A main aftershock induced slope slip prediction method based on multi-parameter regression

Abstract

The application discloses a main aftershock induced slope slip prediction method based on multi-parameter regression, aims to utilize a large amount of ground motion data and a Newmark rigid block method to construct an explicit expression for predicting permanent slip displacement of a slope under the action of a main aftershock from peak ground acceleration, peak velocity and slope yield acceleration. The method comprises the following steps: step 1, selecting ground motion data from a ground motion database NGA-West2 and calculating each ground motion parameter and the slope slip displacement under different slope yield accelerations; step 2, identifying optimal ground motion parameters according to effectiveness criteria and applicability criteria; step 3, obtaining a plurality of prediction models through optimal parameters and a multi-parameter nonlinear regression analysis method and then testing the prediction performance to determine a final slip prediction model; and step 4, predicting the permanent slip displacement of the slope under a given seismic scenario and slope condition. Based on the method, the application develops and discloses a main aftershock induced slope slip prediction model with good accuracy, universality and practicability, and has certain guiding significance for slope stability evaluation and seismic design under the action of a main aftershock.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of slope slip, and particularly relates to a method for predicting permanent slip displacement of a slope under the action of a main shock and aftershocks, and in particular to a method for predicting slope slip induced by a main shock and aftershocks based on multi-parameter regression. BACKGROUND

[0002] Landslide disasters caused by earthquakes can cause casualties and serious economic losses. Slope slip induced by earthquakes is an important evaluation index for evaluating slope stability. When the predicted slope slip exceeds a certain threshold, the slope is likely to be unstable, so it is necessary to predict the slip displacement of the slope under the action of an earthquake.

[0003] In order to calculate the slip displacement of the slope under the action of an earthquake, a Newmark rigid block method considering both the earthquake and the characteristics of the slope is widely used. However, this method requires time history data of the occurred ground motion acceleration for slip calculation, so it cannot evaluate the stability of the slope in the region when no earthquake has occurred. In recent years, many studies have developed empirical slip prediction models based on regression analysis methods and machine learning methods according to existing ground motion records and the Newmark method, which are used for regional earthquake-induced landslide risk assessment and rapid estimation of earthquake-induced slope slip after an earthquake. However, existing slip prediction models only focus on the damage caused by a single main shock, and ignore the further damage caused by aftershocks to the slope. In addition, the use of machine learning methods usually cannot establish an explicit expression, and the practicality for engineering is weak.

[0004] In recent years, with the advancement of technology and the increase in the number of seismic stations, a lot of measured data of main shocks and aftershocks in earthquake events have been recorded, which can be used for the study of the damage caused by main shocks and aftershocks to slopes, and this also promotes the development of empirical slip prediction models that rely on high-quality ground motion record databases. Considering that a large number of studies have found that aftershocks often exacerbate the damage to structures, and that in engineering practice, it is necessary for the prediction model to have an explicit expression and the equation form cannot be too complex, it is necessary to develop a method for predicting slope slip induced by main shocks and aftershocks based on multi-parameter regression. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art and provide a method for predicting slope slip induced by main shocks and aftershocks based on multi-parameter regression, which has high prediction accuracy and strong engineering practicality.

[0006] The technical scheme provided by the present application is as follows:

[0007] A method for predicting slope slip induced by main shocks and aftershocks based on multi-parameter regression, comprising the following steps:

[0008] Step 1, selecting seismic records, calculating seismic parameters, and considering various slope working conditions, obtaining the data set required for the development of the slip prediction model based on the Newmark rigid block method;

[0009] Step 2, identifying the optimal combination of scalar and vector seismic parameters under the main aftershock scenario according to the effectiveness criteria and applicability criteria;

[0010] Step 3, obtaining multiple prediction models through optimal parameters and multi-parameter nonlinear regression analysis, and then determining the final slip prediction model according to the prediction performance of different models through effectiveness criteria and sufficiency criteria;

[0011] Step 4, applying the obtained slip prediction model to slope engineering examples to predict the permanent slip displacement D and standard deviation σ of the slope under a given main aftershock scenario lnD .

[0012] Further, in step 1, seismic data is selected from the global seismic database NGA-West2, seismic parameters are calculated according to the seismic data, and the main aftershock induced slip displacement of different slope working conditions is calculated by the Newmark rigid block method to obtain the seismic parameter and slip data set.

[0013] Further, the seismic parameters include peak ground acceleration PGA, peak ground velocity PGV, Arias intensity Ia, average period Tm, significant duration Ds 5-75 and Ds 5-95 , whose calculation formulas are as follows:

[0014] PGA = max |a(t)|;

[0015] PGV = max |v(t)|;

[0016]

[0017]

[0018] Ds 5-75 = t(0.75Ia) - t(0.05Ia);

[0019] Ds 5-95 = t(0.95Ia) - t(0.05Ia);

[0020] In the formula, a(t) is the seismic acceleration time history; v(t) is the seismic velocity time history; t is the total duration of the seismic motion; g is the acceleration of gravity; C i is the Fourier amplitude of the entire acceleration; f i is the discrete Fourier transform frequency between 0.25 and 20 Hz.

[0021] Further, in the step 1, the seismic record satisfies the following conditions: (1) the ground motion is caused by shallow crustal fault; (2) the measured data is from free-field seismic station; (3) the main aftershock record of the same event should be the measured data of the same station to ensure the consistency of the data; (4) to avoid the influence of small earthquake magnitude on the prediction accuracy of the slip formula, therefore the main aftershock magnitude should be greater than 5Mw.

[0022] Further, in the step 1, the yield acceleration k y The value is set to the range of 0.01g to 1.0g to consider the slope of different stability working conditions.

[0023] Further, the step 2 includes the following sub-steps:

[0024] 2.1 The main earthquake optimal scalar type ground motion parameter is identified according to the Newmark slip displacement calculated from the main earthquake record and the effectiveness criterion, and the single parameter effectiveness is calculated by the following formula:

[0025] ln(D)=c1+c2ln(GM)+c3[ln(GM)] 2

[0026] In the formula, D is the Newmark slip displacement; GM is the ground motion parameter; c1 to c3 are the regression coefficients;

[0027] 2.2 The main earthquake ground motion parameters with poor effectiveness identified by the effectiveness criterion in the sub-step 2.1 are removed, then the remaining parameters are combined one by one, and the effectiveness of each parameter combination is calculated, and the double parameter effectiveness is calculated by the following formula:

[0028] ln(D)=c1+c2ln(GM1)+c3[ln(GM1)] 2 +c4ln(GM2)+c5[ln(GM2)] 2

[0029] In the formula, GM1 and GM2 are two ground motion parameters; c1 to c5 are the regression coefficients;

[0030] 2.3 The aftershock optimal scalar type ground motion parameter is identified according to the Newmark slip displacement calculated from the aftershock record and the effectiveness criterion;

[0031] 2.4 The aftershock ground motion parameters with poor effectiveness identified by the effectiveness criterion in the sub-step 2.3 are removed, then the remaining parameters are combined one by one, and the effectiveness of each parameter combination is calculated;

[0032] 2.5 Calculate the effectiveness of the Newmark sliding displacement calculated according to the main aftershock ground motion record and the parameters identified in sub-steps 2.1 and 2.3, and identify the most effective vector-type ground motion parameter combination under the main aftershock scenario;

[0033] 2.6 Substitute the Newmark sliding displacement calculated in sub-steps 2.1, 2.3 and 2.5 and the main aftershock parameters with good effectiveness into the applicability criterion to calculate the applicability of each parameter under the main, aftershock and sequence earthquake (main shock + aftershock) scenarios, and the applicability is calculated by the following formula:

[0034]

[0035] In the formula, ζ is the correction dispersion, the lower the value, the better the parameter applicability; b is the regression coefficient; β D|ΙΜ is the conditional standard deviation of the regression analysis, which is calculated by the following formula:

[0036]

[0037] In the formula, d i is the i-th peak demand; a is the regression coefficient; IM is the ground motion intensity parameter; N is the total number of displacement data;

[0038] 2.7 Identify the optimal scalar-type ground motion parameter and vector-type ground motion parameter combination under the main aftershock scenario by combining the effectiveness and applicability criteria.

[0039] Further, in step 2.1, the optimal ground motion parameter determined by the effectiveness criterion can reduce the prediction uncertainty of the model.

[0040] Further, the step 3 includes the following sub-steps:

[0041] 3.1 Substitute the optimal parameters identified into the multi-parameter nonlinear regression analysis to obtain a plurality of scalar-type slip prediction models and vector-type slip prediction models; by comparing the effectiveness of different scalar-type prediction models, the scalar-type slip prediction model with the best effectiveness is obtained as:

[0042]

[0043] In the formula, D MA is the predicted main aftershock induced slope slip; k y is the slope yield acceleration; PGA is the peak ground acceleration; subscript M represents the main shock; subscript A represents the aftershock;

[0044] By comparing the effectiveness of different vector-type prediction models, the vector-type slip prediction model with the best effectiveness is obtained as:

[0045]

[0046] In the formula, PGV is peak ground velocity;

[0047] According to the standard deviation result of the vector slip prediction model, the standard deviation σ of the vector slip prediction model is obtained lnDMA The calculation formula is as follows:

[0048]

[0049] 3.2 According to the sufficiency criterion, the prediction performance of different scalar slip prediction models and vector slip prediction models is tested, and the final main aftershock induced slope slip prediction model is screened.

[0050] Further, in step 3, the method for comparing the effectiveness of different prediction models in 3.1 is as follows: the prediction performance of different models is tested according to the effectiveness criterion; the residual error and residual error mean between the predicted slip value and the Newmark slip value are calculated, and then the standard deviation of the model is calculated, and the smaller the standard deviation, the better the effectiveness of the model.

[0051] Further, in step 3, the sufficiency criterion judgment standard in 3.2 is whether the residual error mean changes significantly with the magnitude, fault distance and peak ground velocity, if there is no significant change, it means that the model does not depend on the above three parameters, and the prediction performance of the model is better.

[0052] Compared with the prior art, the present application has the following beneficial effects:

[0053] (1) The present application provides a method for developing a slope slip prediction model under the action of main aftershocks based on a multi-parameter nonlinear regression method, and the obtained model is suitable for different seismic working conditions and slope conditions;

[0054] (2) The prediction model provided by the present application is based on a large number of measured main aftershock ground motion record data, which makes up for the deficiency of traditional displacement prediction models that do not consider the further damage caused by aftershocks to slopes;

[0055] (3) The slip prediction model provided by the present application has a simple and explicit expression, and can quickly evaluate the regional earthquake-induced landslide risk through commonly used software such as ArcGIS. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 is a specific flowchart of the present application;

[0057] Figure 2 is the distribution of the selected ground motion magnitude and fault distance in the embodiment of the present application;

[0058] Figure 3 is a Newmark slider sliding displacement calculation schematic diagram of the present application;

[0059] Figure 4 Figure 6 is a standard deviation result graph of scalar type parameters of main and aftershocks in an embodiment of the present application;

[0060] Figure 5 Figure 7 is a comparison graph of standard deviations of scalar type parameters and vector type parameters in an embodiment of the present application;

[0061] Figure 6 Figure 8 is a standard deviation result graph of vector type parameters of main and aftershocks in an embodiment of the present application;

[0062] Figure 7 Figure 9 is a standard deviation trend graph of a developed scalar slip prediction model in an embodiment of the present application;

[0063] Figure 8 Figure 10 is a standard deviation trend graph of a developed vector slip prediction model in an embodiment of the present application;

[0064] Figure 9 Figure 11 is a distribution graph of slip prediction residuals and residual mean values varying with magnitude in an embodiment of the present application;

[0065] Figure 10 Figure 12 is a distribution graph of slip prediction residuals and residual mean values varying with fault distance in an embodiment of the present application;

[0066] Figure 11 Figure 13 is a distribution graph of slip prediction residuals and residual mean values varying with peak ground velocity in an embodiment of the present application. DETAILED DESCRIPTION

[0067] The present application will be further described below in conjunction with the accompanying drawings and specific embodiments, but is not limited by the same.

[0068] As shown in Figure 1 , the present application provides a method for predicting main and aftershock induced slope slip, comprising the following steps:

[0069] Step 1, selecting ground motion records, calculating various ground motion parameters, and considering various slope working conditions, obtaining the data set required for developing the slip prediction model based on the Newmark rigid block method.

[0070] This step further comprises the following sub-steps:

[0071] 1.1 Select ground motion records from the NGA-West2 database of the Pacific Earthquake Engineering Research Center as shown in Figure 2The ground motion records should meet the following conditions: (1) the ground motion is caused by shallow crustal fault; (2) the measured data is from free-field seismic station; (3) the mainshock and aftershock records of the same event should be measured by the same station to ensure the consistency of the data; (4) to avoid the influence of small magnitude calculation results on the prediction accuracy of the slip formula, the mainshock and aftershock magnitude should be greater than 5Mw.

[0072] 1.2 Calculate the ground motion parameters of mainshock and aftershock according to the selected ground motion: peak ground acceleration (PGA), peak ground velocity (PGV), Arias intensity (Ia), average period (Tm), and significant duration Ds 5-75 and Ds 5-95 The calculation formulas are as follows:

[0073] PGA = maxa(t);

[0074] PGV = maxv(t);

[0075]

[0076]

[0077] Ds 5-75 = t(0.75Ia) - t(0.05Ia);

[0078] Ds 5-95 = t(0.95Ia) - t(0.05Ia);

[0079] In the formula, a(t) is the ground motion acceleration time history; v(t) is the ground motion velocity time history; t is the total duration of ground motion; g is the acceleration of gravity; C i is the Fourier amplitude of the entire acceleration; f i is the discrete Fourier transform frequency between 0.25 and 20 Hz; Ds 5-75 represents the time interval between 5% and 75% of the Arias intensity (Ia) value; Ds 5-95 represents the time interval between 5% and 95% of the Arias intensity (Ia) value.

[0080] 1.3 Set the slope yield acceleration k y value to the range of 0.01g to 1.0g to consider slopes with different stability conditions. Then calculate the corresponding slip value according to the k y value and the Newmark rigid block method. The Newmark block method is based on the limit equilibrium method and the wireless slope model, which simplifies the landslide body into a rigid block and calculates the cumulative sliding displacement of the block under the action of earthquake. As Figure 3 shown, first, the ground motion acceleration time history exceeding k yThe velocity time history of the slider is calculated by integrating the velocity time history of the slider calculated in step 1.4 once, and then the Newmark sliding displacement of the slider is calculated by integrating the velocity time history of the slider once.

[0081] 1.4 Remove the minimum values (e.g. <1x10 -4 cm) in the slip results which have no practical significance for engineering, to avoid affecting the prediction accuracy of the slip prediction model.

[0082] Step 2: Identify the optimal scalar and vector ground motion parameter combination under the main aftershock scenario according to the effectiveness criterion and applicability criterion.

[0083] Effectiveness criterion: The smaller the residual standard deviation calculated by regression analysis, the more effective the parameter is for the slip database used, and the smaller the uncertainty of the slip prediction model developed based on the parameter.

[0084] Applicability criterion: The smaller the applicability evaluation index of the parameter, the more applicable the parameter is for the slip database used.

[0085] This step further includes the following sub-steps:

[0086] 2.1 Identify the optimal scalar ground motion parameter of the main shock such as Figure 4 as shown in the figure, the single parameter effectiveness is calculated by the following formula:

[0087] ln(D) = c1 + c2ln(GM) + c3[ln(GM)] 2

[0088] where GM is the ground motion parameter; c1 to c3 are regression coefficients;

[0089] 2.2 Remove the main shock ground motion parameters with significantly poor effectiveness identified by the effectiveness criterion in sub-step 2.1, then combine the remaining parameters one by one, and calculate the effectiveness of each parameter combination such as Figure 5 as shown in the figure, the two-parameter effectiveness is calculated by the following formula:

[0090] ln(D) = c1 + c2ln(GM1) + c3[ln(GM1)] 2 + c4ln(GM2) + c5[ln(GM2)] 2

[0091] where GM1 and GM2 are two ground motion parameters; c1 to c5 are regression coefficients;

[0092] 2.3 Identify the optimal scalar ground motion parameter of the aftershock such as Figure 4 as shown in the figure.

[0093] 2.4 Remove the aftershock ground motion parameters with significantly poor effectiveness identified by the effectiveness criterion in sub-step 2.3, then combine the remaining parameters one by one and calculate the effectiveness of each parameter combination as follows: Figure 5 As shown.

[0094] 2.5 Based on the Newmark slip displacement calculated from the mainshock and aftershock ground motion records and the parameters with good validity identified in sub-steps 2.1 and 2.3, validity calculations are performed to identify the most effective vector ground motion parameter combination under the mainshock and aftershock scenario, as follows: Figure 6 As shown.

[0095] 2.6 Substitute the Newmark slip displacement and the effective mainshock and aftershock parameters calculated in sub-steps 2.1, 2.3, and 2.5 into the applicability criterion to calculate the applicability of each parameter under the mainshock, aftershock, and sequence earthquake (mainshock + aftershock) scenarios. Applicability is calculated using the following formula:

[0096]

[0097] In the formula, ζ is the corrected dispersion, and the lower the value, the better the applicability of the parameter; b is the regression coefficient; β D|ΙΜ The conditional standard deviation for regression analysis is calculated using the following formula:

[0098]

[0099] In the formula, d i is the i-th peak demand; a is the regression coefficient; IM is the seismic intensity parameter; N is the total number of displacement data. The applicability calculation results are shown in Table 1.

[0100] Table 1. Applicability test results for different seismic motion intensity parameters

[0101]

[0102] 2.7 The optimal combination of scalar and vector ground motion parameters under the mainshock and aftershock scenarios is identified by combining the two criteria of effectiveness and applicability.

[0103] Step 3: Several prediction models are derived through optimal parameter and multi-parameter nonlinear regression analysis. Then, the prediction performance of different models is tested to determine the final slip prediction model.

[0104] This step further includes the following sub-steps:

[0105] 3.1 The identified optimal parameters were substituted into a multi-parameter nonlinear regression analysis, resulting in several scalar and vector slip prediction models. The predictive performance of different models was tested according to the effectiveness criterion (lower model prediction uncertainty). The standard deviation of the model was calculated by determining the residuals and mean residuals between the predicted slip value and the Newmark slip value; a smaller standard deviation indicates better model effectiveness. By comparing the effectiveness of different scalar prediction models, the scalar slip prediction model with the best performance was determined to be:

[0106]

[0107] In the formula, D MA The predicted mainshock and aftershocks induced slope slippage; k y Here, M represents the slope yield acceleration; PGA represents the peak ground acceleration; subscript M indicates the mainshock; subscript A indicates the aftershock. By comparing the effectiveness of different vector-based prediction models, the vector-based slip prediction model with the best effective performance is:

[0108]

[0109]

[0110] In the formula, PGV is the peak ground velocity.

[0111] The trend of standard deviation in scalar slip prediction models is as follows Figure 7 As shown, the standard deviation trend of the vector slip prediction model is as follows: Figure 8 As shown. According to Figure 8 The standard deviation of the vector slip prediction model is obtained. for:

[0112]

[0113] 3.2 The predictive performance of different models was tested based on the sufficiency criterion (mean residuals do not change significantly with magnitude, fault distance, and peak ground velocity). This was achieved through comparison. Figure 7 and Figure 8 It can be seen that the selected scalar slip prediction model is less effective than the vector slip prediction model; therefore, only the sufficiency of the vector slip prediction model is examined. A graph showing the variation of the residual values ​​calculated in sub-step 3.1 with magnitude, fault distance, and peak ground velocity is plotted, as shown below. Figures 9-11 As shown, the residuals of the vector prediction model do not change significantly with the magnitude of the mainshock and aftershock, the fault distance, and the peak ground velocity. This prediction model can well describe the ground motion and corresponding parameters of the mainshock and aftershock. Therefore, this vector slip prediction model is selected as the final prediction model for mainshock-induced slope slip.

[0114] Step 4: Apply the obtained slip prediction model to a slope engineering example to predict the permanent slippage D and standard deviation σ of the slope under a given main shock and aftershock scenario. lnD The ground motion parameters of the site are calculated based on the required earthquake scenario and the ground motion parameter prediction equation. These values ​​are then substituted into the final slip model obtained in step 3 to predict the magnitude of slope slip induced by the mainshock and aftershock.

[0115] The model coefficients of the scalar prediction model and the vector prediction model selected in step 3 of this embodiment are calculated from the main and aftershock ground motion data used in this invention. For different ground motion data, the model coefficients obtained by regression analysis may be slightly different.

[0116] It should be noted that the above are merely preferred embodiments of the present invention and do not limit the implementation methods and protection scope of the present invention. Those skilled in the art should realize that any equivalent substitutions and obvious changes made based on the content of the present invention specification should be included within the protection scope of the present invention.

[0117] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the invention.

Claims

1. A method for predicting slope slippage induced by mainshocks based on multi-parameter regression, characterized in that, Includes the following steps: Step 1: Select ground motion records, calculate ground motion parameters for each location, and consider various slope conditions to obtain the dataset required for developing a slip prediction model based on the Newmark rigid slider method. Step 2: Identify the optimal scalar and vector ground motion parameter combinations under the mainshock and aftershock scenarios based on the effectiveness and applicability criteria. Step 2 includes the following sub-steps: 2.1 Based on the Newmark slip displacement calculated from the mainshock record and the validity criterion, the optimal scalar ground motion parameters of the mainshock are identified. The validity of a single parameter is calculated using the following formula: In the formula, D For Newmark sliding displacement; GM These are the ground motion parameters; c 1 to c 3 represents the regression coefficient; 2.2 Remove the main shock ground motion parameters with significantly poor validity identified by the validity criterion in sub-step 2.1, then combine the remaining parameters one by one and calculate the validity of each parameter combination. The validity of the two parameters is calculated using the following formula: In the formula, GM 1 and GM 2 represents two ground motion parameters; c 1 to c 5 is the regression coefficient; 2.3 Identify the optimal scalar ground motion parameters of aftershocks based on the Newmark slip displacement calculated from aftershock records and the validity criterion; 2.4 Remove the aftershock ground motion parameters with significantly poor effectiveness identified by the effectiveness criterion in sub-step 2.3, and then combine the remaining parameters one by one and calculate the effectiveness of each parameter combination; 2.5 Based on the Newmark slip displacement calculated from the mainshock and aftershock ground motion records and the parameters with better validity identified in sub-steps 2.1 and 2.3, the validity calculation is performed to identify the most effective combination of vector ground motion parameters under the mainshock and aftershock scenario; 2.6 Substitute the Newmark slip displacement and the effective mainshock and aftershock parameters calculated in sub-steps 2.1, 2.3, and 2.5 into the applicability criterion to calculate the applicability of each parameter under the mainshock, aftershock, and sequence earthquake scenarios. The applicability is calculated using the following formula: In the formula, ζ is the corrected dispersion, and the lower the value, the better the applicability of the parameter; b These are the regression coefficients; The conditional standard deviation for regression analysis is calculated using the following formula: In the formula, d i This is the i-th peak demand; a It is the regression coefficient; IM represents the seismic intensity parameter; N This represents the total number of displacement data. 2.7 The optimal combination of scalar and vector ground motion parameters under the mainshock and aftershock scenarios is identified by combining the two criteria of effectiveness and applicability; Step 3: Multiple prediction models are obtained through optimal parameter and multi-parameter nonlinear regression analysis. Then, the prediction performance of different models is tested according to the effectiveness criterion and sufficiency criterion to determine the final slip prediction model. Step 4: Apply the obtained slip prediction model to a slope engineering example to predict the permanent slip displacement of the slope under a given main shock and aftershock scenario. D and standard deviation σ lnD .

2. The method for predicting main shock-induced slope slippage based on multi-parameter regression according to claim 1, characterized in that: In step 1, ground motion data is selected from the global ground motion database NGA-West2, ground motion parameters are calculated based on the ground motion data, and the Newmark rigid slider method is used to calculate the main shock-induced sliding displacement for different slope conditions, so as to obtain ground motion parameters and sliding dataset.

3. The method for predicting slope slippage induced by main shock based on multi-parameter regression according to claim 2, characterized in that: The seismic parameters include peak ground acceleration. PGA Peak ground speed PGV Arias strength Ia Average period Tm Significant duration Ds 5-75 and Ds 5-95 The calculation formula is as follows: ； ； ； ； ； ； In the formula, The time history of ground motion acceleration; The time history of ground motion velocity; t When the total earthquake duration is reached; g It is the acceleration due to gravity; The Fourier amplitude of the entire acceleration; The discrete Fourier transform frequency is between 0.25 and 20 Hz.

4. The method for predicting main shock-induced slope slippage based on multi-parameter regression according to claim 1, characterized in that: In step 1, the earthquake record meets the following conditions: (1) the ground motion is caused by a shallow crustal fault; (2) the measured data comes from a free-field seismic station; (3) the mainshock and aftershock records of the same event should be measured data from the same station; (4) the magnitude of both the mainshock and aftershock should be greater than 5 Mw.

5. The method for predicting main shock-induced slope slippage based on multi-parameter regression according to claim 1, characterized in that: In step 1, the slope yield acceleration is... k y Values ​​are set in the range of 0.01 g to 1.0 g to account for slopes with different stability conditions.

6. The method for predicting slope slippage induced by main shock based on multi-parameter regression according to claim 1, characterized in that: In step 2.1, the optimal ground motion parameters determined by the effectiveness criterion can reduce the prediction uncertainty of the model.

7. The method for predicting slope slippage induced by mainshock based on multi-parameter regression according to claim 1, characterized in that: Step 3 includes the following sub-steps: 3.1 Substituting the identified optimal parameters into a multi-parameter nonlinear regression analysis, several scalar and vector slip prediction models were obtained. By comparing the effectiveness of different scalar prediction models, the scalar slip prediction model with the best effective performance was determined to be: In the formula, D MA The predicted main shock triggered the slope slippage; k y This refers to the slope yield acceleration; PGA Peak ground acceleration; subscript M represents the mainshock; subscript A represents aftershocks; By comparing the effectiveness of different vector-based prediction models, the vector-based slip prediction model with the best effective performance is determined to be: In the formula, PGV Peak ground speed; The standard deviation of the vector slip prediction model is obtained based on the standard deviation results. The formula for calculation is: 3.2 Based on the sufficiency criterion, the predictive performance of different scalar slip prediction models and vector slip prediction models was tested, and the final main shock-induced slope slip prediction model was selected.

8. The method for predicting main shock-induced slope slippage based on multi-parameter regression according to claim 7, characterized in that: In step 3, the method for comparing the effectiveness of different prediction models in 3.1 is as follows: the prediction performance of different models is tested according to the effectiveness criteria; the standard deviation of the model is calculated by calculating the residual and the mean residual between the predicted slip value and the Newmark slip value. The smaller the standard deviation, the better the model effectiveness.

9. The method for predicting main shock-induced slope slippage based on multi-parameter regression according to claim 7, characterized in that: In step 3, the sufficiency criterion in 3.2 is whether the mean residual changes significantly with magnitude, fault distance, and peak ground velocity. If there is no significant change, it means that the model does not depend on the above three parameters and the model prediction performance is good.

Images