1 and compare the geological sensitivity coefficient GDZ with the first threshold Q1 to obtain a first evaluation result, including: When the geological sensitivity coefficient GDZ ≤ the first threshold Q1, it indicates that the geological environment stability of the construction sub-area is qualified, and construction operations are allowed, with continuous monitoring. When the geological sensitivity coefficient GDZ > the first threshold Q1, it indicates that the geological environment stability of the construction sub-area is unqualified, triggering the first early warning instruction and generating the first strategy: activating the disaster-inducing factor monitoring mechanism.

[0032] In this embodiment, by comparing the geological sensitivity coefficient GDZ with a preset first threshold Q1 through the first analysis unit, the system can automatically identify and intelligently determine the geological environment stability of the construction sub-area. When GDZ does not exceed Q1, the system automatically allows construction and continuously monitors. When GDZ exceeds Q1, the system immediately triggers the first warning and generates the first strategy to activate the monitoring mechanism for induced disaster factors. This effectively improves the timeliness of geological disaster risk identification and the level of intelligent response, reduces potential safety hazards caused by delays in human intervention, and enhances the dynamic risk perception and proactive prevention and control capabilities of the construction area. Specific implementation data are shown in the table below:

[0033] Embodiment 6 This embodiment is an explanation and illustration in embodiment 5, specifically, the induced disturbance monitoring module includes a second calculation unit and a second analysis unit; The second calculation unit is used to monitor the geological dynamic response data and induced factor data of the construction sub-area in real time when receiving the first early warning instruction, combine the time sequence change information, perform spatial projection analysis on the input data, after dimensionless processing, fuse the response characteristics of the induced scene, and finally calculate and obtain the induced intensity coefficient YFX, the formula is as follows:

[0034] In the formula, represents the rainfall rate at time t, represents the rainfall sensitive weight value, represents the seismic wave spectrum index score at time t, represents the disturbance intensity generated by construction at x time, represents the disturbance time decay coefficient, represents the disturbance cumulative integral term, the contribution of historical construction disturbance from the starting time t0 to t to the induced intensity; represents the rainfall influence adjustment factor, represents the adjustment microseismic influence amplitude factor, represents the human activity induced factor factor, , , and , represents the base of the natural constant;

[0035] In the formula, N represents the number of sensing channels, represents the instantaneous amplitude collected by the i th seismic source channel, represents the frequency component corresponding to the instantaneous amplitude collected by the i th seismic source channel.

[0036] In this embodiment, the second calculation unit fuses real-time geological dynamic response data and induced factor data after receiving the first early warning instruction, combines time sequence change information and spatial projection analysis, and calculates the induced intensity coefficient YFX, effectively realizing comprehensive perception and intensity evaluation of multiple external causes such as rainfall, earthquake and construction disturbance, significantly improving the description accuracy of the system to the dynamic evolution of the induced scene, ensuring that the system quickly enters the deep monitoring state when the geological environment is unstable, and enhancing the intelligent identification and risk warning ability of the platform to the disaster triggering mechanism under the combined action of complex causes.

[0037] Embodiment 7 The embodiment is an explanation of embodiment 6. Specifically, the second analysis unit is used to preset a second threshold Q2 in advance, and the induced intensity coefficient YFX is compared and analyzed with the second threshold Q2 to obtain a second evaluation result, which includes: When the induced intensity coefficient YFX is less than the second threshold Q2, it indicates that the intensity of the induced disaster in the construction sub-region is normal, and continuous monitoring is performed. When the induced intensity coefficient YFX is greater than or equal to the second threshold Q2, it indicates that the intensity of the induced disaster in the construction sub-region is abnormal, a second warning instruction is triggered, a second strategy is generated: stopping construction activities, generating a disaster evaluation report, and starting a disaster-bearing capacity monitoring mechanism.

[0038] In the embodiment, the second analysis unit compares and analyzes the induced intensity coefficient YFX with the preset second threshold Q2 in real time, which can accurately determine the intensity state of the induced disaster in the construction sub-region. When the induced intensity is abnormal, the warning is triggered in time and the stop work instruction and the disaster evaluation report are automatically generated, and the disaster-bearing capacity monitoring mechanism is started. The intelligent and dynamic management of construction safety control is realized, the geological disaster induced risk is effectively reduced, and the construction safety and personnel and property safety are protected. The specific embodiment data is as follows:

[0039] Embodiment 8 The embodiment is an explanation of embodiment 7. Specifically, the disaster-bearing monitoring module includes a third calculation unit and a third analysis unit. The third calculation unit is used to monitor the ground human activity load, building distribution, population gathering situation and basic evacuation capacity of the construction sub-region in real time when the second warning instruction is received, extract the building volume, accessibility and evacuation potential index combined with the geographic information system GIS, and calculate the disaster-bearing vulnerability coefficient CZCx after dimensionless processing, as follows:

[0040] In the formula, represents the total volume of buildings in the sub-region, represents the maximum building density in the sub-region, represents the real-time population density, represents the maximum historical population density, represents the evacuation accessible area, represents the ideal evacuation coverage area, represents the average evacuation response time, represents a constant to prevent the denominator from being zero, , , and represent weight coefficients, , , , , and .

[0041] In this embodiment, by combining GIS technology through the third computing unit, the human activity load, building distribution, population gathering situation and evacuation capacity in the construction sub-region are monitored and analyzed in real time, the disaster-bearing vulnerability coefficient CZCx is calculated, the regional disaster-bearing risk level can be scientifically quantified, the vulnerability assessment basis for construction safety management is provided, high-risk areas are discovered and warned in time, and thus the geological disaster emergency response capability and personnel evacuation efficiency are effectively improved, and the safety and stability of the construction site and the surrounding area are ensured.

[0042] Embodiment 9 This embodiment is an explanation and description in embodiment 8. Specifically, the third analysis unit is used to preset a third threshold Q3 in advance, and the disaster-bearing vulnerability coefficient CZCx is compared and analyzed with the third threshold Q3 to obtain a third evaluation result, including: When the disaster-bearing vulnerability coefficient CZCx is less than or equal to the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-region is qualified, and the monitoring is continued. When the disaster-bearing vulnerability coefficient CZCx is greater than the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-region is unqualified, a third warning instruction is triggered, a third strategy is generated: a disaster-bearing data report is generated, an emergency command mechanism is started, 50% of emergency shelters are added, and forced evacuation drills and emergency drill plans are started.

[0043] In this embodiment, by comparing the disaster-bearing vulnerability coefficient CZCx with the preset third threshold Q3, the disaster-bearing capacity status of the construction sub-region can be accurately determined, the warning and emergency response measures such as adding emergency shelters and starting forced evacuation drills can be triggered in time, the regional emergency management level and personnel safety protection capability are effectively improved, the risk of personnel casualties and property losses caused by geological disasters is significantly reduced, and specific embodiment data is as follows:

[0044] Embodiment 10 A geological disaster risk assessment method based on big data, please refer to Figure 2 , including the following steps: Step 1: divide the entire mining construction area into several sub-regions, monitor the sub-regions in real time, collect and establish geological data sets, induced disturbance data sets and disaster-bearing data sets; and preprocess the multi-source data, upload to the center platform through the gateway; Step two, a convolutional neural network model is constructed to extract the intermediate layer feature vector, identify the regional geological sensitivity, induced factor response and human activity bearing characteristics, realize the dynamic modeling and joint evaluation of the geological disaster risk of the construction area, with the geological, induced disturbance and disaster bearing data set as the training input; Step three, real-time monitoring of the geological data, topographic data and historical labeled data of the construction sub-area, calculating the geological sensitivity coefficient GDZ and comparing it with the first threshold Q1 to determine whether the geological environment stability of the construction sub-area is qualified, if not, a strategy is given; Step four, real-time monitoring of the geological dynamic response data and induced factor data of the construction sub-area, calculating the induced intensity coefficient YFX and comparing it with the second threshold Q2 to determine whether the intensity of the induced disaster of the construction sub-area is normal, if not, a strategy is given; Step five, real-time monitoring of the surface human activity load, building distribution, population gathering situation and basic evacuation capacity of the construction sub-area, calculating the disaster bearing vulnerability coefficient CZCx and comparing it with the third threshold Q3 to determine whether the disaster bearing capacity of the construction sub-area is qualified, if not, a strategy is given.

[0045] In this embodiment, based on steps one to five, the multi-dimensional dynamic monitoring and evaluation of the geological environment, induced disturbance and disaster bearing capacity of the mining construction area is realized, through big data collection, multi-source heterogeneous data fusion and joint analysis of the convolutional neural network model, the geological disaster risk can be accurately identified, potential hidden dangers can be found in time and corresponding strategies can be automatically triggered, thereby the accuracy and response speed of geological disaster warning are significantly improved, the construction safety is ensured, the personnel and property losses are reduced, and the safe and sustainable development of mining engineering is promoted.

[0046] The size of the threshold is set for easy comparison, and the size of the threshold depends on the number of sample data and the base number set by the person skilled in the art for each group of sample data; as long as it does not affect the proportional relationship of the parameters and the quantized values.

[0047] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value, and the coefficients in the formula are set by the person skilled in the art according to the actual situation, the above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art in the technical field according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, equivalent replacement or change, should be covered within the protection scope of the present application.

Claims

1. A geological hazard risk assessment platform based on big data, characterized in that, include: The system includes a data acquisition module, an AI dynamic risk assessment model establishment module, a geological monitoring module, an induced disturbance monitoring module, and a disaster-bearing monitoring module. The data acquisition module is used to divide the entire mining construction area into several sub-areas, monitor the sub-areas in real time, collect and establish geological datasets, induced disturbance datasets, and disaster-bearing datasets; and preprocess the multi-source data and upload it to the central platform through the gateway. The AI ​​dynamic risk assessment model building module is used to construct a convolutional neural network model. It uses geological, induced disturbance and disaster-bearing datasets as training inputs, extracts intermediate layer feature vectors, identifies regional geological sensitivity, induced factor response and human activity carrying capacity characteristics, and realizes dynamic modeling and joint assessment of geological disaster risks in the construction area. The geological monitoring module is used to monitor the geological data, topographic data and historical annotation data of the construction sub-area in real time, calculate the geological sensitivity coefficient GDZ, and compare it with the first threshold Q1 to determine whether the geological environment stability of the construction sub-area is qualified. If it is not qualified, a strategy is given. The induced disturbance monitoring module is used to monitor the geological dynamic response data and induced factor data of the construction sub-area in real time, calculate and obtain the induced intensity coefficient YFX, and compare and analyze it with the second threshold Q2 to determine whether the intensity of the induced disaster in the construction sub-area is normal. If it is abnormal, a strategy is given. The disaster-bearing monitoring module is used to monitor the surface human activity load, building distribution, population concentration and basic evacuation capacity of the construction sub-area in real time, calculate the disaster vulnerability coefficient CZCx, and compare it with the third threshold Q3 to determine whether the disaster-bearing capacity of the construction sub-area is qualified. If it is not qualified, a strategy is given.

2. The geological disaster risk assessment platform based on big data according to claim 1, characterized in that, The data acquisition module includes a segmentation unit, a geological data acquisition unit, an induced disturbance data acquisition unit, a disaster-bearing data acquisition unit, and a data processing connection unit. The partitioning unit is used to divide the entire mining construction area into several sub-regions of uniform area. These n sub-regions are then divided in the big data geological hazard model according to... , , ... Perform sequential marking; The geological data acquisition unit is used to collect structural geological data, slope morphology data, surface soil structure data, and surface deformation data by setting up sampling points in several sub-regions, deploying multi-band high-resolution geological radar detection equipment, shallow seismic reflectors, and drilling and geophysical exploration combined sampling technology; and to comprehensively establish a geological dataset. The induced disturbance data acquisition unit is used to collect meteorological and hydrological data, earthquake and microseismic data, and human construction activity disturbance data by deploying automatic rain gauges and connecting to the regional meteorological radar system; and to comprehensively establish an induced disturbance dataset. The disaster-bearing data acquisition unit is used to collect building structure data, population aggregation and flow data, and evacuation and accessibility data through UAV oblique photogrammetry modeling and ground LiDAR point cloud scanning system; and to comprehensively establish a disaster-bearing dataset. The data processing connection unit is used for the transmission and caching mechanism between multi-dimensional sensing devices and data interfaces through the EdgeGateway and the central data fusion hub, as well as data standardization and cleaning, spatiotemporal coordinate correction, and timestamp alignment.

3. The geological disaster risk assessment platform based on big data according to claim 2, characterized in that, The AI ​​dynamic risk assessment model building module utilizes a convolutional neural network (CNN) to construct an initial CNN model. This initial model is then trained and tested using geological datasets, induced disturbance datasets, and disaster-bearing datasets. The trained CNN model serves as the AI ​​dynamic risk assessment model. Simultaneously, the intermediate layer outputs from the multi-source data are used as feature vectors to identify information reflecting the sensitivity of regional geological structures, response characteristics to external induced factors, and the carrying capacity of human activities. This acquired feature information is then used to train and test the AI ​​dynamic risk assessment model. The trained AI dynamic risk assessment model is then used as data for geological disaster risk assessment, enabling the joint identification and dynamic assessment of the geological condition response capability, external disturbance triggering risk, and construction pressure status of the field construction area.

4. The geological disaster risk assessment platform based on big data according to claim 3, characterized in that, The geological monitoring module includes a first calculation unit and a first analysis unit; The first calculation unit is used to monitor the geological data, topographic data and historical annotation data of the construction sub-area in real time. Combined with the analysis of the digital elevation model (DEM), after dimensionless processing, the geological sensitivity coefficient (GDZ) is calculated and obtained.

5. A geological disaster risk assessment platform based on big data according to claim 4, characterized in that, The first analysis unit is used to preset a first threshold Q1 and compare the geological sensitivity coefficient GDZ with the first threshold Q1 to obtain a first evaluation result, including: When the geological sensitivity coefficient GDZ ≤ the first threshold Q1, it indicates that the geological environment stability of the construction sub-area is qualified, and construction operations are allowed, with continuous monitoring. When the geological sensitivity coefficient GDZ > the first threshold Q1, it indicates that the geological environment stability of the construction sub-area is unqualified, triggering the first early warning instruction and generating the first strategy: activating the disaster-inducing factor monitoring mechanism.

6. The geological disaster risk assessment platform based on big data according to claim 5, characterized in that, The induced disturbance monitoring module includes a second calculation unit and a second analysis unit; The second calculation unit is used to monitor the geological dynamic response data and inducing factor data of the construction sub-area in real time when the first early warning instruction is received. Combined with the time series change information, the input data is subjected to spatial projection analysis. After dimensionless processing, the response characteristics of the induced scenario are fused, and finally the induced intensity coefficient YFX is calculated and obtained.

7. A geological disaster risk assessment platform based on big data according to claim 6, characterized in that, The second analysis unit is used to pre-set a second threshold Q2 and compare the induced intensity coefficient YFX with the second threshold Q2 to obtain a second evaluation result, including: When the induced intensity coefficient YFX < the second threshold Q2, it indicates that the intensity of the induced disaster in the construction sub-area is normal, and continuous monitoring is required. When the induced intensity coefficient YFX ≥ the second threshold Q2, it indicates that the intensity of the induced disaster in the construction sub-area is abnormal, triggering the second early warning instruction and generating the second strategy: stop construction activities, generate a disaster assessment report, and activate the disaster bearing capacity monitoring mechanism.

8. A geological disaster risk assessment platform based on big data according to claim 7, characterized in that, The disaster monitoring module includes a third calculation unit and a third analysis unit; The third calculation unit is used to monitor the surface human activity load, building distribution, population concentration and basic evacuation capacity of the construction sub-area in real time when the second early warning instruction is received. Combined with the geographic information system (GIS), it extracts the indicators of building volume, accessibility and evacuation potential. After dimensionless processing, it calculates and obtains the disaster vulnerability coefficient CZCx.

9. A geological hazard risk assessment platform based on big data according to claim 8, characterized in that, The third analysis unit is used to pre-set a third threshold Q3, and compare the disaster vulnerability coefficient CZCx with the third threshold Q3 to obtain the third assessment result, including: When the disaster vulnerability coefficient CZCx ≤ the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-area is qualified and will be continuously monitored. When the disaster vulnerability coefficient CZCx > the third threshold Q3, it indicates that the disaster-bearing capacity of the construction sub-area is unqualified, triggering the third early warning instruction and generating the third strategy: generating a disaster data report, activating the emergency command mechanism, adding 50% of emergency refuge points, and implementing mandatory evacuation drills and emergency drill plans.

10. A method for geological hazard risk assessment based on big data, comprising the geological hazard risk assessment platform based on big data as described in any one of claims 1-9, comprising the following steps: Step 1: Divide the entire mining construction area into several sub-areas, monitor the sub-areas in real time, collect and establish geological datasets, induced disturbance datasets, and disaster-bearing datasets; preprocess the multi-source data and upload it to the central platform through the gateway; Step 2: Construct a convolutional neural network model, using geological, induced disturbance, and disaster-bearing datasets as training inputs, extract intermediate layer feature vectors, identify regional geological sensitivity, induced factor response, and human activity carrying capacity characteristics, and realize dynamic modeling and joint assessment of geological disaster risks in the construction area; Step 3: Monitor the geological data, topographic data and historical annotation data of the construction sub-area in real time, calculate the geological sensitivity coefficient GDZ, and compare it with the first threshold Q1 to determine whether the geological environment stability of the construction sub-area is qualified. If it is not qualified, a strategy is given. Step 4: Monitor the geological dynamic response data and inducing factor data of the construction sub-area in real time, calculate and obtain the inducing intensity coefficient YFX, and compare and analyze it with the second threshold Q2 to determine whether the intensity of the induced disaster in the construction sub-area is normal. If it is abnormal, a strategy is given. Step 5: Monitor the surface human activity load, building distribution, population concentration and basic evacuation capacity of the construction sub-area in real time, calculate the disaster vulnerability coefficient CZCx, and compare it with the third threshold Q3 to determine whether the disaster capacity of the construction sub-area is qualified. If it is not qualified, a strategy is given.

Images