Hot dry rock reservoir transformation mutual feedback multistage dynamic earthquake control method based on comprehensive prediction

Abstract

The invention discloses a hot dry rock reservoir transformation mutual feedback multi-stage dynamic earthquake control method based on comprehensive prediction, and relates to the field of enhanced geothermal system development, and the method comprises the following steps: S1, dividing a limited earthquake magnitude into four stages according to a site earthquake control demand; s2, establishing an induced earthquake real-time monitoring system, and providing a real-time induced earthquake catalog; s3, comprehensively predicting and evaluating an induced earthquake risk; s4, according to the different limited earthquake magnitudes determined in the step S1 and the induced earthquake risk comprehensive analysis provided in the step S3, dynamically adjusting hierarchical management and control early warning conditions; s5, performing dynamic limitation on injection parameters by utilizing the earthquake risk analysis in the step S3; and S6, taking corresponding countermeasures according to different dynamic early warning conditions in the step S4 to realize prospective active earthquake control. And S7, through induced earthquake real-time monitoring in S2, mutual feedback type analysis in S3 and S5, dynamic injection limiting parameters are provided, dynamic early warning conditions are provided for S4 in combination with S3, prospective active earthquake control adjustment measures in S6 are realized, a real-time earthquake risk analysis and dynamic parameter adjustment mutual feedback type multi-stage earthquake control system is formed, and finally the purpose of safe earthquake control is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of enhanced geothermal system development, in particular to a dry hot rock reservoir reconstruction mutual feedback multi-level dynamic shock control method based on comprehensive prediction. BACKGROUND

[0002] As a clean energy with great potential, dry hot rock geothermal resources have great development potential, but in the process of dry hot rock development and reservoir reconstruction, high-energy earthquakes are easily induced, which leads to the suspension or even stop of multiple projects. Controlling induced earthquakes has become an important task that dry hot rock development needs to face and solve.

[0003] At present, the injection-induced earthquake control is mainly through the red-green light control system, which is a passive control system. When the induced earthquake exceeds the maximum control magnitude, it will occur. Some scholars propose to introduce a forward-looking induced earthquake prediction model into the red-green light control system, but the prediction model will overestimate the risk level, leading to an increase in injection parameter limitation and a decrease in development effect. SUMMARY

[0004] In view of the above problems, the purpose of the present application is to provide a dry hot rock reservoir reconstruction mutual feedback multi-level dynamic shock control method based on comprehensive prediction, which combines forward-looking induced earthquake comprehensive prediction technology, realizes mutual feedback type dynamic grading control of injection parameter adjustment and earthquake risk analysis, overcomes the passive control of traditional red-green light control system and the limitation of reservoir reconstruction scale based on conservative induced earthquake prediction model, and can maximize the safety of earthquake to realize efficient dry hot rock reservoir development.

[0005] To achieve the above purpose, the present application adopts the following technical scheme: The dry hot rock reservoir reconstruction mutual feedback multi-level dynamic shock control method based on comprehensive prediction comprises the following steps: S1: According to the site shock control requirement, the limited magnitude is divided into four levels; S2: Establish an induced earthquake real-time monitoring system to provide a real-time induced earthquake catalog; S3: Evaluate the induced earthquake risk by comprehensive prediction of seismic b value, seismic release efficiency, maximum magnitude induced probability, etc., to provide dynamic evaluation basis for S4 warning condition, S5 injection parameter control and S6 injection parameter adjustment; S4: According to S1 to determine different limited magnitudes and S3 to provide comprehensive analysis of induced earthquake risk, establish dynamic adjustment grading control warning condition; S5: Use seismic risk analysis combined with fault risk analysis to realize dynamic limitation of injection parameters; S6: According to the different dynamic warning conditions in S4, take corresponding measures to realize proactive active shock control; S7: Through the real-time monitoring of induced earthquakes in S2, the dynamic injection limiting parameters are provided by the mutual feedback analysis of S3 and S5, the dynamic early warning conditions are provided for S4 by S3, the prospective active earthquake control adjustment measures of S6 are realized, the multi-level dynamic earthquake control system of real-time earthquake risk analysis and parameter adjustment mutual feedback is formed, and finally the purpose of safe earthquake control is achieved.

[0006] In step S1: The magnitude limit is divided into four levels, the first three levels are spaced by 1 magnitude, and the third level is spaced by 0.5 magnitude from the fourth level. The threshold of the limited magnitude should be set according to the actual seismic control requirements around the construction site.

[0007] In step S2: Provide near real-time (earthquake catalog report time less than 5 minutes) induced earthquake catalog, including earthquake magnitude, spatial coordinates and earthquake time; The induced earthquake time and injection time should be consistent, and the induced earthquake source should be distributed within 2km of the injection point.

[0008] Step S3 includes: Real-time analysis of seismic risk based on b value: The earthquake b value is calculated by maximum likelihood method: ; Where N is the number of induced earthquakes, Mc is the minimum complete catalog, Mi is the magnitude of the i-th earthquake, the earthquake distribution should meet Gutenberg-Richter relationship, i.e. G-R distribution characteristics, and the sample size should be greater than 150; Seismic release efficiency: The seismic release efficiency is: ; Where ΣE O is the total seismic release energy, E H is the hydraulic energy ( ), where P(t) is the injection pressure at t, and Q(t) is the injection flow rate at t; The seismic release efficiency is less than 10 -2 , which is low risk, 10 -2～ 10 -1 is medium risk, and greater than 10 -1 is high risk; Using statistical model, the probability of induced threshold magnitude is quantitatively analyzed to provide basis for injection parameter adjustment: Combined with the statistical model Shapiro (2010) formula, the induced earthquake probability p is calculated in real time by earthquake b value and seismogenic index: ; Where M is the maximum different limited magnitude set in S1, and b is the b value in S2; Σ is a pregnant earthquake index: Σ = log N M − log V + bM; Wherein N M is the number of earthquakes greater than M magnitude, b is the b value in S2, V is the injection volume; Maximum magnitude and injection volume quantitative analysis: First, according to the formula (1) derived from the earthquake distribution law of Mcgarr (2014), combined with the relationship between cumulative seismic moment and injection volume formula (2), finally using the relationship between seismic moment and moment magnitude formula (3), the relationship between injection volume and maximum injection magnitude formula (4) can be obtained: (1); (2); (3); (4); Wherein Momax is the maximum seismic moment, ΣMo is the cumulative seismic moment, V is the injection volume, Mo is the seismic moment, Mw is the moment magnitude, K is the linear proportional coefficient of cumulative seismic moment and injection volume calculated by least square method, Mwmax is the maximum moment magnitude.

[0009] In step S4: four trigger conditions are set, each trigger condition is based on the limited magnitude set in S1 and the forward-looking comprehensive prediction analysis of S3, the dynamic early warning condition setting is realized, any one of the trigger conditions is triggered, the corresponding early warning condition is started, when different early warning conditions are triggered at the same time, the maximum early warning condition is selected for early warning.

[0010] In step S5; the relationship between injection volume and maximum injection magnitude in step S3 is used to limit the injection volume in real time; The maximum injection pressure is limited by the critical pore pressure of the dangerous fault within 2km around the injection point: ; Wherein P pri is the critical pore pressure, σ is the normal stress of fault plane, is the shear stress of fault plane, μ is the friction coefficient, Pl is the hydrostatic pressure; Then calculate the maximum wellhead injection pressure P max : P max =P pri +△P; Wherein P max is the maximum limited wellhead pressure, △P is the loss pressure; If the fourth early warning condition is triggered and the modification purpose is still not achieved, the injection can be restarted through the above process after the pump is stopped for 6 months.

[0011] In step S6, after the dynamic early warning condition in S4 is triggered, the injection parameters are dynamically limited according to the relationship between the real-time injection parameters in S5 and the induced seismic magnitude, different injection parameter adjustment strategies correspond to different early warning conditions, and the safety control earthquake purpose is achieved. If the fourth early warning condition is triggered and the modification purpose is still not achieved, the injection can be restarted through the above process after the pump is stopped for 3 months.

[0012] In step S7, through real-time monitoring of induced earthquakes in S2, dynamic injection limiting parameters are provided by mutual feedback analysis of S3 and S5, dynamic early warning conditions are provided for S4, proactive control earthquake adjustment measures of S6 are realized, a multi-level dynamic earthquake control system of real-time earthquake risk analysis and parameter adjustment mutual feedback is formed, and finally the safety control earthquake purpose is achieved.

[0013] Compared with the prior art, the dry hot rock reservoir modification mutual feedback multi-level dynamic earthquake control method based on comprehensive prediction provided by the present application combines the forward-looking induced earthquake prediction technology, overcomes the passive control of the traditional red-green light control system, reduces the limitation of the reservoir modification scale based on the conservative induced earthquake prediction model, and can realize efficient dry hot rock reservoir development under the premise of ensuring seismic safety. BRIEF DESCRIPTION OF DRAWINGS

[0014] Other purposes and results of the present application will be more apparent and easy to understand by referring to the following description in conjunction with the drawings and with a more comprehensive understanding of the present application. In the drawings: Figure 1 The flowchart of controlling the induced earthquake magnitude is shown. DETAILED DESCRIPTION

[0015] The specific embodiments of the present application will be further described in detail below in conjunction with the drawings and specific embodiments. The following examples or drawings are used to illustrate the present application, but do not limit the scope of the present application, and the well-known structures and devices are shown in the form of block diagrams.

[0016] The dry hot rock reservoir modification mutual feedback multi-level dynamic earthquake control method based on comprehensive prediction comprises the following steps: S1: According to the site earthquake control requirement, the limited magnitude is divided into four levels; S2: Establish a real-time monitoring system for induced earthquakes to provide a real-time induced earthquake catalog; S3: The induced earthquake risk is evaluated by comprehensive prediction of the earthquake b value, earthquake release efficiency, maximum magnitude induced probability, etc., to provide dynamic evaluation basis for S4 early warning condition, S5 injection parameter control and S6 injection parameter adjustment; S4: According to S1 to determine different magnitude and S3 to provide induced earthquake risk comprehensive analysis, to establish dynamic adjustment grading control early warning conditions; S5: Using earthquake risk analysis, combined with fault risk analysis, to realize dynamic limitation of injection parameters; S6: According to different dynamic early warning conditions in S4, take corresponding measures to realize proactive active earthquake control; S7: Through S2 induced earthquake real-time monitoring, using S3 and S5 through mutual feedback analysis, provide dynamic injection limiting parameters, combined with S3 for S4 to provide dynamic early warning conditions, realize S6 proactive active earthquake control adjustment measures, form real-time earthquake risk analysis and parameter adjustment mutual feedback type multi-level dynamic earthquake control system, finally achieve the purpose of safe earthquake control.

[0017] Step S1 includes: the magnitude is divided into four levels, the first three levels are 1 level apart, the third level is 0.5 level apart from the fourth level, and the fourth level is set according to the actual situation of the surrounding earthquake resistance of the construction site.

[0018] Step S2 includes: high-precision near real-time (earthquake catalog speed report time less than 5 minutes) seismic monitoring network; For M W ≥-1 earthquake events are interpreted and monitored in real time, and microseismic monitoring network belongs to the prior art, which is not described in detail here. Please refer to "Microseismic Ground Monitoring Technical Specification (SY / T 7372-2017)". Induced earthquake catalog is required, including earthquake magnitude, spatial coordinates and earthquake time.

[0019] Eliminate natural earthquakes: The induced earthquake is judged from two aspects: one is that the induced earthquake time and the injection time should be consistent; The second is that the induced earthquake source location should be distributed within 2km of the injection point.

[0020] Step S3 includes: Real-time analysis of b-value earthquake risk: Gutenberg-Richter relationship (i.e. G-R distribution characteristics) has been recognized by seismologists as a proper description of the relationship between magnitude and frequency. B represents the ratio of the frequency of each earthquake, which is one of the important indicators to measure regional seismic activity. It is generally believed that the b value of natural tectonic earthquake is about 1, and the b value of isobaric fracture activity such as water injection is usually greater than 1, so when the b value tends to 1, it means that the earthquake is biased to natural tectonic activation, and the smaller the b value, the more the proportion of large earthquakes in the earthquake catalog, so the change of b value is a qualitative index to judge the increase of earthquake risk; Calculate the b value of earthquake data by maximum likelihood method: ; Where N represents the number of induced earthquakes, Mc is the minimum complete list, Mi is the magnitude of the i-th earthquake, and the earthquake distribution must conform to the Gutenberg-Richard relation, i.e., the GR distribution characteristics. The sample size is selected to be greater than 150, and the lower limit of the magnitude is generally selected as the peak magnitude; Earthquake release efficiency: ; Where ΣE O For the total energy released by the earthquake, E H For hydropower ( Where P(t) is the injection pressure at time t, and Q(t) is the injection velocity at time t, Bentz et al. (2009) compared the seismic injection efficiency of several EGS development projects and found that most hot dry rock development projects I E Less than 10 -2 Increased seismic release efficiency indicates that the injection has a greater impact on the reservoir stress state, thus increasing the seismic risk. Statistical models are used to conduct quantitative analysis of the probability of inducing threshold magnitude, providing a basis for adjusting injection parameters. The probability p of inducing different magnitudes was calculated using Shapiro's (2010) formula: ; Where M is the magnitude whose triggering probability needs to be calculated, b is the b value in S2, and Σ is the earthquake-pregnancy index; Σ=logN M - logV + bM; Where N M The number of earthquakes greater than magnitude M is given, b is the b value in S2, and V is the injection volume.

[0021] Quantitative analysis of maximum magnitude and injection volume: First, based on McGarr's (2014) earthquake distribution pattern, formula (1) is derived. Then, combined with the relationship between cumulative seismic moment and injection volume (2), and finally using the relationship between seismic moment and moment magnitude (3), the relationship between injection volume and maximum injection magnitude (4) can be obtained. (1); (2); (3); (4); Where Momax is the maximum seismic moment, ΣMo is the cumulative seismic moment, V is the injection volume, and Mo is... Seismic moment, Mw is the moment magnitude, K is the linear proportionality coefficient between the cumulative seismic moment and the injected volume calculated using the least squares method, and Mwmax is the maximum moment magnitude; At the beginning of injection, due to the lack of actual injection parameter values, a conservative estimate of the induced maximum magnitude and injection volume can be made using the McGarr (2014) formula, which does not take into account the rock fissure and porosity reduction effect on strain, so that the maximum magnitude predicted by the injection volume is generally larger, Momax=GV, where G is the shear modulus.

[0022] With injection, the ratio K of injection volume to cumulative seismic moment can more truly reflect the stress-strain relationship caused by injection volume, so in subsequent maximum magnitude prediction, the linear proportionality coefficient K of cumulative seismic moment ΣMo and injection volume, and real-time b value, are used to predict the relationship between maximum magnitude and injection volume.

[0023] Step S4 includes: Triggering any of the early warning conditions will start the corresponding early warning condition, and when different early warning conditions are triggered at the same time, the maximum early warning condition is selected and fed back to S6.

[0024] Early warning condition one: the b value in S4 is less than 1.5, but there is no continuous decrease; the earthquake release efficiency is greater than 10 -3 , but not continuously increasing; the first level threshold magnitude occurrence probability defined in S1 calculated in S4 is greater than 90%; inducing greater than or equal to the first level limit magnitude set in S1, and less than the second level limit magnitude value.

[0025] Early warning condition two: the b value in S4 is less than 1.5 and continuously decreasing; the earthquake release efficiency is greater than 10 -2 , and continuously increasing; the probability P of inducing the second level threshold magnitude in S1 is greater than 90%; inducing greater than or equal to the second level limit magnitude set in S1, and less than the third level limit magnitude value.

[0026] Early warning condition three: the b value in S4 is less than 1.2 and continuously decreasing; the earthquake release efficiency is greater than 10 -2 ; the probability P of inducing the third level threshold magnitude in S1 is greater than 90%; inducing greater than or equal to the third level limit magnitude set in S1, and less than the fourth level limit magnitude value.

[0027] Early warning condition four: the b value in S4 is less than or equal to 1; the earthquake release efficiency is greater than 10 -1 , and the order of magnitude increases sharply; the probability P of inducing the fourth level threshold magnitude in S1 is greater than 90%; inducing greater than or equal to the fourth level limit magnitude in S1.

[0028] Step S5 includes: Injection volume limit: The relationship between injection volume and maximum injection magnitude in step S3 is used to limit the injection volume in real time.

[0029] Maximum wellhead pressure limit: The maximum injection pressure is limited by calculating the critical pore pressure of faults within 2km from the wellhead. First, the critical pore pressure of faults within 2km from the wellhead that can induce the fourth level of limited magnitude earthquake is calculated ; where P pri is the critical pore pressure, σ is the normal stress of the fault plane, is the shear stress of the fault plane, and μ is the friction coefficient.

[0030] Then the maximum limited wellhead injection pressure P max P max = P pri +△P; where P max is the maximum limited wellhead pressure, and △P is the loss pressure.

[0031] After triggering the S4 early warning condition, the wellhead pressure is limited according to the corresponding management measures of S6.

[0032] Step S6 includes: First, if no early warning condition is triggered, the injection volume and injection pressure are limited according to the initial injection volume limit and wellhead pressure limit in step S5 according to the reservoir reconstruction plan.

[0033] Triggering early warning condition one in S4, taking measure one: controlling the injection flow rate to ensure that the injection pressure does not increase rapidly, and if the reconstruction process needs to be pressurized, the wellhead pressure increase is less than 5% / h. The injection volume limit is determined by the real-time injection volume and the maximum magnitude relationship in S5, where the maximum magnitude is selected as the first level of magnitude value defined in S1.

[0034] Triggering early warning condition two in S4, taking measure two: not increasing the injection pressure and injection flow rate. The injection volume limit is determined by the real-time injection volume and the maximum magnitude relationship in S5, where the maximum magnitude is selected as the second level of magnitude value defined in S1.

[0035] Triggering early warning condition three in S4, taking measure three: slowly reducing the injection displacement, and keeping the wellhead pressure below the maximum wellhead pressure before triggering early warning condition three. The injection volume limit is determined by the real-time injection volume and the maximum magnitude relationship in S5, where the maximum magnitude is selected as the third level of magnitude value defined in S1.

[0036] Trigger S4 warning condition four, take measures four: slow down the injection displacement, keep the wellhead pressure slowly down, and the wellhead pressure reduction is kept less than 5% / h, until the pump is stopped. The injection volume is limited by the real-time injection volume and the maximum magnitude relationship in S5, wherein the maximum magnitude is selected as the fourth level magnitude value defined in S1. If the injected volume is greater than the real-time limited injection volume after the pump is stopped, the back flushing can be performed after the pump is stopped until the injection volume is less than the limited injection volume analyzed by S5.

[0037] Step S7 includes: Through the induced earthquake real-time monitoring of S2, the dynamic injection limiting parameters are provided by the mutual feedback analysis of S3 and S5, the dynamic warning conditions are provided for S4 by S3, the prospective active shock control adjustment measures of S6 are realized, the multi-level dynamic shock control system of real-time earthquake risk analysis and dynamic parameter adjustment mutual feedback is formed, and finally the safety shock control purpose is achieved. Embodiments

[0038] The dry hot rock reservoir reconstruction mutual feedback multi-level dynamic shock control method based on comprehensive prediction includes the following steps: S1: According to the site shock control demand, the limited magnitude is divided into four levels; S2: Establish an induced earthquake real-time monitoring system to provide a real-time induced earthquake catalog; S3: Evaluate the induced earthquake risk by the comprehensive prediction of the earthquake b value, earthquake release efficiency, and maximum magnitude induced probability, and provide dynamic evaluation basis for S4 warning condition, S5 injection parameter control, and S6 injection parameter adjustment; S4: According to the different limited magnitudes determined by S1 and the comprehensive analysis of the induced earthquake risk provided by S3, establish dynamic adjustment grading control warning conditions; S5: Use the earthquake risk analysis to realize dynamic limitation of the injection parameters in combination with the fault danger analysis.

[0039] S6: According to the different dynamic warning conditions in S4, take corresponding measures to realize the prospective active shock control; S7: Through the induced earthquake real-time monitoring of S2, the dynamic injection limiting parameters are provided by the mutual feedback analysis of S3 and S5, the dynamic warning conditions are provided for S4 by S3, the prospective active shock control adjustment measures of S6 are realized, the multi-level dynamic shock control system of real-time earthquake risk analysis and dynamic parameter adjustment mutual feedback is formed, and finally the safety shock control purpose is achieved.

[0040] Step S1 includes: Firstly, the local induced earthquake bearing capacity is investigated and analyzed. Since the local distance to the town is relatively close, the demand for earthquake resistance is extremely high, and the maximum induced magnitude needs to be controlled within M W 2.5, so the limited magnitude is divided into four levels: the first level limited magnitude is the moment magnitude MW 0 level; the second level limit set is M W 1 level; the third level limit is M W 2 level, the fourth level limit is M W 2.5 level.

[0041] Step S2 includes: A high-precision near real-time seismic monitoring network is established. For M W ≥-1 level earthquake events, real-time interpretation and monitoring are carried out, and an induced earthquake catalog is provided within 5 minutes of the occurrence of the earthquake, including earthquake magnitude, spatial coordinates and earthquake time, a total of 1548 induced earthquake times are recorded.

[0042] Step S3 includes: The earthquake catalog fed back from S2 is analyzed, and the b value is calculated in real time. It is found that the distribution of earthquakes greater than-0.3 level must meet the Gutenberg-Richter relationship, that is, the G-R distribution characteristics, and the earthquakes greater than-0.3 level are selected as the b value calculation earthquake catalog.

[0043] The formula is used to calculate the real-time earthquake release efficiency, and the overall earthquake release efficiency is less than 10 -4 .

[0044] The probability of the four limited magnitudes 0, 1, 2 and 2.5 in S1 is calculated in real time.

[0045] The maximum injection volume limited by the fourth level limited magnitude 2.5 in S1 is calculated in real time.

[0046] Step S4 includes: According to the limited magnitude of S1 and the dynamic risk analysis of S3, the warning condition is established Warning condition one: the b value in S4 is less than 1.5, but there is no continuous decrease; the earthquake release efficiency is greater than 10 -3 , but it does not continuously increase; the probability of inducing Mw0 threshold magnitude is greater than 90%; inducing earthquakes greater than Mw0 level and less than Mw1.0 level.

[0047] Warning condition two: the b value in S4 is less than 1.5 and continuously decreases; the earthquake release efficiency is greater than 10 -2 , and continuously increases; the probability of inducing Mw1.0 earthquake P is greater than 90%; inducing earthquakes greater than Mw1.0 and less than Mw2.0.

[0048] Warning condition three: the b value in S4 is less than 1.2 and continuously decreases; the earthquake release efficiency is greater than 10 -2 ; the probability of inducing Mw2 earthquake is greater than 90%; inducing earthquakes greater than Mw2.0 and less than Mw2.5.

[0049] Early warning condition four: b value in S4 is less than or equal to 1 and continuously decreases; earthquake release efficiency is greater than 10 -1 , and increases by an order of magnitude; P probability of inducing a 2.5 magnitude earthquake is greater than 90%; and an earthquake greater than or equal to Mw2.5 is induced.

[0050] Step S5 includes: Injection volume limitation: Initial injection volume limitation: using the relationship between injection volume and maximum injection magnitude in step S3, the initial injection volume is limited with a maximum magnitude of 2.5, after triggering the S4 early warning condition, according to the corresponding measures in S6, the maximum magnitude and injection volume quantitative relationship in step S3 is used to limit the injection volume in real time; Maximum wellhead pressure limitation: The maximum injection pressure is limited by calculating the critical pore pressure of the fault within a range of 2 km near the wellhead, and the maximum limited wellhead pressure is 60 MPa.

[0051] Step S6 includes: Measure one: control injection flow rate to ensure that injection pressure does not increase rapidly, and if pressure boosting is required during the modification process, the wellhead pressure should be increased by less than 5% / h. The injection volume limitation is determined by the real-time injection volume and maximum magnitude relationship in S5, and the maximum magnitude is selected as the 0 magnitude earthquake defined in S1.

[0052] Triggering early warning condition two in S4, measure two: do not increase injection pressure and injection flow rate. The injection volume limitation is determined by the real-time injection volume and maximum magnitude relationship in S5, and the maximum magnitude is selected as the 1.0 magnitude earthquake defined in S1.

[0053] Triggering early warning condition three in S4, measure three: slowly reduce injection flow rate and keep wellhead pressure below the maximum wellhead pressure before triggering early warning condition three. The injection volume limitation is determined by the real-time injection volume and maximum magnitude relationship in S5, and the maximum magnitude is selected as the 2 magnitude earthquake defined in S1.

[0054] Triggering early warning condition four in S4, measure four: slowly reduce injection flow rate and keep wellhead pressure slowly decreasing, and the wellhead pressure decrease is kept less than 5% / h until the pump is stopped. The injection volume limitation is determined by the real-time injection volume and maximum magnitude relationship in S5, and the maximum magnitude is selected as the 2.5 magnitude earthquake defined in S1. If the injected volume is greater than the real-time limited injection volume after the pump is stopped, back flushing can be performed after the pump is stopped until the injection volume is less than the limited injection volume analyzed in S5.

[0055] Step S7 includes: First, according to S5, the initial injection volume and maximum magnitude relationship is used to limit the injection volume to 210m 3The dangerous fault within 2km of the periphery is analyzed, and the wellhead pressure is limited to 60Mpa. After injection, real-time analysis of induced earthquakes by S3 is used to feed back to S5 to limit the injection volume in real time.

[0056] When injected to 104m 3 , a greater than 0.03 magnitude earthquake was induced, triggering S4 warning condition one, and S6 control measures were taken, the pressure was slowly increased, and the planned injection arrangement was not exceeded, and the injection was continued.

[0057] When injected to 1256 m 3 , the probability of inducing a 1.0 magnitude earthquake reached 90%, triggering S4 warning condition two, and S6 measure two was taken, and the injection pressure was no longer increased. The injection volume is limited within the S5 real-time limited volume.

[0058] When injected to 2670m 3 , the probability of inducing a 2.0 magnitude earthquake reached 70%, and the earthquake b value continuously decreased to 1.18, triggering warning condition three, and response measure three was taken, keeping the wellhead pressure below the maximum wellhead pressure before triggering the warning condition three, less than 50MPa. The volume is limited within the S5 real-time limited volume.

[0059] Continue to inject to the predetermined 4600m 3 , the earthquake b value is 1.1 and stable, the maximum induced magnitude is 1.7, and the probability of inducing a 2.5 magnitude earthquake is 83%, without triggering the warning condition four.

[0060] The final injection plan is successfully completed, and the magnitude is controlled within 2.5 magnitude.

[0061] The above is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A dry hot rock reservoir reconstruction mutual feedback multi-stage dynamic control method based on comprehensive prediction, characterized in that, Comprise the following steps: S1: According to the site control earthquake demand, the magnitude limit is divided into four levels; S2: Establish induced earthquake real-time monitoring system, provide real-time induced earthquake catalog; S3: Through the earthquake b value, earthquake release efficiency, maximum magnitude induced probability, etc. Comprehensive prediction of induced earthquake risk, provide dynamic evaluation basis for S4 early warning condition, S5 injection parameter control and S6 injection parameter adjustment; S4: According to S1 to determine different magnitude limit and S3 to provide comprehensive analysis of induced earthquake risk, establish dynamic adjustment grading control early warning condition; S5: Using earthquake risk analysis, combined with fault risk analysis, realize dynamic limitation of injection parameters; S6: According to the different dynamic early warning conditions in S4, take corresponding measures to realize proactive active control of earthquake; S7: Through S2 induced earthquake real-time monitoring, using S3 and S5 through mutual feedback analysis, provide dynamic injection limit parameters, combined with S3 for S4 to provide dynamic early warning condition, realize S6 proactive active control of earthquake adjustment measures, form real-time earthquake risk analysis and parameter adjustment mutual feedback type multi-level dynamic control system, finally achieve the purpose of safe control of earthquake.

2. The integrated prediction based hot dry rock reservoir reformation cross-feeding multi-stage dynamic control seismic method according to claim 1, characterized in that, In the step S1: The magnitude limit is divided into four levels, the interval between the first three levels is 1 level, the interval between the third level and the fourth level is 0.5 level, the threshold of the magnitude limit needs to be set according to the actual earthquake control requirements around the construction site.

3. The integrated prediction based hot dry rock reservoir reformation cross-feeding multi-stage dynamic control seismic method according to claim 1, characterized in that, In the step S2: Provide near real-time (earthquake catalog speed report time less than 5 minutes) induced earthquake catalog, including earthquake magnitude, spatial coordinates and earthquake time; The induced earthquake time and injection time need to be consistent, and the induced earthquake source needs to be distributed within 2km of the injection point.

4. The integrated prediction based hot dry rock reservoir reformation cross-feeding multi-stage dynamic control seismic method according to claim 1, characterized in that, In the step S3, it includes: Based on b value earthquake risk real-time analysis: The earthquake b value uses maximum likelihood method to calculate the b value of earthquake data: ； Where N is the number of induced earthquakes, Mc is the minimum complete catalog, Mi is the magnitude of the i-th earthquake, the earthquake distribution should meet the Gutenberg-Richter relationship, that is, the G-R distribution characteristics, and the sample size is greater than 150; Earthquake release efficiency: The earthquake release efficiency is: I E =E O / E H； where ΣE O is the total seismic release energy, E H is the hydraulic energy ( ), where P(t) is the injection pressure at time t and Q(t) is the injection flow rate at time t. Seismic release efficiency less than 10 -2 Low risk, 10 -2～ 10 -1 Medium risk, greater than 10 -1 High risk; Using statistical model, quantitative analysis is carried out on the induced threshold magnitude probability to provide basis for injection parameter adjustment: Combined with statistical model Shapiro (2010) formula, the induced earthquake probability p is calculated in real time through earthquake b value and seismic index: ； Where M is the maximum different magnitude limit set in S1, b is the b value in S2; Σ is the seismic index: ∑ = log N M − log V + bM; where N M is the number of earthquakes greater than M, b is the b-value in S2, and V is the volume of fluid injected. Maximum magnitude and injection volume quantitative analysis: Firstly, according to the formula (1) derived from Mcgarr (2014) earthquake distribution law, combined with the relationship formula (2) between cumulative seismic moment and injection volume, finally using the relationship formula (3) between seismic moment and moment magnitude, the relationship formula (4) between injection volume and maximum injection magnitude can be obtained: （1）； （2）； ； ； Where Momax is the maximum seismic moment, ΣMo is the cumulative seismic moment, V is the injection volume, Mo is the seismic moment, Mw is the moment magnitude, K is the linear proportionality coefficient of cumulative seismic moment and injection volume calculated by least square method, Mwmax is the maximum moment magnitude.

5. The integrated prediction based hot dry rock reservoir reformation cross- feed multi-stage dynamic control seismic method of claim 1, wherein, In the step S4: Four trigger conditions are set, each of which is based on the limited magnitude set by S1 and the prospective comprehensive prediction analysis of S3 to achieve dynamic early warning condition setting. Triggering any of the early warning conditions will start the corresponding early warning condition. When different early warning conditions are triggered at the same time, the maximum early warning condition is selected for early warning.

6. The integrated prediction based hot dry rock reservoir reformation cross-feeding multi-stage dynamic control seismic method according to claim 4, characterized in that, In the step S5; The relationship between the injection volume and the maximum injection magnitude in step S3 is used to limit the injection volume in real time; The maximum injection pressure is limited by the critical pore pressure of the dangerous fault within a range of 2 km around the injection point: ； where P pri is the critical pore pressure, σ is the normal stress on the fault plane, is the shear stress on the fault plane, μ is the friction coefficient, and Pl is the hydrostatic pressure. The maximum wellhead injection pressure P is then calculated according to the formula max : P max =P pri +△P; where P max is the maximum allowable wellhead pressure, and ΔP is the pressure loss. If the modification purpose is still not achieved after triggering the fourth early warning condition, the injection can be restarted after stopping the pump for 6 months through the above process.

7. The integrated prediction based hot dry rock reservoir reformation cross- feed multi-stage dynamic control seismic method of claim 1, wherein, In the step S6: After triggering the dynamic early warning condition in S4, the injection parameters are dynamically limited based on the relationship between the real-time injection parameters and the induced magnitude in S5. Different early warning conditions correspond to different injection parameter adjustment strategies to achieve the purpose of safe earthquake control. If the modification purpose is still not achieved after triggering the fourth early warning condition, the injection can be restarted after stopping the pump for 3 months through the above process.

8. The integrated prediction based hot dry rock reservoir reformation cross feed multi-stage dynamic control seismic method of claim 1, wherein, In the step S7: Through real-time monitoring of induced earthquakes in S2, dynamic injection limiting parameters are provided by mutual feedback analysis of S3 and S5. Dynamic early warning conditions are provided for S4 by S3 to achieve proactive earthquake control adjustment measures in S6. A multi-level dynamic earthquake control system is formed through real-time earthquake risk analysis and parameter adjustment mutual feedback, and the purpose of safe earthquake control is ultimately achieved.

Images