Risk management method and system for inducing earthquake based on hydraulic fracturing technology

Abstract

The invention relates to the technical field of induced earthquake risk management, and particularly discloses a risk management method and system for induced earthquakes based on a hydraulic fracturing technology, and the method comprises the following steps: determining a risk tolerance standard; constructing a multi-parameter probability risk assessment model, and calculating expected influences of earthquake induction under different earthquake magnitudes based on Monte Carlo simulation; generating a spatial variation red light magnitude threshold graph; implementing a dynamic traffic light protocol based on the threshold graph, and triggering green light, yellow light or red light response measures according to the magnitude monitored in real time; and performing retrospective calibration and optimization based on historical data. The risk management system comprises a data input and storage unit, a risk modeling and Monte Carlo simulation calculation unit, a spatial analysis and drawing unit, a real-time monitoring data interface and a decision support and alarm output unit. According to the method, the threshold setting is associated with the social acceptable influence, so that the risk control is more scientific and reasonable.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of induced seismic risk management, in particular to traffic light protocol implementation technology based on quantitative risk assessment and spatial variation threshold, and more particularly to a risk management method and system for induced seismicity based on hydraulic fracturing technology. BACKGROUND

[0002] Hydraulic fracturing technology is widely used in the development of unconventional oil and gas resources, but this technology has the risk of inducing earthquakes. The current industry generally uses the "traffic light protocol" as a risk management tool, which triggers different levels of response measures according to real-time monitoring of earthquake magnitude, such as "green-yellow-red". However, the existing traffic light protocol mostly uses a uniform magnitude threshold (such as M3.0 or M4.0), without fully considering the risk differences in different regions. This "one-size-fits-all" threshold setting has significant defects: in high population density areas, the existing threshold may lead to an overestimation of risk, while in remote areas, the threshold may be too strict, causing unnecessary work interruptions and economic waste. SUMMARY

[0003] In view of the above technical problems in the related art, the present application provides a risk management method and system for induced seismicity based on hydraulic fracturing technology, which can solve the above problems.

[0004] To achieve the above technical purposes, the technical scheme of the present application is as follows:

[0005] A risk management method for induced seismicity based on hydraulic fracturing technology, comprising the following steps:

[0006] S1, determining a risk tolerance standard, including a nuisance tolerance and a damage tolerance;

[0007] S2, constructing a multi-parameter probabilistic risk assessment model, and calculating the expected impact of induced seismicity under different magnitudes based on Monte Carlo simulation;

[0008] S3, generating a spatial variation red light magnitude threshold map;

[0009] S4, implementing a dynamic traffic light protocol based on the threshold map, and triggering green light, yellow light or red light response measures according to real-time monitoring of the magnitude;

[0010] S5, performing retrospective calibration and optimization based on historical data.

[0011] Further, the nuisance tolerance in step S1 is: based on the community's decimal intensity level, defining the number of households that can tolerate feeling the vibration and its probability of occurrence; and the damage tolerance is: based on the damage state level, defining the number of households that can tolerate minor damage and its probability of occurrence.

[0012] Furthermore, the multi-parameter probabilistic risk assessment model in step S2 includes:

[0013] Earthquake sequence module: Employs a probabilistic tail magnitude prediction model to estimate the maximum magnitude that may be subsequently transmitted after the red light is triggered and production is halted. ;

[0014] Source and Propagation Module: Sets the initial source depth based on the depth of the target stratum and simulates the propagation of earthquake motion;

[0015] Site Effects Module: Quantifies the impact of near-surface geology on seismic motion using the global VS30 dataset and error estimation;

[0016] Impact Conversion Module: Employs disturbance and vulnerability functions to convert seismic motion parameters into disturbance and damage probabilities;

[0017] Exposure module: Integrates high-resolution population spatial distribution data to calculate the expected number of affected households.

[0018] Furthermore, the maximum magnitude that may be sent subsequently. The calculation is performed using the following steps: Based on the distribution characteristics of the tail portion of the earthquake sequence, using the formula... Calculate the magnitude increment dM, where R F The proportion of earthquakes after production shutdown, b is the frequency-magnitude distribution parameter, and q is the confidence level parameter; the maximum magnitude observed before production shutdown is used. Based on the baseline, the magnitude increment dM is superimposed to obtain the maximum possible subsequent magnitude. .

[0019] Furthermore, step S3 specifically includes: establishing a spatial grid in the target work area and performing Monte Carlo simulation in step S2 for each grid point; calculating the red light magnitude value that makes the expected impact meet the tolerance of step S1 through interpolation, and finally generating a spatial threshold map reflecting the change of the red light magnitude threshold with geographical location.

[0020] Furthermore, step S4 specifically includes: using the generated spatial variation red light magnitude threshold map as the basis for on-site operation decisions, monitoring induced seismic activity in real time, comparing the monitored earthquake magnitude with the local red light magnitude threshold, and if the monitored earthquake magnitude is lower than the local red light magnitude threshold by a certain range, it is a green light; if the monitored earthquake magnitude is lower than but close to the local red light magnitude threshold, it is a yellow light, and mitigation measures are initiated; if the monitored earthquake magnitude reaches or exceeds the local red light magnitude threshold, it is a red light, and operations must be stopped.

[0021] Furthermore, step S5 specifically includes: conducting a retrospective assessment using historical earthquake data to verify the accuracy of the set risk tolerance and corresponding threshold map, and optimizing model parameters based on the assessment results to improve the applicability of the protocol.

[0022] Furthermore, it also includes reserving interfaces in the multi-parameter probabilistic risk assessment model for integrating the quantitative impact of mitigation measures on earthquake sequences.

[0023] A risk management system for earthquakes induced by hydraulic fracturing technology includes:

[0024] Data input and storage unit: used for inputting and storing multivariate spatial data, including geological data, seismic data, population data, disturbance functions, and vulnerability functions;

[0025] Risk Modeling and Monte Carlo Simulation Calculation Unit: Used to execute multi-parameter probabilistic risk assessment models;

[0026] Spatial Analysis and Mapping Unit: Used to generate a visualized spatial variation red light magnitude threshold map;

[0027] Real-time monitoring data interface: used to access the real-time data stream of the on-site earthquake monitoring network;

[0028] Decision support and alarm output unit: used to compare real-time earthquake events with local red light magnitude thresholds and output alarm signals.

[0029] The beneficial effects of this invention are as follows: By linking threshold settings to socially acceptable impacts, this application makes risk control more scientific and reasonable; based on different risk conditions in different regions, this application sets red-light seismic magnitude thresholds in a refined manner, achieving differentiated management with strict requirements in high-risk areas and lenient requirements in low-risk areas; through Monte Carlo simulation, this application clearly quantifies various uncertainties in risk prediction, enhancing the transparency and reliability of decision-making; the risk tolerance standards and models of this application can be adjusted according to regulatory requirements or new data updates, exhibiting high flexibility and adaptability. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] The present invention will now be described in further detail with reference to the accompanying drawings.

[0032] Figure 1 This is a flowchart of a risk management method for earthquakes induced by hydraulic fracturing technology, as described in an embodiment of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0034] like Figure 1 As shown, this invention discloses a risk management method for earthquakes induced by hydraulic fracturing technology, comprising the following steps: S1, determining risk tolerance standards, including disturbance tolerance and damage tolerance; S2, constructing a multi-parameter probabilistic risk assessment model, and calculating the expected impact of induced earthquakes at different magnitudes based on Monte Carlo simulation; S3, generating a spatially variable red-light magnitude threshold map; S4, implementing a dynamic traffic light protocol based on the threshold map, triggering green, yellow, or red light response measures according to the real-time monitored magnitude; S5, performing retrospective calibration and optimization based on historical data.

[0035] The system for implementing the aforementioned risk management method is also disclosed, comprising: a data input and storage unit for inputting and storing multivariate spatial data, including geological data, seismic data, population data, disturbance functions, and vulnerability functions; a risk modeling and Monte Carlo simulation calculation unit for executing a multi-parameter probabilistic risk assessment model; a spatial analysis and mapping unit for generating a visualized spatial variation red-light magnitude threshold map; a real-time monitoring data interface for accessing real-time data streams from the on-site seismic monitoring network; and a decision support and alarm output unit for comparing real-time seismic events with local red-light magnitude thresholds and outputting alarm signals.

[0036] Example 1:

[0037] This application establishes risk tolerance standards, namely, defining the acceptable upper limit of earthquake impact, including disturbance tolerance and damage tolerance.

[0038] Disturbance tolerance is defined as follows: based on the decimal intensity level of the community (earthquake intensity scale), the number of households that can tolerate feeling the vibration (e.g., CDI≥3) and the probability of it occurring (e.g., no more than 300,000 households with a 50% probability).

[0039] Damage tolerance: Based on the damage status level, it defines the number of households that can tolerate minor damage (e.g., DS≥1) and the probability of it occurring (e.g., no more than 30 households with a 50% probability).

[0040] Example 2:

[0041] The multi-parameter probabilistic risk assessment model in this application includes an earthquake sequence module, a source and propagation module, a site effect module, an impact transformation module, and an exposure module.

[0042] Earthquake sequence module (predicting earthquake events themselves): First, based on historical data and operational parameters (such as the magnitude before shutdown, R...). F (b value), predict the maximum magnitude that may still occur after triggering the red light production stoppage, clarify the boundary of the earthquake event for risk assessment (i.e. the range of the maximum magnitude to be assessed), and provide a magnitude input benchmark for subsequent modules.

[0043] The source and propagation module (simulating the process of ground motion from underground to the surface): Based on the maximum magnitude output by the earthquake sequence module, combined with the geological data of the target work area (such as stratum depth and rock velocity), the initial source depth is set, the propagation path and energy attenuation of seismic waves in underground rock strata are simulated, and the ground motion parameters (such as peak ground acceleration PGA and seismic intensity) at different locations on the surface are output, completing the transformation from "abstract magnitude" to "concrete ground motion intensity".

[0044] The site effect module (correcting the impact of surface geology on vibration) receives the "basic ground motion parameters" output by the source and propagation module, combines them with the global VS30 dataset (near-surface shear wave velocity, reflecting the softness and hardness of the surface soil layer), and quantifies the "amplification / attenuation effect" of near-surface geology. For example, loose sedimentary layers (low VS30 value) will amplify the vibration, while hard bedrock (high VS30 value) will attenuate the vibration. Finally, it outputs "accurate ground motion parameters after site correction" to ensure that subsequent impact assessments match the actual surface conditions.

[0045] Impact Conversion Module (converting seismic intensity into risk probability): Based on the ground motion parameters corrected by the site effect module, probability conversion is performed using two types of tools: "disturbance function" and "vulnerability function". The disturbance function converts seismic intensity (e.g., CDI≥3) into "the probability that residents will feel the vibration"; the vulnerability function converts parameters such as PGA into "the probability that buildings will suffer minor damage (e.g., DS≥1)", thus completing the conversion from "physical vibration" to "social impact probability".

[0046] Exposure module (calculates the actual size of the affected population): integrates the "probability of harassment / damage" output by the impact conversion module with high-resolution population spatial data (such as the distribution of the number of households), calculates the "expected value of the number of affected households" by "probability × population base" (such as 300,000 households feeling the vibration at a 50% probability), and finally outputs "quantitative results that can be directly used for risk tolerance comparison" (matching the tolerance standard in step S1).

[0047] The earthquake sequence module employs a probabilistic tail magnitude prediction model to estimate the maximum magnitude that may occur after production shutdown. The specific calculation is performed through the following steps:

[0048] Step 1: Calculate the magnitude increment dM (key intermediate variable).

[0049] Based on the distribution characteristics of the tail of the earthquake sequence, using the formula Calculate the magnitude increment dM, where R F The first part represents the proportion of earthquakes after production stoppage, b is the frequency-magnitude distribution parameter, and q is the confidence level parameter. R reflects the impact of production stoppages on the upper limit of earthquake magnitude. F The smaller the value (the stronger the production halt inhibition effect), the smaller the value; the latter half This reflects the confidence level adjustment; the higher the q, the more reliable the requirement. The smaller the value, the larger the final dM.

[0050] Step 2: Calculate the subsequent maximum magnitude .

[0051] The maximum magnitude observed before production ceased Based on the baseline, the magnitude increment dM is superimposed to obtain the maximum possible subsequent magnitude. .

[0052] For example, if the maximum magnitude of the earthquake before production was halted R F =0.2 (only 20% of seismic activity continues after production stoppage), b=1.0 (b value for typical areas), q=0.95 (95% confidence level), then , (That is, with a 95% confidence level, the maximum subsequent magnitude is about 3.6).

[0053] The disturbance function typically takes earthquake intensity (CDI, community seismic intensity) as input parameter (CDI is an intensity rating based on the subjective feelings of the population; for example, CDI≥3 corresponds to "most people indoors felt the tremor, and some people were awakened") and outputs "the probability that residents in a certain area felt this intensity," which can be achieved using the following log-normal distribution function (or empirical statistical curve):

[0054]

[0055] Where I is the ground motion intensity (e.g., the intensity after correction by the site effect module); c is the disturbance threshold (e.g., CDI=3, representing "significant disturbance"); , The median and standard deviation are statistically obtained (based on regional population earthquake sensation survey data). It is the standard normal cumulative distribution function.

[0056] Vulnerability functions typically use peak ground acceleration (PGA) as input (PGA is a key physical quantity for measuring the strength of earthquakes) and output the probability that a building will reach a certain state of damage (such as minor or moderate damage). A log-normal distribution function (or empirical statistical curve) can be used as an example.

[0057]

[0058] Where DS represents the damage state (e.g., DS=1 corresponds to "minor damage", such as wall cracks); d represents the damage threshold (e.g., DS=1). , The median and standard deviation are statistically obtained (based on regional building earthquake damage data). It is the standard normal cumulative distribution function.

[0059] Example 3:

[0060] This application transforms the "risk tolerance standard" into a "dynamically varying red light magnitude threshold based on geographical location" through "spatial gridding + point-by-point simulation + interpolation calculation," ultimately generating a visualized spatially varying red light magnitude threshold map. This provides a decision-making basis for differentiated traffic light protocols, specifically including the following steps:

[0061] The first step is to divide the target work area into uniform "spatial grids" (similar to the pixel grids of an electronic map) based on the actual work area, population density, and geological complexity. The grid size can be flexibly adjusted. For example, high-risk areas (such as the periphery of densely populated towns or geologically fractured zones) can be set with fine grids (such as 500m×500m) to improve accuracy, while remote low-risk areas can be set with coarse grids (such as 2km×2km) to improve calculation efficiency. This transforms the "macro-risk assessment of the entire work area" into "micro-risk calculation of each grid point," avoiding the problem of traditional "one-size-fits-all" thresholds ignoring local differences.

[0062] The second step involves performing a Monte Carlo simulation (S2) for each grid point individually. The core of this simulation is to "traverse different magnitudes and calculate whether the expected impact at that point meets the risk tolerance of S1." The input parameters are specific data for each grid point (such as the near-surface VS30 value, population density, distance from the fracturing zone, and velocity of the passing rock strata). Through thousands of random samplings (the core logic of Monte Carlo simulation), traversing different magnitudes (e.g., M1.0 to M4.0), the simulation calculates the ground motion parameters (PGA, seismic intensity) at each grid point under each magnitude, corrected for "source-propagation-site effect," the disturbance probability transformed by the "disturbance function / vulnerability function," the damage probability, and the expected number of affected households calculated by the "exposure module" (e.g., "at magnitude M2.5, there is a 50% probability that 250,000 households at this grid point will feel the vibration").

[0063] The third step involves, for each grid point, deriving the magnitude value that "makes the expected impact just meet the S1 risk tolerance" based on the simulation results from the second step. This is the red light threshold for that grid point. For example, if S1 is set to "the number of affected households is ≤300,000 with a 50% probability of disturbance tolerance," simulations show that when the magnitude is M2.8, the expected number of affected households at that grid point is 290,000 (meeting the tolerance); when the magnitude is M2.9, the expected number of affected households is 310,000 (exceeding the tolerance). Therefore, the red light magnitude threshold for that grid point is M2.8 (just meeting the maximum magnitude of the tolerance; exceeding it triggers the red light). Since the grid points are discrete (e.g., spaced 500m apart), the thresholds of adjacent grid points may differ. Spatial interpolation algorithms (e.g., Kriging interpolation, inverse distance weighted interpolation) are used to fill in the threshold data between grid points, ensuring a continuous and smooth threshold distribution throughout the work area (avoiding unreasonable situations of "abrupt changes in thresholds between adjacent grids").

[0064] The fourth step is to correlate the red-light magnitude thresholds of all grid points (including those after interpolation) with their geographic coordinates (latitude and longitude) and generate a visualization map using drawing tools (such as GIS software). Different colors or numerical values ​​can be used to label the red-light thresholds for each area. Lower thresholds are assigned to high-population-density areas and areas with soft geological conditions (where site effects are amplified significantly), while higher thresholds are assigned to remote, sparsely populated areas and areas with hard bedrock, thus visually reflecting the characteristics of "spatial variability."

[0065] Example 4:

[0066] This application uses the generated spatial variation red-light magnitude threshold map as the basis for on-site operational decisions, monitors induced seismic activity in real time, compares the monitored earthquake magnitude with the local red-light magnitude threshold, and then performs corresponding operations. To facilitate on-site operations, the threshold map can be simplified into several discrete threshold levels (such as M2.0, M2.5, M3.0, etc.). Specifically, based on the distribution characteristics of the original continuous spatial threshold map, "areas with similar thresholds" can be merged into the same level. For example, high-risk areas (densely populated + soft geology, original continuous threshold M2.0~M2.2) can be unified into discrete level M2.0; medium-high risk areas (relatively densely populated + moderate geology, original continuous threshold M2.3~M2.7) can be unified into discrete level M2.5; medium-low risk areas (sparsely populated + hard geology, original continuous threshold M2.8~M3.2) can be unified into discrete level M3.0; and low-risk areas (remote and unpopulated + bedrock area, original continuous threshold M3.3~M3.8) can be unified into discrete level M3.5.

[0067] Based on discrete thresholds, responses are categorized into green, yellow, and red colors according to magnitude differences, as shown in the table below:

[0068]

[0069] Example 5:

[0070] This application uses historical earthquake data for retrospective assessment to verify the accuracy of the set risk tolerance and corresponding threshold map, and optimizes the model parameters based on the assessment results to improve the applicability of the protocol.

[0071] Reasonableness test of risk tolerance: Determine whether the "disturbing tolerance" (e.g., ≤300,000 households feel the vibration with a 50% probability) and "damage tolerance" (e.g., ≤30 households suffer minor damage with a 50% probability) set in step S1 are consistent with the actual situation. For example, in historical data, an M2.5 earthquake in a certain area actually caused disturbance to 400,000 households, far exceeding the original tolerance, indicating that the tolerance setting is too lenient; or an M3.0 earthquake in a certain area only caused damage to 10 households, far below the original tolerance, indicating that the tolerance may be too strict.

[0072] Accuracy verification of spatial variation threshold maps: Compare the "actual magnitude of historical earthquake events" with the "original red light threshold corresponding to the location of the event" to determine whether the threshold can effectively match the actual risk. For example, if the original red light threshold for a certain location is M2.8, and a historical M2.7 earthquake occurred at that location, the actual number of affected households exceeded the tolerance limit, indicating that the original threshold was set too high (too lenient); or if the original threshold for a certain location is M3.0, and a historical M3.1 earthquake occurred, no damage occurred, indicating that the original threshold was set too low (too strict).

[0073] Validation of parameters in a multi-parameter model: Verify the key parameters of the model in step S2 (such as b-value, R-value, etc.). F Whether the value of b (μ and σ values ​​of the disturbance function / vulnerability function, and VS30 error estimate) fits the historical data—for example, the actual b value of the historical earthquake sequence is 1.1, but the model is initially set to 0.9, which leads to a deviation in magnitude prediction, and this parameter needs to be corrected.

[0074] The entire optimization process follows a standardized procedure of "data collection - comparative analysis - parameter correction - threshold update", and can be repeated.

[0075] The first step is to collect and organize historical data.

[0076] The collected data includes historical hydraulic fracturing operation data of the target work area and similar work areas, corresponding induced earthquake monitoring data (magnitude, focal depth, time of occurrence, location), actual impact data after the earthquake (number of households disturbed / damaged, building damage status recorded in on-site surveys), and model parameters and threshold maps used in historical risk assessments.

[0077] The second step is quantitative comparative analysis.

[0078] Risk tolerance comparison: Based on the magnitude of historical earthquakes and the actual number of affected households, the "tolerance corresponding to historical events" is deduced and compared with the original S1 tolerance to calculate the deviation (e.g., the original tolerance "50% probability ≤ 300,000 households", while the actual number of households affected by the same magnitude in historical events is 450,000, with a deviation rate of 50%).

[0079] Threshold map comparison: For each location where a historical earthquake occurred, extract the red light threshold from the original threshold map and compare it with "whether the actual impact of the earthquake exceeds the tolerance". Statistical analysis of "threshold failure cases" (such as threshold M2.8 corresponding to an earthquake whose actual impact exceeds the tolerance, or threshold M3.0 corresponding to an earthquake whose impact does not exceed the tolerance).

[0080] Model parameter comparison: Substitute historical data into a multi-parameter model, and through fit analysis (such as least squares method), calculate the goodness of fit between the current model parameters and the historical data, and identify parameters with large fit deviations (such as R). F The actual value was 0.2, but the model was set to 0.3, which led to an overestimation of the maximum magnitude.

[0081] The third step is to correct the model parameters.

[0082] For parameters with large fitting bias, recalibration is performed based on historical data: for example, the b-value is recalculated using the magnitude distribution of historical earthquake sequences (following the Gutenberg-Rickett law); and R-values ​​are recalculated by comparing seismic activity before and after historical production shutdowns. F Values ​​(proportion of earthquakes after production stoppage); μ (median) and σ (standard deviation) of the disturbance function / vulnerability function are refitted using actual earthquake sensation / damage data from historical earthquakes; error estimates of the site effect module are corrected using historical VS30 measured data.

[0083] Parameter integration via reserved interfaces: If historical data includes the implementation effects of mitigation measures (such as reducing injection rate), the impact of these mitigation measures on R can be quantified through the model's reserved interfaces. F The influence of values ​​and earthquake sequences should be added to the model parameters (e.g., if the RF value drops from 0.2 to 0.1 after mitigation measures are implemented, the default value of this parameter needs to be updated).

[0084] Step 4: Update the threshold graph and protocol.

[0085] Based on the corrected model parameters, steps S2 (Monte Carlo simulation) and S3 (spatial grid interpolation) are re-executed to generate a new spatial variation red light magnitude threshold map. For example, the original threshold M2.8 for a certain region is updated to M2.6 after parameter correction because the historical data verification is too lenient; the original threshold M3.0 for a certain region is updated to M3.2 because the verification is too strict.

[0086] Synchronously adjust the response boundaries of the dynamic traffic light protocol (such as the magnitude difference between green and yellow lights): If the overall threshold is reduced after correction, the gap between the green light and the threshold can be appropriately narrowed (such as from ≥0.5 magnitude units to ≥0.3) to ensure that the risk control level matches the actual situation.

[0087] Step 5: Verify the optimization effect.

[0088] The updated model parameters and threshold maps are applied to subsequent operations, new actual data are continuously collected, and the above steps are repeated to form an iterative closed loop of "evaluation-optimization-re-evaluation" to avoid model failure due to changes in geological conditions or upgrades in operation processes.

[0089] Example 6:

[0090] This application reserves an interface in the multi-parameter probabilistic risk assessment model for integrating the quantitative impact of mitigation measures on earthquake sequences. This reserved interface serves as a connection channel between the model and "quantitative data on mitigation measures." Essentially, it provides a port for data input and parameter correction within the multi-parameter probabilistic risk assessment model (especially the "earthquake sequence module"), specifically for receiving "quantitative impact data on seismic activity from mitigation measures." For example, traditional mitigation measures during the yellow light phase (such as reducing the injection rate) are often based on experience and do not quantify their suppressive effect on subsequent earthquake sequences (such as upper magnitude limit and earthquake frequency). The interface can incorporate the "quantitative impact" of these measures into the model, allowing the risk assessment results to reflect the actual risks after the measures are implemented. Furthermore, different mitigation measures (such as reducing the rate and skipping segments) have different suppressive effects, and different levels of the same measure (such as reducing the rate by 30% vs. 50%) also have different impacts. The interface can integrate quantitative parameters of multiple measures, allowing the model to calculate the "expected earthquake impact under different combinations of measures," providing a basis for selecting the optimal mitigation scheme on-site.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A risk management method for inducing earthquakes based on hydraulic fracturing technology, characterized in that, Comprising the following steps: S1, determining risk tolerance standards, including harassment tolerance and damage tolerance; S2, constructing a multi-parameter probabilistic risk assessment model, calculating the expected impact of induced earthquakes under different magnitudes based on Monte Carlo simulation; S3, generating a spatially variable red light magnitude threshold map; S4, implementing a dynamic traffic light protocol based on the threshold map, triggering green light, yellow light or red light response measures according to real-time monitored magnitude; S5, retrospective calibration and optimization based on historical data.

2. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, characterized in that, The harassment tolerance in step S1 is: based on the community's decimal intensity scale, define the number of households that can tolerate feeling the vibration and its probability; the damage tolerance is: based on the damage state level, define the number of households that can tolerate minor damage and its probability.

3. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, characterized in that, The multi-parameter probabilistic risk assessment model in step S2 includes: Earthquake sequence module: uses probabilistic aftershock magnitude prediction model to estimate the maximum magnitude that can be sent after a red light is triggered and production is halted ; Seismic source and propagation module: set the initial source depth according to the depth of the target stratum, and simulate the propagation of ground motion; Site effect module: use global VS30 dataset and error estimation to quantify the influence of near-surface geology on ground motion; Impact conversion module: use harassment function and vulnerability function to convert ground motion parameters into harassment probability and damage probability; Exposure module: integrate high-resolution population spatial distribution data to calculate the expected value of affected households.

4. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 3, characterized in that, Maximum magnitude of possible subsequent event The maximum magnitude of possible subsequent event is calculated by the following steps: according to the tail distribution characteristics of the earthquake sequence, the formula Calculate the magnitude increment dM, where R F is the ratio of earthquakes after the shutdown, b is the frequency-magnitude distribution parameter, q is the confidence level parameter; the maximum magnitude observed before the shutdown is the reference, superimposed with the magnitude increment dM, to obtain the maximum magnitude of possible subsequent event .

5. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, wherein, Step S3 specifically includes: establishing a spatial grid in the target work area, and performing Monte Carlo simulation for each grid point in step S2; calculate the red light magnitude value that makes the expected impact meet the tolerance of step S1, and finally generate a spatial threshold map reflecting the change of red light magnitude threshold with geographical location.

6. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, wherein, Step S4 specifically includes: using the generated spatially variable red light magnitude threshold map as the basis for on-site operation decision-making, real-time monitoring of induced seismic activity, comparing the monitored earthquake magnitude with the local red light magnitude threshold, if the monitored earthquake magnitude is lower than the local red light magnitude threshold within a certain range, it is green light; if the monitored earthquake magnitude is lower than but close to the local red light magnitude threshold, it is yellow light, and mitigation measures are started; if the monitored earthquake magnitude reaches or exceeds the local red light magnitude threshold, it is red light, and the operation must be suspended.

7. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, wherein, Step S5 specifically includes: retrospective evaluation using historical earthquake data to test the accuracy of the set risk tolerance and corresponding threshold map, and optimize the model parameters according to the evaluation results to improve the applicability of the protocol.

8. The risk management method for inducing earthquake based on hydraulic fracturing technology according to claim 1, wherein, Also includes reserving an interface in the multi-parameter probabilistic risk assessment model for integrating the quantitative impact of mitigation measures on earthquake sequences.

9. A system for implementing the risk management method according to any one of claims 1 to 8, characterized in that, Comprising: Data input and storage unit: for inputting and storing multi-element spatial data, including geological data, seismic data, population data, harassment function, vulnerability function; Risk modeling and Monte Carlo simulation calculation unit: for executing multi-parameter probabilistic risk assessment model; Spatial analysis and mapping unit: for generating a visual spatially variable red light magnitude threshold map; Real-time monitoring data interface: for accessing real-time data stream of the on-site seismic monitoring network; Decision support and alarm output unit: for comparing real-time earthquake events with local red light magnitude threshold, and outputting alarm signals.

Images