Dangerous zoning method for active fault engineering influence

Abstract

The invention discloses a risk zoning method for active fault engineering influence, which relates to the technical field of engineering construction evaluation, and comprises the following steps: I, fusing multi-source data of a detection fault, constructing a three-dimensional geometric model of the fault, and analyzing and correcting a fault distribution rule by utilizing a GIS (Geographic Information System) space; iI, combining fault sliding rate, paleoearthquake recurrence period and deep engineering crustal stress inversion, and evaluating activity characteristics of the hidden fracture in the engineering service period. Through multi-source data fusion, three-dimensional model construction and GIS correction, multi-factor overlay analysis and a dynamic correction mechanism, multivariate characteristics of the active fault and complex geological conditions of an engineering site are comprehensively considered, the problem of limitation of a zoning result is solved, the accuracy of risk assessment is improved, limitation of a single index is broken through, and the risk assessment efficiency is improved. Closed-loop management of data acquisition-risk assessment-engineering response is realized, and scientific and reliable technical support is provided for safe site selection, optimization design and operation and maintenance of major infrastructures.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering construction evaluation, in particular to a method for partitioning the risk of the engineering influence of active faults. BACKGROUND

[0002] An active fault refers to a fault structure that may have significant displacement within the service life of an engineering project. Its engineering influence involves the safe site selection and design of major infrastructure. As an active zone of crustal movement, active faults can cause intense ground shaking, surface rupture, and a series of secondary geological disasters such as sand liquefaction and landslides when geological events such as earthquakes occur, posing a serious threat to nearby engineering facilities. In the past, a series of methods for partitioning the risk of the engineering influence of active faults have been developed.

[0003] However, existing methods are based on single-factor partitioning methods, such as considering only the peak ground acceleration of seismic motion or simply partitioning based on the distance between the fault and the engineering site. Although these methods improve the accuracy of partitioning to some extent, they fail to comprehensively consider the multi-element characteristics of active faults and the complex geological conditions of engineering sites, resulting in significant limitations in the partitioning results and an inability to accurately reflect the true risk conditions in different regions. SUMMARY

[0004] The purpose of the present application is to overcome the above-mentioned deficiencies and provide a method for partitioning the risk of the engineering influence of active faults, in order to solve the problem that existing partitioning methods fail to comprehensively consider the multi-element characteristics of active faults and the complex geological conditions of engineering sites, resulting in significant limitations in the partitioning results and an inability to accurately reflect the true risk conditions in different regions.

[0005] To solve the above technical problems, the technical solution adopted by the present application is as follows: a method for partitioning the risk of the engineering influence of active faults, comprising the following steps: Step I: fuse multi-source data of detected faults, construct a three-dimensional geometric model of the fault, and use GIS spatial analysis to correct the fault distribution law; Step II: combine the fault slip rate, paleoseismic recurrence period, and deep engineering stress inversion to evaluate the activity characteristics of hidden faults during the service life of the engineering project; Step III: establish a fault zone partitioning evaluation model based on fault data, combine the thickness of the overburden layer, soft soil / bedrock type, and liquefaction potential, and use the site amplification coefficient to adjust the seismic motion parameters to identify high amplification areas; Step IV: construct a near-fault seismic motion pulse model to perform fine analysis of near-fault seismic motion parameters and evaluate the seismic risk probability; Step V: evaluate the dynamic response and stability of hydraulic structures; Step VI, the fault avoidance zone, high PGA area, site amplification area and secondary disaster potential area are analyzed by GIS overlay analysis to generate a comprehensive risk zoning map, and the zoning boundary and risk level are dynamically corrected according to long-term monitoring data; Step VII, real-time monitoring and early warning of fault activity and ground motion, combined with LSTM machine learning model to predict fault activation trend and optimize emergency plan.

[0006] Preferably, in step I, the multi-source data for detecting faults are fused, and the shallow fault fracture zone is identified by high-density resistivity method, which uses electrode array inversion formula as follows: ; In the formula: is apparent resistivity, is electrode spacing, is potential difference, is current, is device coefficient.

[0007] Preferably, in step I, the construction of fault three-dimensional geometric model and the correction of fault distribution law by GIS spatial analysis include the following steps: A1, integrate multi-source data and preprocess the data; A2, draw the fault surface trace in GIS to generate a line feature layer, and establish a three-dimensional geometric model of the fault based on seismic reflection profile and drilling data; A3, superimpose the surface trace, geophysical anomaly area and deep fault model to identify spatial inconsistent areas, assign weights according to data confidence, calculate the optimal fault position, and generate a buffer zone with the deep fault projection as the center, and adjust the fault trace to the overlapping area; A4, divide the fault into several segments, and count the strike and dip angle changes of each segment to analyze whether it conforms to the regional tectonic stress field direction.

[0008] More preferably, in step A1, the preprocessing includes unifying the coordinate system, aligning different resolution data through interpolation, and constructing a fault attribute table.

[0009] More preferably, in step A4, the fault is divided into several segments, each 5km.

[0010] Preferably, in step II, the fault slip rate uses displacement-time probability model to calculate the fault slip probability in T years of engineering service period: ; In the formula: is the slip rate.

[0011] Preferably, in step II, the deep geostress inversion is based on drilling hydraulic fracturing data, and the horizontal principal stress is calculated : ; wherein: is the fracture pressure, is the rock tensile strength, is the Poisson's ratio, is the vertical stress.

[0012] Preferably, in step III, the fault data establishes the division standard of the core broken zone in the fault zone division evaluation model as: rock quality designation RQD: ; The permeability coefficient is calculated by Darcy's law: ; wherein: Q is the flow rate, L is the sample length, A is the cross-sectional area, is the water head difference.

[0013] Preferably, in step IV, the evaluation of the seismic risk probability comprises the following steps: B1, embedding the pulse model into the probabilistic seismic risk analysis framework to evaluate the exceeding probability of the engineering site seismic parameters; B2, near-fault correction is modified by directionality and slippage effect; B3, the exceeding probability within the engineering service period T is calculated, and a seismic risk report is output.

[0014] Preferably, in step V, the evaluation of the dynamic response and stability of the hydraulic structure comprises the following steps: C1, inputting the seismic parameters, fault displacement parameters and structure parameters into the finite element model to prepare for dynamic response analysis; C2, performing time history analysis under the action of seismic waves to output displacement response, stress distribution and acceleration response; C3, performing quasi-static analysis under the action of fault displacement.

[0015] Preferably, in step VI, the GIS superposition analysis comprises the following steps: D1, weight distribution is performed on the fault avoidance zone, high PGA area, site amplification area and secondary disaster potential area; D2, comprehensive risk value calculation is performed, and grade division is performed according to the risk calculation result; D3, a comprehensive risk zoning map is generated according to the threshold classification in GIS.

[0016] Preferably, in step VI, the specific process of dynamically correcting the partition boundaries and risk levels based on long-term monitoring data is as follows: the correction is performed using a risk value update formula, and the correction process is as follows: first, abnormal areas are identified and marked as areas to be corrected, then local recalculation is performed, overlay analysis is rerun in the areas to be corrected, the corresponding layers are updated, and finally the threshold is adjusted. If the risk value increases for three consecutive cycles, the orange area is upgraded to the red area.

[0017] Beneficial effects of this invention: 1. This invention comprehensively considers the diverse characteristics of active faults and the complex geological conditions of engineering sites through multi-source data fusion, 3D model construction and GIS correction, multi-factor overlay analysis and dynamic correction mechanism. It solves the problem of the limitations of zoning results, improves the accuracy of risk assessment, breaks through the limitations of single indicators, and realizes closed-loop management of "data collection-risk assessment-engineering response". It provides scientific and reliable technical support for the safe site selection, optimized design and operation and maintenance of major infrastructure. 2. This invention uses a displacement-time probability model to calculate the probability of fault slippage during its service life using historical fault slip rate data, providing a quantitative basis for risk assessment. It also improves the accuracy of assessment by using GIS overlay analysis and LSTM model to achieve dynamic correction and trend prediction of zoning boundaries. 3. This invention provides a complete solution for infrastructure such as hydraulic structures, from hazard analysis to stability evaluation. It uses a finite element model to input fault slip parameters for quasi-static analysis to evaluate the structure's resistance to slippage, filling the gap in the existing standards for evaluating the direct effects of fault slippage. 4. This invention solves the problem that existing zoning methods fail to comprehensively consider the diverse characteristics of active faults and the complex geological conditions of engineering sites, resulting in significant limitations in the zoning results and an inability to accurately reflect the true risk status of different areas. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for zoning the hazards of active fault engineering. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0020] Example 1: As Figure 1 As shown, a method for hazard zoning of engineering impacts from active faults includes the following steps: Step 1: Integrate multi-source data of the detected faults, construct a three-dimensional geometric model of the faults, and use GIS spatial analysis to correct the fault distribution pattern; In the multi-source data fusion, the high-density resistivity method is used to identify shallow fault fracture zones. The electrode array inversion formula is as follows: ; In the formula: Apparent resistivity The electrode spacing, For potential difference, For current, For device coefficients; The specific process for constructing a three-dimensional geometric model of a fault and using GIS spatial analysis to correct the fault distribution pattern is as follows: First, multi-source data are integrated and preprocessed, including unifying the coordinate system, aligning data of different resolutions through interpolation, and constructing a fault attribute table. Then, the fault surface traces are drawn in GIS to generate a line feature layer. Based on seismic reflection profiles and borehole data, a three-dimensional geometric model of the fault is established. Next, the surface traces, geophysical anomaly zones, and deep fault models are overlaid to identify spatially inconsistent areas. Weights are assigned according to data confidence levels to calculate the optimal fault location. A buffer zone is generated centered on the deep fault projection and intersected with the surface traces. The fault traces are then adjusted to the overlapping area. The fault is divided into several segments, each 5 km long. The strike and dip changes of each segment are statistically analyzed to determine whether they conform to the direction of the regional tectonic stress field.

[0021] Step 2: Combine fault slip rate, paleoseismic recurrence period and deep engineering in-situ stress inversion to evaluate the activity characteristics of hidden faults during the service life of the project. The deep in-situ stress inversion is based on borehole hydraulic fracturing data to calculate the horizontal principal stress. and : ; In the formula: For the rupture pressure, For the tensile strength of rock, Poisson's ratio, Vertical stress; The fault slip potential assessment uses a displacement-time probability model to calculate the probability of fault slip within the engineering service period T years: ; In the formula: The sliding speed is denoted as .

[0022] Step 3: Establish a fault zone zoning evaluation model based on fault data, and combine the overburden thickness, soft soil / bedrock type and liquefaction potential to identify high-magnification areas by adjusting the ground motion parameters using the site amplification factor. The criteria for dividing the core fracture zone in the mechanical zoning of the fracture zone are as follows: Rock Quality Index (RQD): ; The permeability coefficient is calculated by Darcy's law: ; In the formula, Q is the flow rate, L is the sample length, A is the cross-sectional area, and Δh is the water head difference.

[0023] Step four, constructing a near-fault seismic pulse model to conduct a fine analysis of near-fault seismic motion parameters and evaluate the probability of seismic hazard; specifically including the following steps: ①Embedding the pulse model into the framework of probabilistic seismic hazard analysis to evaluate the exceeding probability of the engineering site's seismic motion parameters; ②The near-fault correction is modified by directionality and slippage effect; ③Calculate the exceeding probability within the engineering service period T and output the seismic hazard report Step five, evaluating the dynamic response and stability of the hydraulic structure; wherein the dynamic response and stability evaluation includes the following steps: ①Input the seismic motion parameters, fault displacement parameters and structure parameters into the finite element model to prepare for dynamic response analysis; ②Conduct time history analysis under the action of seismic motion to output displacement response, stress distribution and acceleration response; ③Conduct quasi-static analysis under the action of fault displacement.

[0024] Step six, superimpose the fault avoidance zone, high PGA area, site amplification area and secondary disaster potential area through GIS analysis to generate a comprehensive risk zoning map, and dynamically correct the zoning boundaries and risk levels according to long-term monitoring data; wherein the GIS superimposition analysis includes the following steps: 1) Weight distribution of fault avoidance zone, high PGA area, site amplification area and secondary disaster potential area, the weights are fault avoidance zone 0.4, high PGA area 0.3, site amplification area 0.2, secondary disaster potential area 0.1 in turn; 2) Calculate the comprehensive risk value, the calculation formula is: ; In the formula, S is the standardized score of each factor. According to the risk calculation results, the threshold values are set as follows:

[0025] 3) Generate a comprehensive risk zoning map according to the threshold classification in GIS.

[0026] The dynamic correction algorithm is corrected by using a risk value updating formula, and the correction process is as follows: first, the abnormal area is identified and marked as a to-be-corrected area, then the local recalculation is performed, the superposition analysis is re-run in the to-be-corrected area, the corresponding layer is updated, and finally the threshold adjustment is performed, if the risk value rises for three consecutive periods, the orange area is upgraded to the red area. Risk value updating formula: In the formula, is the risk increment calculated according to the monitoring data change, that is, the monitoring data weight.

[0027] Step seven, real-time monitoring and early warning of fault activity and ground motion, combining the LSTM machine learning model to predict the fault activation trend and optimize the emergency plan; the specific process is as follows: Input features: Time series: past 365 days of strain data (daily average), monthly average sliding rate, historical earthquake events (magnitude / time, one-hot encoding), precipitation / temperature (affecting ground stress state), a total of 12-dimensional features.

[0028] Model training: TensorFlow framework is adopted, the hidden layer is 2 layers (each 128 units), the dropout rate is 0.2, 80% of the training data is used for training, 20% is used for verification, the loss function is binary cross entropy, and the iteration is converged after 200 times (verification set accuracy 88%).

[0029] Emergency plan optimization: When the predicted activation probability in the next year is greater than 60%, the monitoring frequency is automatically adjusted (strain gauge 1 time / minute→1 time / 10 seconds), and the reinforcement suggestion (such as adding 5 anti-slide piles, budget adjustment ±15%) is pushed to the engineering management system.

[0030] Example 2: see Figure 1 The active fault engineering influence risk zoning method provided by the application realizes full-process coverage from data acquisition, model construction, risk assessment to dynamic early warning through 7 core steps, and the core logic is multi-source data fusion→fault characteristic evaluation→site effect analysis→structure response evaluation→comprehensive risk zoning→dynamic correction and early warning, and the specific core technical principles are as follows: 1. Multi-source data fusion principle Through the combination of multiple methods such as high-density resistivity method (shallow exploration), seismic reflection profile (deep structure), and drilling data (stratum verification), the detection blind area of a single method (such as shallow resistivity method cannot identify deep hidden faults, which needs to be combined with deep seismic reflection data) is solved, and the accuracy of the fault distribution model is ensured (error≤5%) through GIS spatial analysis and weighted correction. ​

[0031] 2. Fault activity quantification principle Stress inversion: using hydraulic fracturing method, the horizontal principal stress is inversely deduced by rock fracture pressure, and it is judged whether the fault is in a critical stress state (when - When the rock shear strength is exceeded, the fault is easy to slide).

[0032] Slip probability model: assuming that the fault slip conforms to the Poisson process, the slip rate λ is a constant, and the slip probability increases exponentially with time during the service period, which is suitable for long-term stable sliding faults (such as creep-type faults).

[0033] 3. Site amplification effect and ground motion correction principle The thickness of the overburden layer and the characteristics of the soft soil cause the amplification of the ground motion (such as the thick soft soil layer, which can amplify the 5-10 Hz ground motion by 2-3 times), and the site amplification coefficient is calculated by a one-dimensional equivalent linear model to correct the bedrock ground motion parameters to the ground surface.

[0034] Near-fault directionality effect: the peak value of the ground motion in the front of the fault rupture direction is 30%-50% higher than that in the rear due to wave focusing, and the correction needs to be made by the rupture direction angle weight coefficient.

[0035] 4. Comprehensive risk zoning and dynamic correction principle The analytic hierarchy process (AHP) is used to determine the weight of each factor, ensuring that the fault avoidance zone (direct threat) has the highest weight and the secondary disaster area has the lowest weight, which conforms to the principle of engineering safety priority.

[0036] Dynamic correction is based on a data-driven model, which updates the risk value in real time through monitoring data, solving the static limitations of traditional zoning methods.

[0037] 5. LSTM prediction principle Using the sequence data processing capability of LSTM, the long-term dependence relationship of fault activity (such as stress accumulation-release cycle) is captured, and compared with traditional time series models (such as ARIMA), it can more accurately predict the nonlinear and non-stationary fault activation trend (prediction accuracy is improved by 15%-20%).

[0038] The present application constructs a three-dimensional model by fusing multi-source data of the fault, and uses GIS to correct the distribution law, combines the fault slip rate, the recurrence period of paleoearthquakes, and the stress inversion to evaluate the activity characteristics, establishes a fault zone zoning evaluation model and identifies high amplification areas, constructs a near-fault ground motion pulse model to evaluate the seismic hazard probability, evaluates the dynamic response and stability of hydraulic structures, and generates a comprehensive risk zoning map through GIS overlay analysis and dynamic correction. Finally, real-time monitoring and early warning are combined with the LSTM model to predict the fault activation trend, breaking through the limitations of single factor, realizing multi-factor comprehensive evaluation, and improving the accuracy of zoning; Through the pulse model embedded in the probability seismic risk analysis framework, considering the directionality and slip effect (making the local peak ground motion magnified by 20%-30%), the problem of underestimation of the traditional method for near-fault strong pulse effect is solved, which is especially suitable for the evaluation of high-risk areas within 5km from the fault, combined with the thickness of the overburden layer, soft soil / bedrock type and liquefaction potential, using Darcy's law and site amplification factor model, the high amplification area of ground motion (such as the amplification factor of 20m thick soft soil layer area can reach 2.5 times) is located, and the design parameter deviation caused by the simplification of site conditions is avoided. Through weight allocation and risk value formula, multi-factor quantitative superposition is realized; combined with long-term monitoring data, abnormal areas are automatically marked, local recalculation and threshold adjustment are performed, the static lag problem of traditional zoning is solved, machine learning is used to capture nonlinear characteristics such as stress accumulation and microseismic sequence, the prediction accuracy is improved by 15%-20% compared with traditional time series model, and the fault activation risk is warned 5-10 years in advance, providing dynamic optimization support for engineering emergency plan. Through the design of a dedicated dynamic response analysis process: through the finite element model input fault displacement parameters for quasi-static analysis, the fault direct displacement effect is evaluated, which fills the gap in the existing specification, especially suitable for the stability design of cross-fault hydraulic structures, from data acquisition (high-density resistivity method / drilling hydraulic fracturing) to output (comprehensive risk zoning map), each step is closely related to the engineering site selection, design, construction and operation requirements. For example: the fault is analyzed by 5km segmentation to match the tectonic stress, which provides a basis for segmented seismic design of linear engineering (such as high-speed rail and oil pipeline); the risk zoning map is directly related to the engineering construction standard (red zone prohibited construction, orange zone strengthened seismic measures), realizing the integration of "evaluation-design-construction"; Through early and accurate zoning, the cost of later reinforcement or reconstruction caused by blind construction in high-risk areas is avoided (estimated to reduce 30%-50% of additional engineering costs); the dynamic warning system reduces the operation loss caused by sudden geological disasters, especially for major facilities such as nuclear power plants and reservoirs, which can greatly improve the disaster prevention and reduction capacity.

[0039] The above embodiments are only preferred technical solutions of the present application, and should not be regarded as limiting the present application, the protection scope of the present application should be based on the technical solutions claimed in the claims, including the equivalent replacement solutions of the technical features claimed in the claims. That is, within this range, equivalent replacement improvements are also within the protection scope of the present application.

Claims

1. A method for hazard zoning of engineering impacts from active faults, characterized in that, Includes the following steps: Step 1: Integrate multi-source data of the detected faults, construct a three-dimensional geometric model of the faults, and use GIS spatial analysis to correct the fault distribution pattern; Step II: Combine fault slip rate, paleoseismic recurrence cycle and deep engineering in-situ stress inversion to evaluate the activity characteristics of hidden faults during the service life of the project. Step III: Establish a fault zone zoning evaluation model based on fault data, and combine the overburden thickness, soft soil / bedrock type and liquefaction potential to identify high-magnification areas by adjusting the ground motion parameters using the site amplification factor. Step IV: Construct a near-fracture ground motion pulse model to conduct a refined analysis of near-fracture ground motion parameters and assess the probability of earthquake hazard; Step V: Evaluate the dynamic response and stability of hydraulic structures; Step VI: The fault avoidance zone, high PGA area, site enlargement area and secondary disaster potential area are analyzed by GIS overlay to generate a comprehensive risk zoning map. At the same time, the zoning boundaries and risk levels are dynamically corrected based on long-term monitoring data. Step VII: Conduct real-time monitoring and early warning of fault activity and ground motion, and optimize emergency response plans by combining LSTM machine learning models to predict fault activation trends.

2. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step I, the multi-source data of the fault detection are fused together, and the high-density resistivity method is used to identify the shallow fault fracture zone. The electrode array inversion formula is as follows: ； In the formula: Apparent resistivity The distance between the electrodes. For potential difference, For current, This is the device coefficient.

3. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step I, the construction of a three-dimensional geometric model of the fault and the correction of the fault distribution pattern using GIS spatial analysis include the following steps: A1. Integrate multi-source data and preprocess the data; A2. Draw the fault surface traces in GIS, generate line feature layers, and establish a three-dimensional geometric model of the fault based on seismic reflection profiles and borehole data. A3. Overlay the surface traces, geophysical anomaly zones, and deep fault models to identify spatially inconsistent areas. Assign weights based on data confidence levels, calculate the optimal fault location, generate a buffer zone centered on the deep fault projection, intersect it with the surface traces, and adjust the fault traces to the overlapping area. A4. Divide the fault into several segments, statistically analyze the changes in the strike and dip angle of each segment, and analyze whether they conform to the direction of the regional tectonic stress field.

4. The method for hazard zoning of engineering impacts from active faults according to claim 3, characterized in that, In step A1, the preprocessing includes unifying the coordinate system, aligning data of different resolutions through interpolation, and constructing a fault attribute table.

5. The method for hazard zoning of engineering impacts from active faults according to claim 3, characterized in that, In step A4, the fault is divided into several segments, each 5 km long.

6. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step II, the fault slip rate is calculated using a displacement-time probability model to determine the probability of fault slip within the engineering service life T years: ； In the formula: The sliding speed is denoted as .

7. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step II, the deep in-situ stress inversion is based on borehole hydraulic fracturing data to calculate the horizontal principal stress. : ； In the formula: For the rupture pressure, For the tensile strength of rock, Poisson's ratio, This represents vertical stress.

8. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step III, the criteria for dividing the core fracture zone in the fault zone zoning evaluation model established by the fault data are as follows: Rock Quality Index (RQD): ； The permeability coefficient is calculated using Darcy's law: ； In the formula: Q is the flow rate, L is the sample length, and A is the cross-sectional area. This is due to head difference.

9. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step IV, the assessment of earthquake hazard probability includes the following steps: B1. Embed the pulse model into the probabilistic seismic hazard analysis framework to assess the exceedance probability of ground motion parameters at the engineering site; B2. Near-fault correction is performed using directionality and thrust slip effects; B3. Calculate the exceedance probability during the service life T of the project and output a seismic hazard report.

10. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, Step V, evaluating the dynamic response and stability of hydraulic structures, includes the following steps: C1. Input the ground motion parameters, fault slip parameters and structural parameters into the finite element model to prepare for dynamic response analysis; C2. Perform time history analysis under seismic action and output displacement response, stress distribution and acceleration response; C3. Perform quasi-static analysis under fault faulting.

11. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step VI, the GIS overlay analysis includes the following steps: D1. Weight allocation is performed for fault avoidance zones, high PGA zones, site enlargement zones, and secondary disaster potential zones; D2. Calculate the comprehensive risk value and classify the risk level based on the risk calculation results; D3. Generate a comprehensive risk zoning map in GIS by classifying according to thresholds.

12. The method for hazard zoning of engineering impacts from active faults according to claim 1, characterized in that, In step VI, the specific process of dynamically correcting the zoning boundaries and risk levels based on long-term monitoring data is as follows: the correction is performed using a risk value update formula. The correction process is as follows: first, abnormal areas are identified and marked as areas to be corrected; then, local recalculation is performed; overlay analysis is rerun in the areas to be corrected to update the corresponding layers; finally, the threshold is adjusted. If the risk value increases for three consecutive periods, the orange zone is upgraded to the red zone.

Images