CN109143370B - Correction method for base line drift of seismic oscillation acceleration record - Google Patents
Correction method for base line drift of seismic oscillation acceleration record Download PDFInfo
- Publication number
- CN109143370B CN109143370B CN201810826680.0A CN201810826680A CN109143370B CN 109143370 B CN109143370 B CN 109143370B CN 201810826680 A CN201810826680 A CN 201810826680A CN 109143370 B CN109143370 B CN 109143370B
- Authority
- CN
- China
- Prior art keywords
- acceleration
- drift
- baseline
- time course
- baseline drift
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001133 acceleration Effects 0.000 title claims abstract description 114
- 238000000034 method Methods 0.000 title claims abstract description 98
- 238000012937 correction Methods 0.000 title claims abstract description 34
- 230000010355 oscillation Effects 0.000 title description 3
- 238000006073 displacement reaction Methods 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims description 12
- 230000010354 integration Effects 0.000 claims description 10
- 230000004083 survival effect Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 description 17
- 230000000694 effects Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000012886 linear function Methods 0.000 description 2
- 239000000919 ceramic Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to a correction method for base line drift of earthquake acceleration recording, which is characterized in that a convex optimization solving method of L1 norm regularization is used for a multi-segment linear fitting speed time course to identify the acceleration base line drift, and the speed time course and the displacement time course are recovered according to the acceleration base line drift; the acceleration baseline drift identification method specifically comprises the steps of adopting a multi-segment linear fitting drifting speed time course, taking the minimum fitting residual error and the minimum sought baseline drift as an objective function, and obtaining the baseline drift through multiple iteration solving. The method can well solve the randomness problem of baseline drift, can reduce human intervention, and enables the baseline correction method to be more objective and universal.
Description
Technical Field
The invention relates to the technical field of seismic data processing, in particular to a correction method for base line drift of seismic acceleration record.
Background
The observation of the strong seismograph is an effective means for acquiring high-precision earth surface deformation (displacement, speed and acceleration), the strong seismograph is easy to acquire high-resolution acceleration, and the speed and the displacement after the integral of the strong seismograph have deviation due to the existence of an acceleration baseline drift error. In the current more general baseline correction method, it is generally assumed that the acceleration drifts within a given period of time, and the number of drifts is also given by subjective experience (one, two or more). The velocity time course is usually obtained by integrating the original records, and the baseline drift of the acceleration time course is obtained by using a method of piecewise linear fitting velocity according to the baseline drift of the velocity time course.
However, when the seismic acceleration recorded by the strong seismic instrument drifts, the seismic acceleration is random, that is, the acceleration baseline drift is represented by a step function for realizing the number of uncertain segmentation, the subjective intervention has great influence on the baseline correction processing result, and the number and the position of the baseline drift are judged by human intervention, so that the baseline correction program is too strong in subjectivity and is not standardized.
Disclosure of Invention
The invention aims to provide a correction method for the base line drift of the earthquake acceleration record, which utilizes an L1 norm regularization optimization method to obtain the optimized base line drift by taking the goal that the variance of a multi-segment linear fitting speed time course is as small as possible and the solved base line drift is also as small as possible. The method can well solve the randomness problem of baseline drift, can reduce human intervention, and enables the baseline correction method to be more objective and universal.
In order to achieve the purpose, the technical scheme adopted by the invention is a correction method for recording the baseline drift of the earthquake acceleration, the correction method uses a method of L1 norm regularization convex optimization solution for multi-segment linear fitting speed time interval to identify the acceleration baseline drift, and the speed time interval and the displacement time interval are recovered according to the acceleration baseline drift; the acceleration baseline drift identification method specifically comprises the steps of adopting a multi-segment linear fitting drifting speed time course, taking the minimum fitting residual error and the minimum sought baseline drift as an objective function, and obtaining the baseline drift through multiple iteration solving.
In an improved technical scheme, the method comprises the following steps:
1) obtaining a speed time course by recording and integrating the original acceleration;
2) the L1 norm regularization method is used in the process of least square piecewise linear fitting speed time interval to obtain the optimized acceleration baseline drift; the L1 norm is normalized into a dual-rule formula, and the residual error is Euclid norm (| | Ax-b | |)2) Measuring, regularizing with L1 norm | | x | | luminance1Specifically, the formula is adopted:
Minimize(||Ax-b||2+λ||x||1) Carrying out correction operation;
in the formula, A is a sparse operator, b is a fitting target, namely the speed of ground motion during earthquake; x is the acceleration baseline drift of the ground motion during the earthquake; λ is a regularization parameter (non-negative);
in the formula, I | Ax-b | count hair is obtained by adjusting lambda2And sparsity of vector x to obtain an optimized acceleration baseline drift x_opt；
3) And correcting the acceleration time course according to the optimized acceleration baseline drift curve.
In one refinement, the formula uses the CVX convex optimization toolkit in MATLAB to optimize the acceleration baseline drift x \uoptAnd (6) solving.
The invention also discloses a method for verifying the effect of the correction method for the base line drift of the seismic dynamic acceleration record, which comprises the steps of firstly selecting a proper seismic dynamic acceleration record as an original record, adding artificial base line drift as noise, then processing the acceleration record after the noise is added by using the correction method, and calculating the base line drift as the artificially added known base line drift for comparison. The main improvement of this verification method is the establishment of a baseline drift noise model.
The verification method proves that the method can accurately identify the baseline drift position and the drift size.
The baseline correction method based on the L1 norm regularization does not need to artificially select a fixed time point of baseline drift in advance, depends on that the final speed is as close to zero as possible and the baseline drift is as small as possible as a target, obtains the baseline drift position and the drift size through algorithm optimization calculation, can well solve the randomness problem of the baseline drift, can reduce human intervention, and enables the baseline correction method to be more objective and universal.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without inventive efforts.
FIG. 1 is a schematic diagram illustrating an embodiment of a method for correcting a seismic dynamic acceleration recording baseline wander according to the present invention;
FIG. 2 is a schematic diagram of another embodiment of the method for correcting the base line drift of seismic dynamic acceleration recording according to the present invention;
FIG. 3 is a diagram illustrating a step of verifying the effect of the calibration method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a noise model of acceleration baseline drift in accordance with the present invention;
FIG. 5 is a graph illustrating the recognition and processing results of the two-stage baseline noise model according to the calibration method of the present invention;
FIG. 6 shows the initial partial processing results of El Centro (NS) records in the noise model of the present invention;
FIG. 7 is a graphical illustration of the results of the identification and processing of a one-segment baseline noise model by the calibration method of the present invention;
FIG. 8 is a graphical illustration of the results of the recognition and processing of a three-segment baseline noise model by the calibration method of the present invention;
FIG. 9 is a graph of the time course of velocity recovered in the noise model of the present invention for El Centro (NS) recordings with different noise models added.
Detailed Description
In order to make the purpose, technical solution and beneficial effects of the present application more clear and more obvious, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The invention relates to a correction method for recording baseline drift of earthquake acceleration, which is characterized in that an L1 norm regularization method is used in a multi-segment linear fitting speed time course process to identify the acceleration baseline drift, and the speed time course and the displacement time course are recovered according to the acceleration baseline drift; the acceleration baseline drift identification method specifically comprises the steps of adopting a multi-segment linear fitting drifting speed time course, taking the minimum fitting residual error and the minimum sought baseline drift as an objective function, and obtaining the baseline drift through multiple iteration solving.
Embodiments of the present invention are described below with reference to the drawings. As shown in fig. 1, the method mainly comprises the following steps:
1) obtaining a speed time course by recording and integrating the original acceleration;
2) mixing L withThe 1 norm regularization method is used for obtaining the optimized acceleration baseline drift in the process of least square piecewise linear fitting speed time interval; wherein the L1 norm is normalized into a dual-criterion formula, and the residual error is Euclid (| Ax-b |)2) Representing, regularization L1 norm | | | x | | luminance1Specifically, the formula is adopted:
Minimize(||Ax-b||2+λ||x||1) Carrying out correction operation;
in the formula, A is a sparse operator, b is a fitting target, namely the target speed of ground motion during earthquake to be fitted; x is the acceleration baseline drift of the ground motion during the earthquake; λ is a regularization parameter;
the L1 norm is the sum of the absolute values of the vector elements, i.e.
The fitted residual is expressed in Euclidid norm (Euclidean norm) as the sum of squares of the absolute values of the vector elements, i.e. the reopening power
In the formula, | | Ax-b | | survival rate is obtained by adjusting λ2And sparsity of vector x to obtain an optimized acceleration baseline drift x _opt(ii) a The regularization parameter lambda is used for adjusting a fitting error term | Ax-b | counting2And regularization term | | x | | non-woven ceiling1A model-dependent hyper-parameter of the relative contribution; when lambda is 0, no regularization is performed, an overfitting phenomenon occurs at the moment, and the solved x is an acceleration time course; the greater the value of λ, the greater the number of zeros (or small values) in the acceleration baseline drift vector x. In a preferred example of the present invention, λ is 1, i.e. the fitting error term and the regularization term have equally important contributions to solving the objective function minimum; the L1 norm regularization is added, so that the optimization calculation amount is effectively reduced, a more accurate speed fitting straight line is obtained, and the times and the size of baseline drift are controlled.
3) And correcting the acceleration time course according to the optimized acceleration baseline drift curve.
It should be noted that, in view of the various causes of baseline drift: 1) the acceleration baseline drift may be random, multiple times; 2) the larger the value of the acceleration is, the larger the degree of drift of the acceleration is, and conversely, the smaller the degree of drift is; 3) the drift amount can be kept unchanged after the acceleration baseline drifts at a certain moment, then the acceleration baseline changes suddenly at another moment and then is kept unchanged for a period of time, namely the acceleration drift amount is a piecewise linear function (step function) of the time; 4) the velocity is zero after the earthquake motion is finished. On the basis of the influences, for the idea of finding the acceleration baseline drift by using the commonly used least square piecewise linear fitting speed time course, the optimal baseline drift is solved by using an L1 norm regularization method, namely the formula, the method aims to obtain the | Ax-b | M | by using the method that the variance of the multi-segment linear fitting speed time course is as small as possible and the solved baseline drift x is also as small as possible2And an approximation of the optimal trade-off curve between the sparsity of the vector x, solving for x. In some examples, the above formula may employ the CVX convex optimization toolkit in MATLAB to optimize the acceleration baseline drift x _ \optAnd (6) solving. For the actual seismic acceleration signal Acc (vector of n × 1, signal sampling time interval DT), the MATLAB code is embodied as follows:
by adding the L1 norm as the regularization norm, namely, constraining the objective function, the optimization calculation amount can be effectively reduced, a more accurate speed fitting straight line can be obtained, and the times and the size of baseline drift are effectively controlled.
Fig. 2 shows another embodiment of the method for correcting the baseline drift of the acceleration record according to the invention, comprising the following steps:
1) acquisition of raw data
1.1) acquiring an acceleration signal of the target seismic wave through an acceleration recorder to obtain an acceleration record;
the acceleration recorder measures a 3-dimensional acceleration value through a 3-axis piezoelectric ceramic accelerometer, measures and records the absolute maximum value of the 3-axis acceleration at a time interval set by a user, and can calculate the maximum value, the minimum value, the average value and the 3-axis vector sum; the recorded seismic information is then sent to a data processing unit for data processing in the following steps.
1.2) to the original acceleration a1Recording the speed time course obtained by carrying out primary integration;
2) the L1 norm regularization method is used in the process of least square piecewise linear fitting speed time interval to obtain the optimized acceleration baseline drift x;
3) correcting the acceleration time course, wherein the corrected acceleration value a2Is a1-x；
3.1) identifying a baseline drift position and a drift size according to the optimized acceleration baseline drift curve;
and comparing the original acceleration recording curve with the optimized acceleration baseline drift curve, and intuitively identifying the baseline drift position and the drift size.
3.2) correcting the drift size of the position according to the corresponding position of the acceleration baseline drift to obtain a corrected acceleration time interval;
4) and respectively carrying out primary and secondary integration on the corrected acceleration time interval to obtain a corrected speed time interval and a corrected displacement time interval.
In one example of a calculation, at 0.4s seismic ground motion occurrence time, the original acceleration is recorded as 0.010924749, the original velocity is recorded as 0.0002162, the baseline x obtained according to the formula of the invention is-0.007851165, and the corrected acceleration is a1(original acceleration) -x (baseline wander), 0.018775914, and further integrated once, with corrected velocity 0.004816582 and corrected displacement 0.00081327.
The invention also checks whether the method can identify the added noise or not by identifying the manually added baseline wander noise, and verifies the effect of identifying the baseline wander by the method.
FIG. 3 shows the specific implementation steps of the effect verification of the above method of the present invention, including
(1) Selecting a proper seismic oscillation acceleration record as an original record, establishing a noise model of baseline drift for the original record, and obtaining an acceleration record added with noise;
(2) obtaining a speed time interval according to the acceleration record added with the noise, and calculating the acceleration baseline drift by using an L1 norm regularization method in the process of multi-segment linear fitting of the speed time interval;
(3) the acceleration baseline drift was compared to the manually added known baseline drift and analyzed.
In step (1), for the acceleration record, a relatively representative seismic record is selected as an initial seismic record for identifying the artificial baseline drift noise by the test text method. Corresponding data are downloaded from a strong earthquake database of the American Pacific earthquake engineering research center and are earthquake records which are subjected to baseline correction, filtering and the like. This example detailed analysis is exemplified by NS components of 129 TCU's collected in 1999, EW components of Bolu's Duzce earthquake in 1999, and NS components of El Central Array # 9's El Central Ro waves in 7.1 class Imperial Valley earthquake in 1940, with the detailed information shown in Table 1. The prior part (the arrival time of the seismic wave P) is calculated by adopting an STA/LTA method proposed by Allen (1978,1982), and the rest information is obtained from seismic metadata (NGA Flatfile (public version7.3)) in the PEER NGA-West 2.
Table 1 example seismic record information
For baseline wander noise models, the baseline wander can be roughly generalized into four categories according to the typical baseline wander noise model proposed by Akkar and Boore (2009) in discussing monte carlo-based baseline correction schemes: single-stage, two-stage, multi-stage, and diagonal. However, the most common baseline wander correction methods in the current mainstream are two-stage correction methods and improvement and correction methods thereof. Given that the common form of baseline wander, and the form of baseline wander solved by the L1 norm regularization, is a piecewise linear function (determined by the sparsity of the L1 norm), the baseline wander noise model chosen herein is single-segment, two-segment, and multi-segment, as shown in FIG. 4.
In the specific setting of the baseline wander noise model, acceleration is greater than 50gal at the first and last times (denoted t respectively1、t2And recording the final acceleration time as tend) Is most likely to have a baseline drift, and acceleration is initially from time t to time t1The time instants do not drift. Therefore, the specific method for establishing the noise model of baseline drift for the raw record in step (1) is as follows:
the single-segment noise model is [ t ]1，tend]Adding 5gal, i.e. a shown in section M1 in FIG. 4f＝5gal；
Two-stage noise model is [ t ]1，t2]Adding 5gal, [ t ]2，tend]Adding-2 gal, i.e. segment a of M2 in FIG. 4m1＝5gal，af＝2gal；
The three-segment noise model is [ t ]1，(t1+t2)/2]Adding 2gal [ (t)1+t2)/2，t2]Adding-2 gal, [ t ]2，tend]Adding 5gal, i.e. a shown in section M3 in FIG. 4m1＝2gal，am2＝-2gal，af＝5gal。
For example, the two-segment noise model has an original acceleration of-141.2073368 at a time t, where t is [ t [ ]1，t2]In the range, the acceleration after adding the noise is-136.2073368; and if it is at a certain time point t, its original acceleration is-8.78458555, t belongs to [ t2，tend]In the range, the acceleration after noise addition is-10.7848555.
After adding the baseline drift noise, the influence is not obvious in the acceleration time course (such as a1-a2 in FIG. 5), but the velocity time course and the displacement time course obtained by integration have obvious drift (such as c1-c2 in FIG. 5).
And next, processing the data obtained by combining the selected seismic acceleration record and the baseline drift noise model pairwise by using the method provided by the invention.
Baseline correction results and analysis
1. Identifying and processing results for two-segment baseline noise model
The results are shown in FIG. 5. Overall, the method has good performance in identifying the two-segment baseline noise M2, the piecewise linear fitting speed time course and the recovery displacement time course. From the results of identifying the location of the two drifts of baseline drift noise M2 and the magnitude of the degree of drift (row c in fig. 5), the baseline drift from the process of this document for recording TCU129(NS) almost completely fits the added known noise model; recording results obtained by Bolu (EW) processing can be well matched at the time points when the baseline drifts twice, and the baseline drift degree has deviation of-1.1 gal within a short period of time; while the recorded El Centro (NS) process gave slight deviations, with a maximum deviation of +1.3gal from the baseline drift after a delay of approximately 0.5 seconds from the initial position where the first drift occurred, but with accurate agreement between the position and the drift at the second drift. The reason for this is that the prior portion of these several records are significantly different (see prior portion in table 1), where TCU129(NS) record has a sufficient prior portion (20.65 seconds), bolu (ew) has a small prior portion (5.39 seconds), and El Centro (NS) record has a significant drop-head phenomenon.
As shown in FIG. 6, the El Centro (NS) recording is significantly lacking in the prior part and its initial recording part is not zero (5.8 gal mean in the first 0.8 seconds and 8.3gal mean in the first 2 seconds). The baseline drift of this initial portion was found to be 2.4gal (dotted line in fig. 6) when processed by the inventive method. That is, when the acceleration recording is performed without enough preceding part and the initial recording part is not zero, the preceding part is pulled toward the zero line as much as possible (as shown by the thin line in fig. 6, the result after the processing). This means that the more complete the prior portion of the seismic recording, the more beneficial it is to identify the exact location and magnitude of the baseline wander, and thus the more beneficial it is to recover the velocity and displacement time courses.
In fig. 5, row d, the black solid line is the displacement obtained by the acceleration recording integration of the raw data, the gray dashed line is the displacement obtained by the method performing the baseline correction process on the acceleration after adding the noise model and performing the two integrations, and the effect after recording TCU129 (the NS process is much better than that of recording El Centro (NS) and the effect after recording bolu (ew) process is between them, for recording El Centro (NS), the processed displacement deviates from the displacement of the raw data during the period from the initial moment of the displacement time course obtained by the method to the second drift of the baseline (0-30 seconds) (row d column 3 in fig. 5) because the baseline drift of the acceleration recognized by the method of the present invention in this period slightly deviates from the known drift noise drift model at the drift moment and drift degree (row c column 3 in fig. 5), but all are near the known baseline of drift, with little difference.
2. Identifying and processing results for single-and three-segment baseline noise models
For the simpler one-segment baseline wander noise model M1, as shown in fig. 7, the method also performs better on identifying baseline wander M1, fitting velocity time interval and recovery displacement time interval. When the TCU129(NS) and Bolu (EW) velocity drifts are recorded, the drift position and the drift degree of the noise model M1 are well matched, and the displacement time course can be well matched with the displacement of the original data. For the record El Centro (NS), the method lags behind the initial position where the baseline drift is identified by about 0.5 seconds, and the baseline drift is biased to +1.4gal to the maximum extent. The reason for this is also because of the absence of the prior portion, and the acceleration value of the initial recording portion is not zero.
The multi-segment mode is a more complex mode in the baseline drift, and as can be seen from fig. 8, the method of the present invention can sharply capture the drift tendency (slope change) of the speed time course, and as shown in the row c in fig. 8, one-time segment fitting is performed between every two gray dashed lines, which means that the speed fitting has more segment times, and therefore, the speed time course with the drift can be fitted more accurately. The resulting number of acceleration baseline drift segments will also be greater, as shown in row d of FIG. 8. The baseline correction method based on L1 norm regularization provided by the invention uses a convex optimization tool to continuously iterate optimization solution, and can automatically calculate the speed time course fitting line segment with the fitting error as small as possible and the appropriate number of segments. Compared with the method that the small change trend is usually ignored in manual interpretation and identification, the method is more sensitive to the drift of the speed time course, and can effectively identify the speed drift phenomenon (namely the trend change of the overall speed drift). For recording TCU129(NS) and Bolu (EW), the displacement time course processed by the method can be well matched and matched with the original displacement time course; for El Centro (NS) records lacking a prior part, the displacement time course recovered by the method of the invention has a certain deviation from the original displacement time course.
The method can well fit the drifting speed record by processing the seismic motion record of the baseline drifting noise added with single-section M1, two-section M2 and multi-section M3, thereby effectively identifying the acceleration baseline drifting position and the drifting degree. After baseline correction processing is carried out by the method, the displacement obtained by twice integration of data records TCU129(NS) and Bolu (EW) added with noise can be well matched with the initial original data displacement time course. The method of the invention has very good performance on speed time course recovery under each noise model, as shown in FIG. 9. The PGV obtained by integration of El Centro (NS) raw recordings was-29.66 cm/s, and the PGV obtained by processing the data after addition of noise models M1, M2 and M3 was-30.66 cm/s, -30.66cm/s and-30.53 cm/s.
The correction method of the invention is characterized in that an L1 norm regularization method is used, the baseline drift is not manually fixed into two sections in advance, but the appropriate number of sections is automatically optimized and selected under the conditions that the variance of the control fitting speed is as small as possible and the drift of the acceleration baseline is as small as possible, so that the subjectivity of manual intervention is reduced. The method can more sensitively and intelligently find the baseline drift, and can better and automatically identify various baseline drift noises in the recording of a certain prior part.
The above-described embodiments do not limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the above-described embodiments should be included in the protection scope of the technical solution.
Claims (6)
1. A correction method for the base line drift of the earthquake acceleration record is characterized in that the correction method uses a method of L1 norm regularized convex optimization solution in a multi-segment linear fitting speed time course process to identify the acceleration base line drift, and recovers the speed time course and the displacement time course according to the acceleration base line drift; the method specifically comprises the steps of adopting a multi-section linear fitting drifting speed time course, taking the minimum fitting residual error and the minimum sought baseline drifting as much as possible as a target function, and obtaining the baseline drifting through multiple times of iterative solution;
the method comprises the following steps:
1) obtaining a speed time course by recording and integrating the original acceleration;
2) the L1 norm regularization method is used in the process of least square piecewise linear fitting speed time interval to obtain the optimized acceleration baseline drift; the L1 norm is normalized into a dual-rule formula, and the residual error of the dual-rule formula is Euclid norm | Ax-b | | survival light2Representing, regularization L1 norm | | | x | | luminance1Specifically, the formula is adopted:
Minimize(||Ax-b||2+λ||x||1) Carrying out correction operation;
in the formula, A is a sparse operator, b is a fitting target, namely the velocity of ground motion during earthquake to be fitted; x is the acceleration baseline drift of the ground motion during the earthquake; λ is a regularization parameter;
3) and correcting the acceleration time course according to the optimized acceleration baseline drift curve.
2. The correction method of claim 1, wherein in the formula, the regularization parameter λ is 1.
3. The calibration method according to claim 1, wherein the formula is employed in MATLABCVX convex optimization toolkit versus optimized acceleration baseline drift x _optAnd (6) solving.
4. The correction method according to claim 1, characterized in that said step 1) comprises in particular:
1.1) acquiring an acceleration signal of the target seismic wave through an acceleration recorder to obtain an acceleration record;
1.2) carrying out primary integration on the original acceleration record to obtain a speed time course.
5. The correction method according to claim 1, characterized in that said step 3) comprises in particular
3.1) identifying a baseline drift position and a drift size according to the optimized acceleration baseline drift curve;
and 3.2) correcting the drift size of the position according to the corresponding position of the acceleration baseline drift to obtain a corrected acceleration time interval.
6. The calibration method of claim 1, wherein the method further comprises
4) And respectively carrying out primary integration and secondary integration on the corrected acceleration time interval to obtain a corrected speed time interval and a corrected displacement time interval.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810826680.0A CN109143370B (en) | 2018-07-25 | 2018-07-25 | Correction method for base line drift of seismic oscillation acceleration record |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810826680.0A CN109143370B (en) | 2018-07-25 | 2018-07-25 | Correction method for base line drift of seismic oscillation acceleration record |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109143370A CN109143370A (en) | 2019-01-04 |
| CN109143370B true CN109143370B (en) | 2020-03-31 |
Family
ID=64798953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810826680.0A Active CN109143370B (en) | 2018-07-25 | 2018-07-25 | Correction method for base line drift of seismic oscillation acceleration record |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109143370B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110389379B (en) * | 2019-07-12 | 2020-09-29 | 中国地震局地球物理研究所 | Near-fault seismic acceleration time-course fitting method capable of representing ground permanent displacement |
| CN111044405B (en) * | 2020-01-02 | 2022-06-28 | 西安建筑科技大学 | A Correction Method for Drift Error of Thermogravimetric Curve |
| CN112083481A (en) * | 2020-09-11 | 2020-12-15 | 杭州叙简科技股份有限公司 | Method for acquiring speed and displacement through acceleration |
| CN113049202B (en) * | 2021-03-08 | 2022-07-12 | 中国地震局工程力学研究所 | Local weighted regression correction method and system for acceleration integral displacement |
| CN114296150B (en) * | 2021-12-27 | 2024-12-20 | 河北环鼎石油设备有限责任公司 | An acceleration correction method and correction device for an array lateral instrument |
| CN115876413A (en) * | 2023-02-14 | 2023-03-31 | 济南城建集团有限公司 | Acceleration-Based Method for Estimating Bridge Displacement Under the Action of Moving Vehicles |
| CN117607967B (en) * | 2024-01-19 | 2024-03-26 | 中国建筑西南设计研究院有限公司 | Genetic algorithm-based earthquake motion baseline correction method and electronic equipment |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003056296A2 (en) * | 2001-08-30 | 2003-07-10 | Klakamp Scott L | Improved methods for determining binding affinities |
| US8301230B2 (en) * | 2008-09-25 | 2012-10-30 | Air Products And Chemicals, Inc. | Method for reducing baseline drift in a biological signal |
| CN101950034B (en) * | 2010-08-16 | 2011-11-09 | 中国石油天然气股份有限公司 | Natural potential curve baseline drift correction method for seismic inversion |
| CN103018194A (en) * | 2012-12-06 | 2013-04-03 | 江苏省质量安全工程研究院 | Asymmetric least square baseline correction method based on background estimation |
| CN107390267B (en) * | 2017-07-27 | 2019-03-01 | 西安交通大学 | A kind of synchronous seismic data attenuation compensation method for squeezing transform domain |
| CN108197345B (en) * | 2017-12-06 | 2021-03-30 | 华北水利水电大学 | A dynamic time-history analysis and modeling method for long-line water conveyance structures based on coupling theory |
| CN108267782B (en) * | 2017-12-29 | 2019-07-26 | 中国地震局地球物理研究所 | Earth's surface absolute displacement test macro and method based on machine vision and numeric value analysis |
-
2018
- 2018-07-25 CN CN201810826680.0A patent/CN109143370B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109143370A (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109143370B (en) | Correction method for base line drift of seismic oscillation acceleration record | |
| CN114998109B (en) | Hyperspectral imaging method, system and medium based on dual RGB image fusion | |
| CN108346159B (en) | A Tracking-Learning-Detection-Based Visual Object Tracking Method | |
| US10317549B2 (en) | Systems and methods for non-parametric autopicking of seismic wave features from seismic data | |
| CN117238058A (en) | Starter monitoring method for automobile based on data analysis | |
| CN117496359B (en) | Plant planting layout monitoring method and system based on three-dimensional point cloud | |
| CN112990106B (en) | Underwater object detection method, device, computer equipment and storage medium | |
| CN109446962B (en) | Method and system for detecting interannual changes in land cover | |
| Barz et al. | Maximally divergent intervals for extreme weather event detection | |
| CN106324614A (en) | New TAVI combination algorithm | |
| JP2017130067A (en) | Automatic image processing system for improving position accuracy level of satellite image and method thereof | |
| CN119272187A (en) | Anomaly detection method for satellite remote sensing application system | |
| CN1514239A (en) | A method for suppressing drift of a gas sensor | |
| CN114913086A (en) | Face image quality enhancement method based on generation countermeasure network | |
| CN119131690A (en) | A time-series remote sensing vegetation monitoring method for grassland desertification ecological restoration | |
| CN114611553B (en) | An identity detection method and system based on inertial sensor | |
| CN108492283B (en) | Hyperspectral image anomaly detection method based on band-constrained sparse representation | |
| CN117372407A (en) | Hyperspectral anomaly detection method and system based on plug-and-play tensor low-rank approximation | |
| CN110990383A (en) | A Similarity Calculation Method Based on Industrial Big Data Sets | |
| CN109709624B (en) | Method for determining flash elements of infrared detector based on LSTM model | |
| CN105184275B (en) | Infrared local face key point acquisition method based on binary decision tree | |
| CN107730535A (en) | A kind of cascaded infrared video tracing method of visible ray | |
| CN114067123A (en) | Instrument automatic identification system, identification method and computer readable storage medium | |
| CN113723179A (en) | Digital display instrument data recording method using target tracking algorithm | |
| CN116027405B (en) | A method and system for identifying gravity disturbance signals before earthquake |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |