Method for constructing seismic oscillation parameter prediction model in shale gas exploitation area

Abstract

The invention discloses a construction method of a shale gas exploitation area seismic oscillation parameter prediction model, and belongs to the technical field of seismic safety evaluation. Comprising the following steps of: by supplementing seismic line data such as existing destructive earthquakes and adopting a long and short axis seismic intensity joint attenuation relation model, performing regression to obtain an intensity attenuation relation of a moderate-intensity region; based on an NGA-West2 strong earthquake database, screening bedrock records of a free site of the reference area, and establishing an earthquake motion parameter prediction equation based on the hypo-central distance; then, converting the reference region prediction equation into a hypo-central distance prediction equation suitable for a moderately strong region through an intensity mapping method; and utilizing broadband speed records around the shale gas exploitation area, and adopting a step-by-step regression method to correct the magnitude item to obtain a corrected seismic oscillation prediction equation. According to the method, the adaptability of shallow source induced earthquakes is remarkably improved, the lower limit of the earthquake magnitude is expanded to 4.0, the regional characteristics are fully considered, and the method is suitable for earthquake safety evaluation of the southern Sichuan shale gas exploitation area.

Description

Technical Field

[0001] The present invention relates to the technical field of earthquake safety assessment, and in particular to a method for constructing a seismic parameter prediction model for a shale gas mining area. Background Art

[0002] The seismic motion prediction equation is a model that uses formulas to characterize how seismic parameters (such as peak acceleration and response spectrum) vary with magnitude, distance, and site conditions, and provides estimates of their uncertainty. Determining the seismic motion prediction equation is a crucial step in earthquake safety assessments. It is primarily used to predict seismic parameters for free-bedrock sites, providing a basis for determining site seismic parameters. Because most seismic parameters (such as peak acceleration and response spectrum) increase with magnitude and decrease with distance (a phenomenon known as "attenuation"), the seismic motion prediction equation was previously often referred to as the seismic motion attenuation relationship.

[0003] Seismic parameters vary significantly across regions, with magnitude and distance. Seismic parameter attenuation models for different regions cannot be applied to each other. Commonly used seismic parameter prediction equations in my country are generally obtained through strong motion data regression or intensity data conversion, and are targeted at earthquakes in shallow crustal active areas. For seismic parameter prediction in the shale gas mining area of ​​southern Sichuan, the above methods have the following three problems:

[0004] 1. From the perspective of focal depth, the average focal depth of shallow crust earthquakes is different from that of induced earthquakes in shale gas production areas, as the focal depth of induced earthquakes is shallower. Currently, the earthquake safety assessment work commonly used in China mostly uses the earthquake motion prediction equations using the epicenter distance as the distance parameter, which does not adequately consider the focal depth and cannot meet the requirements for accurate estimation of earthquake motion parameters in shale gas production areas.

[0005] 2. From the perspective of earthquake magnitude range, the commonly used seismic motion parameter prediction equations in my country are generally applicable to magnitudes above Ms5.0, and extrapolation is used for estimates of lower magnitudes. For earthquakes in shale gas production areas, due to their shallow epicenters, even lower magnitude earthquakes can cause significant damage. Therefore, more accurate seismic motion parameter estimates for small and medium magnitudes are needed;

[0006] 3. From the perspective of data scope, the seismic motion parameter prediction equations currently commonly used in my country are usually targeted at a large seismic motion zoning range. The seismic motion parameter estimation of a small area in the southern Sichuan shale gas mining area does not fully consider the regional nature of seismic motion attenuation. Summary of the Invention

[0007] In response to the problems raised by the prior art, the purpose of the present invention is to provide a method for constructing a seismic parameter prediction model for shale gas mining areas.

[0008] To achieve the above objectives, the present invention provides the following technical solutions:

[0009] The present invention provides a method for constructing a seismic parameter prediction model for a shale gas mining area, comprising the following steps:

[0010] Step 1: Determine the intensity attenuation relationship in the moderate-strong area

[0011] The existing destructive earthquake isoseismal data were obtained and combined with the fifth-generation zoning map data to screen the earthquake event data in the moderate-strong earthquake zone. The long and short axis earthquake intensity joint attenuation relationship model was adopted. By filling in the near-field and far-field felt range points, the long and short axis data of the isoseismal lines after the near-field and far-field filling points were used. Based on the idea of ​​joint regression of the long and short axis of the ellipse, the least squares method was used for statistical regression to obtain the intensity attenuation relationship of the moderate-strong zone after the newly added data were obtained.

[0012] Step 2: Determine the prediction equation based on the reference area and the focal distance

[0013] Based on the strong earthquake data in the NGA-West2 strong earthquake database, the free-site bedrock records in the reference area were screened, and shallow-source large earthquakes in other areas were selected as supplements. The high-magnitude part was constrained, and a piecewise linear model was used to establish the focal distance prediction equation for the horizontal ground motion parameters of the bedrock site. Finally, the model coefficients were obtained through maximum likelihood regression.

[0014] Step 3: Obtain the prediction equation for the focal distance in the moderate-strong area

[0015] Using the intensity mapping method, the intensity attenuation relationship of the reference area is equivalently matched with the intensity attenuation relationship of the moderate-strong area in step 1. Combined with the ground motion equation of the reference area in step 2, the focal distance prediction equation of the moderate-strong area is obtained.

[0016] Step 4: Local data correction in shale gas mining areas

[0017] Broadband velocity records were collected around the shale gas production area, and the magnitude term of the prediction equation in step 3 was corrected using the stepwise regression method to obtain the corrected seismic motion prediction equations for the long and short axis directions of the shale gas production area.

[0018] Furthermore, in step 1, the joint attenuation relationship model of the major and minor axis earthquake intensity is:

[0019] I=a+bM+c1lg(R1+R oa )+c2lg(R2+R ob )+ε

[0020] Among them, I is the earthquake intensity, M is the magnitude, and R oa 、R obare the near-field saturation factors for intensity attenuation in the long and short axis directions, R1 and R2 are the lengths of the major and minor axes of the elliptical isoseismal line of intensity I, a, b, c1, and c2 are regression coefficients, and ε is a random variable representing uncertainty in the regression analysis, which is usually assumed to be a log-normal distribution with a mean of zero and a standard deviation of σ. In this model, the intensity of the long and short axes is equal when R = 0, while the difference in the intensity of the long and short axes is still maintained at the intermediate distance, and the isoseismal lines are also made circular in the far field.

[0021] Furthermore, in step 2, when screening data, free-field bedrock records of shallow earthquakes in active areas were retained, and the screening was performed according to the following conditions:

[0022] To remove the influence of terrain effects, only records on free surface or on or near the bottom floor of light buildings with less than 4 floors were selected;

[0023] Records with Vs30 above 500 m / s were selected as bedrock sites, which are equivalent to Class A, B, and BC sites in the NEHRP standard;

[0024] Small and medium magnitude events in other areas were removed, and to control the impact of remaining events in other areas on the attenuation properties of the model, only records within 80 km were retained;

[0025] Records with incomplete three-component information and time histories were eliminated, as were events with fewer than four records and magnitudes less than 6.5, and the distance of the dataset was controlled to be within 200 km;

[0026] The surface wave magnitude Ms of each event is determined mainly based on the GlobalCMT earthquake catalog and supplemented by the NEIC earthquake catalog.

[0027] Furthermore, in step 2, when using the piecewise linear model, the distance parameter is modified to the focal distance, and the depth term is accordingly incorporated into the model. The focal distance prediction equation is:

[0028] When M<6.5, LgSa(T)=A1+B1M-C lg(R+D exp(E*M)),

[0029] When M≥6.5, LgSa(T)=A2+B2M-C lg(R+D exp(E*M)),

[0030] Where Lg represents the logarithm with base 10, M is the surface wave magnitude, R is the focal distance, and A1, A2, B1, B2, C, D, and E are model coefficients.

[0031] Furthermore, the specific steps of correcting the local data of the shale gas mining area in step 4 include:

[0032] Step 41: Screen broadband velocity records around the shale gas mining area and remove records with limited amplitude, unknown location, and low signal-to-noise ratio. The magnitude range of the data is Ms4.0-6.0, and the focal distance is up to 200.0 km.

[0033] Step 42: Use the stepwise regression method to fit the model parameters. For earthquakes below magnitude 6.5, use the regression data to correct the magnitude term. The attenuation relationship model is:

[0034]

[0035] Where i represents the i-th earthquake, j represents the j-th record, and E ij =1 when the jth record belongs to the i-th earthquake, otherwise =0, C, D and E take the values ​​obtained in the previous step;

[0036] Step 43: Use the attenuation relationship model to regress all data and obtain the magnitude term A i , using piecewise linear function to A i Perform regression, that is:

[0037] When M≤6.5,

[0038] A i =a1+b1M i

[0039] When M>6.5,

[0040] A i =a2+b2M i

[0041] Step 44: Use the above method to perform a joint regression on the parameters of the major and minor axes, and finally obtain the earthquake parameter prediction equation for the horizontal free-bedrock site in the shale gas mining area, which is corrected using local data, including the peak acceleration PGA, the peak velocity PGV, and the response spectrum of the 5% damping ratio.

[0042] Furthermore, during regression, the following constraints are imposed:

[0043] The magnitude terms calculated from the coefficients below 6.5 and the magnitude terms calculated from the coefficients above 6.5 have the same value at Ms6.5;

[0044] The corrected coefficients of the major axis and minor axis are such that the calculated value of the seismic motion parameter of the major axis is not less than the calculated value of the seismic motion parameter of the minor axis at a magnitude of 4.0 and a focal distance of 1.0 km.

[0045] Based on the above technical solution, the embodiments of the present invention can produce at least the following technical effects:

[0046] (1) The present invention adopts a seismic motion prediction equation based on focal distance, and reasonably includes the focal depth into the input parameters. This method significantly improves the adaptability to shallow-source induced earthquakes in shale gas mining areas, overcomes the prediction bias caused by ignoring depth in existing models, and makes the seismic motion parameter estimation more consistent with actual geological conditions. It can reasonably reflect the shallow earthquake depth in shale gas mining areas in earthquake safety assessment work.

[0047] (2) The present invention uses bedrock seismic records of small and medium earthquakes around the shale gas mining area to participate in the determination of the seismic motion prediction equation, and corrects the magnitude term through the step-by-step regression method, effectively expanding the available lower limit of magnitude to 4.0, meeting the engineering assessment needs of small and medium induced earthquakes; by constraining the model parameters through local data, the prediction equation is deeply consistent with the regional characteristics of seismic motion attenuation in the shale gas mining area in southern Sichuan, avoiding the problem of insufficient applicability of large-scale partition models. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0049] Figure 1 It is a technical flow chart of the present invention;

[0050] Figure 2 This is a comparison diagram of the intensity attenuation curve of the medium-strong area of ​​the present invention and actual data;

[0051] Figure 3 is a magnitude-distance distribution diagram of the strong earthquake data of the present invention;

[0052] Figure 4 This is a graph showing the relationship between the number of available recording points in the response spectrum of the present invention and the period. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in this field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0054] The object of the present invention is achieved through the following technical solutions:

[0055] The present invention inherits the earthquake motion parameter prediction equation model commonly used in my country's current earthquake safety assessment and modifies the distance parameter to the focal distance parameter so that the prediction equation can take into account the influence of the focal depth. Using the intensity conversion method, the earthquake motion prediction equation for the focal distance of the reference area (the western United States) was converted into a earthquake motion prediction equation for the focal distance of the reference area (the major and minor axis models), which is applicable to the shale gas production area in southern Sichuan. This equation is referred to as the "basic model" below. The broadband velocity records and strong motion records in the surrounding area of ​​the region were used to complete local corrections to the "basic model" so that the final earthquake motion prediction equation can reasonably reflect the regional characteristics of the earthquake motion in the shale gas production area. The flowchart of the specific technology is shown below. Figure 1 shown.

[0056] In terms of technical process, it is mainly divided into the following four steps:

[0057] Step 1: Determination of intensity attenuation relationship in the moderate-strong area;

[0058] Step 2: Determination of the reference area based on the focal distance prediction equation;

[0059] Step 3: Conversion results of the medium-strong area based on the earthquake focal distance prediction equation;

[0060] Step 4: Correction of local data in shale gas mining areas.

[0061] Example:

[0062] 1. Determination of intensity attenuation relationship in moderate-strong areas

[0063] When fitting the intensity attenuation relationship using seismic isoseismal data, the selected data required independent measurements of both magnitude (M) and intensity (I). Therefore, only earthquakes with magnitudes measured using modern instruments were selected, while historical earthquakes whose magnitudes were calculated from epicenter intensity were discarded. Furthermore, earthquakes lacking macroscopic field survey data and for which some isoseismal lines are outlined with dashed lines were carefully selected.

[0064] The intensity data used in the fifth-generation seismic parameter zoning map is up to December 2008. The present invention supplements the collection and compilation of isoseismal data of 83 existing destructive earthquake events from January 2009 to December 2022, most of which are from the official website of the China Earthquake Administration and the "China Earthquake Yearbook", and a small part is directly from post-earthquake investigation reports and publicly published literature; in addition, the isoseismal data used in the fifth-generation map were reorganized from the "Catalogue of Modern Earthquakes in China: 1912-1990 AD, MS ≥ 4.7" and the "China Earthquake Yearbook".

[0065] According to the preliminary seismic zoning plan for the new generation of regional maps, moderate-to-strong earthquake zones include the middle Yangtze River seismic belt, the Youjiang River seismic belt, the Ordos seismic belt, the Alxa earthquake statistical area, and the Xing'an Mountains seismic belt. The southern Sichuan shale gas production area, located in the middle Yangtze River seismic belt, falls within the moderate-to-strong earthquake zone within these zoning plans.

[0066] The updated intensity data includes 55 earthquake events and 108 isoseismal lines in the moderate-to-strong earthquake zone. The combined attenuation relationship model for the major and minor axes of earthquake intensity, the same as the fifth-generation earthquake map, is used. This model ensures that the major and minor axes have equal intensity when R = 0, while maintaining the difference in intensity at intermediate distances. It also circularizes the isoseismals in the far field. The attenuation equation for the combined attenuation model is:

[0067] I=a+bM+c1lg(R1+R oa )+c2lg(R2+R ob )+ε (Formula 1)

[0068] Among them, I is the earthquake intensity, M is the magnitude, and R oa 、R ob are the near-field saturation factors of intensity attenuation in the long and short axis directions, R1 and R2 are the lengths of the major and minor axes of the elliptical isoseismal line of intensity I, a, b, c1, and c2 are regression coefficients, and ε is a random variable representing uncertainty in the regression analysis, which is usually assumed to be log-normally distributed with a mean of zero and a standard deviation of σ.

[0069] Field survey results of intensity show that the intensity at any point within the meiseisonic zone should be the same, the influence of far-field seismogenic structures disappears, and isoseismal lines tend to be circular. To reflect these characteristics, near-field supplementary points were carried out when the epicenter intensity in the meiseisonic zone was above VII and the radius of the innermost isoseismal line was greater than 5 km. Control points were supplemented within the far-field reference felt range. The felt intensity value was III-IV. The relationship between felt radius and magnitude is shown in Table 1:

[0070] Table 1 Relationship between felt magnitude and earthquake magnitude

[0071]

[0072] Using the isoseismal major and minor axis data after filling in the near and far fields, and based on the idea of ​​joint regression of the major and minor axes of the ellipse, the least squares method was used for statistical regression to obtain the intensity attenuation relationship of the moderate-strong area after the newly added data. Table 2 gives the correlation coefficients:

[0073] Table 2 Earthquake intensity attenuation coefficient (moderate-strong area)

[0074]

[0075] like Figure 2As shown in the figure, the comparison between the regression results and the actual data is shown. The red lines from bottom to top in the figure represent the intensity attenuation curves of MS=5.5 and 6.5 respectively. It can be seen from the figure that the newly fitted intensity attenuation relationship is in good agreement with the data, and can better represent the distribution of earthquake intensities of different magnitudes. At the same time, it meets the constraints of the felt range in the far field, and the regression results are reasonable and reliable.

[0076] 2. Determination of the reference area based on the earthquake source distance prediction equation

[0077] The intensity scale used by the United States is similar to the Chinese intensity scale, and the United States has a long history of strong earthquake records and rich data. Therefore, the western United States was selected as a reference area. Appropriate shallow-source large earthquakes from other parts of the world were selected as supplementary data to help constrain the high-magnitude portion. Based on strong earthquake data from the NGA-West2 strong earthquake database, through appropriate data selection and using a piecewise linear model that considers the near-field saturation characteristics of large earthquakes, a prediction equation for free surface horizontal seismic parameters (peak acceleration, peak velocity, and 5% damping ratio response spectrum) was determined for bedrock sites in the western United States. To account for the influence of focal depth, the new prediction equation uses focal distance (the distance from the focal point to the field point) as the distance parameter.

[0078] When screening data, the main purpose was to preserve free-field bedrock records of shallow earthquakes in active areas, so the screening was carried out according to the following criteria:

[0079] ① Remove the influence of terrain effects and select only records on free surfaces or on or near the bottom floor of light buildings with less than 4 floors;

[0080] ②Following the bedrock standard of the fifth generation diagram, records with Vs30 above 500 m / s are selected as bedrock sites, which are equivalent to Class A, B, and BC sites in the NEHRP standard;

[0081] ③ Earthquake events in Japan and small and medium magnitude events in other parts of the world were removed. At the same time, in order to control the impact of remaining events in other regions on the attenuation properties of the model, only records within 80 km were retained;

[0082] ④ Eliminate records with incomplete three-component information and time history, exclude events with fewer than 4 records and magnitudes less than 6.5, and control the distance of the data set to within 200 km;

[0083] ⑤ The surface wave magnitude Ms of each event is determined mainly based on the Global CMT earthquake catalog and supplemented by the earthquake catalogs of the National Earthquake Information Center (NEIC, PDE).

[0084] In actual work, aftershock events were removed and truncation effects were accounted for. The final regression dataset contained 21 events, totaling 570 records. To increase the number of statistical sample points and reflect the discrete nature of ground motion parameters, two horizontal records from the same station were treated as independent records. The available frequency of the response spectrum was controlled by the high-pass cutoff frequency during data processing. Figure 3 and Figure 4 The distribution relationship between magnitude and distance and period of regression data is given.

[0085] In order to consider the impact of focal depth and to connect with the zoning map and the commonly used models for earthquake safety assessment, the piecewise linear model is used in succession, but the distance parameter is modified to the focal distance. Correspondingly, the depth term can be incorporated into the model. The regression model is shown as follows:

[0086] When M < 6.5, LgSa(T) = A1 + B1M - C lg(R + D exp(E*M)) (Formula 2-a)

[0087] When M≥6.5, LgSa(T)=A2+B2M-C lg(R+D exp(E*M)) (Formula 2-b)

[0088] Where Lg represents the logarithm with base 10, M is the surface wave magnitude, R is the focal distance, and A1, A2, B1, B2, C, D, and E are model coefficients.

[0089] The dataset was regressed using the maximum likelihood method. Considering that the near-field saturation factor reflects the influence of source scale and should be constant for the same earthquake, the saturation factor for different periods was set to be constant during the regression. From a conservative perspective, the results at any period point were set to avoid oversaturation.

[0090] 3. Conversion results based on the focal distance prediction equation in the medium-strong area

[0091] At present, the domestic earthquake prediction equation mainly adopts the earthquake motion and earthquake intensity attenuation relationship in the western region of the United States with rich earthquake motion records, combined with the earthquake intensity data of various regions in my country, to obtain the earthquake motion attenuation relationship of each region to meet the needs of engineering. The mapping method is used for conversion, that is, for any point P (MA, RA) in the area with rich earthquake motion records (region A), there is a corresponding point Q (MB, RB) in the area with insufficient earthquake motion records (region B), so that the earthquake intensity I and earthquake motion parameter Y caused by the mapping pair P and Q in their respective regions are equal, that is:

[0092] IA(MA,RA)=IB(MB,RB) and YA(MA,RA)=YB(MB,RB) (Equation 3)

[0093] M and R represent the earthquake's magnitude and spatial distance, respectively. Using the intensity attenuation relationship and equivalent earthquake pairs, we can uniquely identify two corresponding points in the reference and target areas, where the seismic parameters are identical. The average focal depth in China and the United States is 10 km, which allows for conversion between focal distance and epicenter distance.

[0094] The attenuation relationship of earthquake intensity in the western United States is based on Chandra's research results on isoseismal data of 10 earthquakes in the San Andreas area between 1906 and 1966:

[0095] I=I0+2.014-0.00659R-2.014Lg(R+10)σ I =0.274 (Formula 4)

[0096] Where: R≤330km;

[0097] The earthquake intensity attenuation relationship in the target area adopts the result of the previous step 1, and the corresponding seismic parameter prediction equation adopts the result of step 2. Through the intensity conversion method, the prediction equation based on the focal distance in the moderate-strong area can be obtained (refer to formulas 2-a and 2-b). The results in this area are applicable to the magnitude range Ms4.0-7.0 and the focal distance 0-200km.

[0098] 4. Local data correction in shale gas mining areas

[0099] The present invention selected data from the Sichuan-Yunnan region (mainly concerned with the magnitude range of 4.0-6.0) with high density broadband stations and strong recent seismic activity around the shale gas mining area (roughly Kangding, Ya'an, Deyang, Yibin, Zhaotong, etc.). The data was acquired from January 2015 to February 2021. From them, broadband velocity records of small and medium-sized earthquakes in the near and medium fields of the study area were selected. Since the current seismic instrument installation specifications in my country require bedrock sites, this type of velocity record can be regarded as a good bedrock record. The main focus is on whether it is a free surface, underground or high-rise building, etc., and classification is performed. After obtaining the broadband velocity record, a unified data processing program is used to complete the removal of instrument response and baseline correction, and a high-pass filter with a cutoff frequency of 0.025Hz is performed, and it is differentiated to obtain the acceleration time history. Using a multi-group cross-check approach, we manually checked the velocity and acceleration waveform data and eliminated those with low signal-to-noise ratio, amplitude limiting, erroneous records, etc. We then calculated common seismic parameters such as PGA, PGV, and acceleration response spectrum with a 5% damping ratio, and used these data as the basic database.

[0100] The data of each earthquake event were individually reviewed in the basic database, and some earthquake data were removed. The reasons were:

[0101] (1) Some near-field earthquake records were clipped, making the data unusable and therefore deleted;

[0102] (2) The locations of some earthquake records were unclear and it was impossible to distinguish whether they belonged to the free surface, so they were eliminated;

[0103] (3) Some earthquake events only contained records from three or fewer stations, which were too rare, so earthquake events with only records from three or fewer stations were excluded;

[0104] (4) Some earthquake events deviated significantly from other records of the same magnitude and distance. It was initially suspected that the instrument response did not match the nominal value in the file, so they were eliminated from a conservative perspective.

[0105] (5) Due to the truncation effect, records with higher amplitudes may be deleted, while records with smaller amplitudes are systematically retained. Therefore, data with near-field amplitudes significantly smaller than those of longer-distance records for the same earthquake event are eliminated.

[0106] After completing a detailed review of the data according to the above steps, the regression database finally included a total of 93 earthquake events, 2246 broadband velocity records in the horizontal direction (the two horizontal directions were considered independent records), the data magnitude range was Ms4.0-6.0, and the focal distance was up to 200.0 km.

[0107] The model parameters are fitted using the stepwise regression method. For earthquakes below magnitude 6.5, the magnitude term is corrected using regression data. Specifically, the attenuation relationship model is written as:

[0108]

[0109] Where i represents the i-th earthquake, j represents the j-th record, and E ij =1 when the jth record belongs to the ith earthquake, otherwise =0, C, D and E take the values ​​obtained in the previous step. Use (Equation 5) to regress all data and get the magnitude term A i , using piecewise linear function to A i Perform regression, that is:

[0110] When M≤6.5,

[0111] A i =a1+b1M i

[0112] When M>6.5,

[0113] A i =a2+b2M i (Equation 6)

[0114] During regression, the following constraints are imposed:

[0115] ① The magnitude term calculated by the coefficient below 6.5 is equal to the magnitude term calculated by the coefficient above 6.5 at Ms6.5;

[0116] ② The coefficients of the major and minor axes after correction are such that at a magnitude of 4.0 and a focal distance of 1.0 km, the calculated value of the seismic motion parameters of the major axis is not less than the calculated value of the seismic motion parameters of the minor axis.

[0117] Using the above method to perform a joint regression on the major and minor axis parameters, we ultimately obtain a prediction equation for the horizontal free-bedrock ground motion parameters in the southern Sichuan shale gas development area, corrected using local data. This includes the peak acceleration (PGA), peak velocity (PGV), and the response spectrum for a 5% damping ratio. Tables 3 and 4 present the final, locally corrected parameter tables.

[0118] Table 3. Coefficients of the shale gas region ground motion parameter prediction model after local correction (source distance model, long axis direction)

[0119]

[0120] Table 4. Coefficients of the shale gas region ground motion parameter prediction model after local correction (source distance model, short axis direction)

[0121]

[0122]

[0123] To date, using newly collected and collated strong earthquake data from the western United States and newly supplemented domestic moderate-strength intensity data, a seismic parameter prediction equation for the moderate-strength region based on the focal distance model was developed through intensity conversion. Furthermore, by collecting broadband seismic records of small and moderate earthquakes surrounding shale gas production areas in the Sichuan-Yunnan region, a seismic parameter prediction equation for small and moderate earthquakes associated with these areas (horizontally in free-bedrock sites) was established through local data correction. The equation includes peak acceleration (PGA), peak velocity (PGV), and a 5% damping ratio response spectrum. The applicable range is Ms4.0-7.0 and the focal distance is 0-200 km.

[0124] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a seismic parameter prediction model for a shale gas mining area, characterized in that: The following steps are involved: Step 1: Determine the intensity attenuation relationship in the moderate-strong area The existing destructive earthquake isoseismal data were obtained and combined with the fifth-generation zoning map data to screen the earthquake event data in the moderate-strong earthquake zone. The long and short axis earthquake intensity joint attenuation relationship model was adopted. By filling in the near-field and far-field felt range points, the long and short axis data of the isoseismal lines after the near-field and far-field filling points were used. Based on the idea of ​​joint regression of the long and short axis of the ellipse, the least squares method was used for statistical regression to obtain the intensity attenuation relationship of the moderate-strong zone after the newly added data were obtained. Step 2: Determine the prediction equation based on the reference area and the focal distance Based on the strong earthquake data in the NGA-West2 strong earthquake database, the free-site bedrock records in the reference area were screened, and shallow-source large earthquakes in other areas were selected as supplements. The high-magnitude part was constrained, and a piecewise linear model was used to establish the focal distance prediction equation for the horizontal ground motion parameters of the bedrock site. Finally, the model coefficients were obtained through maximum likelihood regression. Step 3: Obtain the prediction equation for the focal distance in the moderate-strong area Using the intensity mapping method, the intensity attenuation relationship of the reference area is equivalently matched with the intensity attenuation relationship of the moderate-strong area in step 1. Combined with the ground motion equation of the reference area in step 2, the focal distance prediction equation of the moderate-strong area is obtained. Step 4: Local data correction in shale gas mining areas Broadband velocity records were collected around the shale gas production area, and the magnitude term of the prediction equation in step 3 was corrected using the stepwise regression method to obtain the corrected seismic motion prediction equations for the long and short axis directions of the shale gas production area.

2. The method for constructing a seismic parameter prediction model for a shale gas mining area according to claim 1, characterized in that: In step 1, the joint attenuation relationship model of the major and minor axis earthquake intensity is: I=a+bM+c1lg(R1+R oa )+c2lg(R2+R ob )+ε Among them, I is the earthquake intensity, M is the magnitude, and R oa 、R ob are the near-field saturation factors for intensity attenuation in the long and short axis directions, R1 and R2 are the lengths of the major and minor axes of the elliptical isoseismal line of intensity I, a, b, c1, and c2 are regression coefficients, and ε is a random variable representing uncertainty in the regression analysis, which is usually assumed to be a log-normal distribution with a mean of zero and a standard deviation of σ. In this model, the intensity of the long and short axes is equal when R = 0, while the difference in the intensity of the long and short axes is still maintained at the intermediate distance, and the isoseismal lines are also made circular in the far field.

3. The method for constructing a seismic parameter prediction model for a shale gas mining area according to claim 1, characterized in that: In step 2, during data screening, free-field bedrock records of shallow earthquakes in active areas were retained, and the screening was performed according to the following criteria: To remove the influence of terrain effects, only records on free surface or on or near the bottom floor of light buildings with less than 4 floors were selected; Records with Vs30 above 500 m / s were selected as bedrock sites, which are equivalent to Class A, B, and BC sites in the NEHRP standard; Small and medium magnitude events in other areas were removed, and to control the impact of remaining events in other areas on the attenuation properties of the model, only records within 80 km were retained; Records with incomplete three-component information and time histories were eliminated, as were events with fewer than four records and magnitudes less than 6.5, and the distance of the dataset was controlled to be within 200 km; The surface wave magnitude Ms of each event is determined mainly based on the GlobalCMT earthquake catalog and supplemented by the NEIC earthquake catalog.

4. The method for constructing a seismic parameter prediction model for a shale gas mining area according to claim 1, characterized in that: In step 2, when using the piecewise linear model, the distance parameter is modified to the focal distance, and the depth term is accordingly incorporated into the model. The focal distance prediction equation is: When M<6.5, LgSa(T)=A1+B1M-Clg(R+Dexp(E*M)), When M≥6.5, LgSa(T)=A2+B2M-Clg(R+Dexp(E*M)), Where Lg represents the logarithm with base 10, M is the surface wave magnitude, R is the focal distance, and A1, A2, B1, B2, C, D, and E are model coefficients.

5. The method for constructing a seismic parameter prediction model for a shale gas mining area according to claim 1, characterized in that: The specific steps of correcting the local data of the shale gas mining area in step 4 include: Step 41: Screen broadband velocity records around the shale gas mining area and remove records with limited amplitude, unknown location, and low signal-to-noise ratio. The magnitude range of the data is Ms4.0-6.0, and the focal distance is up to 200.0 km. Step 42: Use the stepwise regression method to fit the model parameters. For earthquakes below magnitude 6.5, use the regression data to correct the magnitude term. The attenuation relationship model is: Where i represents the i-th earthquake, j represents the j-th record, and E ij =1 when the jth record belongs to the i-th earthquake, otherwise =0, C, D and E take the values ​​obtained in the previous step; Step 43: Use the attenuation relationship model to regress all data and obtain the magnitude term A i , using piecewise linear function to A i Perform regression, that is: When M≤6.5, A i =a1+b1M i When M>6.5, <h2 style=";text-align:left;direction:ltr">A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> =a2+b2M<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; Step 44: Use the above method to perform a joint regression on the parameters of the major and minor axes, and finally obtain the earthquake parameter prediction equation for the horizontal free-bedrock site in the shale gas mining area, which is corrected using local data, including the peak acceleration PGA, the peak velocity PGV, and the response spectrum of the 5% damping ratio.

6. The method for constructing a seismic parameter prediction model for a shale gas mining area according to claim 5, characterized in that: During regression, the following constraints are imposed: The magnitude terms calculated from the coefficients below 6.5 and the magnitude terms calculated from the coefficients above 6.5 have the same value at Ms6.5; The corrected coefficients of the major axis and minor axis are such that the calculated value of the seismic motion parameter of the major axis is not less than the calculated value of the seismic motion parameter of the minor axis at a magnitude of 4.0 and a focal distance of 1.0 km.

Images