CN113655521B - Wave selection method based on discrete latin hypercube sampling - Google Patents

Wave selection method based on discrete latin hypercube sampling Download PDF

Info

Publication number
CN113655521B
CN113655521B CN202110788027.1A CN202110788027A CN113655521B CN 113655521 B CN113655521 B CN 113655521B CN 202110788027 A CN202110788027 A CN 202110788027A CN 113655521 B CN113655521 B CN 113655521B
Authority
CN
China
Prior art keywords
empty
seismic
subintervals
wave
equal
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
Application number
CN202110788027.1A
Other languages
Chinese (zh)
Other versions
CN113655521A (en
Inventor
季静
陈珊珊
吴梓楠
韩小雷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN202110788027.1A priority Critical patent/CN113655521B/en
Publication of CN113655521A publication Critical patent/CN113655521A/en
Application granted granted Critical
Publication of CN113655521B publication Critical patent/CN113655521B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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. analysis, for interpretation, for correction
    • G01V1/288—Event detection in seismic signals, e.g. microseismics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems

Abstract

The invention discloses a discrete latin-based hypercube sampling wave selection method which comprises the following steps: s1, selecting a seismic wave library, and then selecting the number of seismic waves and seismic wave parameters; s2, obtaining a seismic wave parameter database by calculating seismic wave parameter values; s3, obtaining a plurality of equally divided intervals according to the seismic wave parameter database; s4, dividing the seismic waves into a plurality of subintervals by the plurality of equal division intervals and numbering the subintervals; s5, dividing the multiple subintervals into non-empty subintervals and empty subintervals; s6, determining an equal division interval of the least non-empty subintervals and corresponding seismic wave parameters; s7, respectively extracting non-empty subintervals from the equal intervals of the seismic wave parameters; and S8, respectively extracting seismic waves from the non-empty subintervals. The method can fully consider the discreteness of the seismic waves and reduce the number of the selected waves, obviously reduces the calculation time of structural vulnerability analysis, improves the reliability of the result, does not depend on the experience of a user during sampling, and retains the natural characteristics of the seismic waves.

Description

Wave selection method based on discrete latin hypercube sampling
Technical Field
The invention relates to the technical field of structural seismic time-course analysis, in particular to a discrete Latin hypercube sampling wave selection method.
Background
The earthquake is a natural disaster with great harmfulness, and has important significance in the research of the earthquake-resistant vulnerability of the structure, and the reasonable selection of earthquake waves is a very important part in the earthquake-resistant analysis of the structure.
The research on the vulnerability of the structure refers to the evaluation of the seismic performance of the structure under various seismic waves. The research on the vulnerability of the structure requires that when seismic waves are selected, the discreteness of the seismic waves and the characteristics of a field are fully considered. The existing method for selecting seismic waves comprises the steps of 1, selecting waves in a selected seismic wave parameter range, for example, selecting seismic waves according to wave selection standards proposed by the building structure performance coefficient evaluation guide ATC-63 issued by the American Application Technology Council (ATC) in 2009, selecting seismic waves according to the magnitude and the range of the seismic distance, and the like. 2. Wave selection is carried out according to seismic wave response spectrums (such as standard response spectrums), and the method is insufficient in consideration of the discreteness of seismic waves. 3. By selecting 22 seismic waves recommended by ATC-63, the method cannot consider the characteristics of the field.
In order to fully consider the site characteristics and the dispersion of seismic waves, the land innovation proposes that when the structural vulnerability is researched and the seismic waves are selected, the seismic waves are selected from a reasonable seismic wave library, the more the number of the selected seismic waves is, the more comprehensive the consideration on the dispersion of the seismic waves is, and the more reasonable the seismic vulnerability of the structure can be evaluated. However, too many seismic waves are selected, which causes problems of large calculation amount, long calculation time and the like. In view of the above disadvantages, some researchers introduce the latin hypercube into the research of structural vulnerability, for example, in the research of structural vulnerability by professor of lugda of the university of harbin industry, the latin hypercube sampling is used to sample the structural sample, so as to reduce the whole vulnerability analysis calculation amount. However, Latin hypercube sampling is difficult to be popularized to sampling of seismic waves, because Latin hypercube sampling requires that all parameters are continuously distributed in respective definition domains, values can be arbitrarily taken in the definition domains, and all parameters have no correlation and can be randomly combined. The parameters of the seismic waves are distributed discretely, so that the seismic waves cannot be taken at will in a distribution interval, and the correlation among the parameters is high, so that random combination cannot be performed. In order to make up for the defects of the method, after the Latin hypercube is used for sampling the seismic wave parameter combination, the Ghotbi adjusts the seismic waves, but the adjusted index weight depends on the experience of a user seriously, and the adjusted seismic waves lose the original natural characteristics, so that the Latin hypercube is difficult to be widely applied to the wave selection process of the structural vulnerability.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a discrete Latin-based hypercube sampling wave selection method.
The purpose of the invention is realized by the following technical scheme: the discrete Latin-based hypercube sampling wave selection method comprises the following steps:
s1, selecting a seismic wave library containing M seismic waves, selecting N seismic waves from the seismic wave library, and selecting K seismic wave parameters, wherein M is larger than or equal to N, K is larger than or equal to 1, and both M, N and K are positive integers;
s2, obtaining Q seismic wave parameter values by calculating K seismic wave parameters of M seismic waves to form a seismic wave parameter database, wherein Q is M multiplied by K;
s3, obtaining P equal intervals according to the seismic wave parameter database, wherein P is N multiplied by K;
s4, dividing M seismic waves into N seismic waves by P equal intervalsKWithin a sub-interval, and for NKNumbering each of the subintervals;
s5, according to the number of seismic waves in the subintervals, dividing NKEach subinterval is marked as a non-empty subinterval and an empty subinterval;
s6, determining the minimum equally divided intervals according to the number of the non-empty subintervals of the P equally divided intervals, and determining the corresponding seismic wave parameters according to the positions of the minimum equally divided intervals;
s7, arranging the equal intervals of the seismic wave parameters in the step S6 in sequence from less to more according to the number of non-empty subintervals to obtain N sequentially arranged equal intervals;
s8, sequentially extracting 1 non-empty subinterval from the N equal subintervals in the step S7 to obtain N non-empty subintervals, and then respectively and randomly extracting 1 seismic wave from the N non-empty subintervals to obtain N seismic waves.
Preferably, step S8 includes the following steps:
s81, randomly extracting one non-empty subinterval by one equally divided interval according to the equally divided interval arrangement sequence of the step S7, and recording the marked equally divided interval of the non-empty subinterval;
s82, according to the arrangement sequence of the equal partition sections in the step S7, the other equal partition section eliminates the non-empty subinterval containing the marked equal partition section, then one non-empty subinterval is randomly extracted from the equal partition section, the marked equal partition section of the non-empty subinterval is recorded, when the equal partition section has no selectable non-empty subinterval, the selected non-empty subinterval and the marked equal partition section are abandoned, the step S81 is restarted, one non-empty subinterval is obtained, and the marked equal partition section of the non-empty subinterval is recorded; s83, repeating the step S82 to obtain N-1 non-empty subintervals, wherein the N-1 non-empty subintervals and the non-empty subintervals in the step S81 form N non-empty subintervals;
s84, randomly extracting 1 seismic wave from the N non-empty subintervals in the step S83 respectively to obtain N seismic waves.
More preferably, K is greater than or equal to 2 in step S1.
Preferably, M is greater than or equal to N in step S1K。
Preferably, in engineering application, N is 3-7.
Preferably, in the vulnerability research, the N is more than or equal to 10.
Preferably, the seismic wave parameter in step S1 is a seismic wave response spectrum maximum.
Preferably, the magnitude of the seismic wave in step S1 is greater than 6, and the ground peak acceleration of the seismic wave is above 30 gal.
Preferably, the acceleration adjustment formula of the seismic waves is
Figure BDA0003159760480000041
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention can fully consider the discreteness of seismic waves and reduce the wave selecting number by the discrete Latin hypercube sampling wave selecting method, thereby obviously reducing the calculation time of structural vulnerability analysis, improving the reliability of results, and not depending on the experience of users and reserving the natural characteristics of the seismic waves during sampling.
Drawings
FIG. 1 is a schematic flow chart of a discrete Latin hypercube sampling wave selection method according to the present invention;
FIG. 2 is a distribution diagram of seismic waves of a seismic wave library in a seismic wave parameter database in example 1;
FIG. 3 is a number and distribution diagram of non-empty subintervals and empty subintervals of example 1;
fig. 4 is a schematic diagram of non-null sub-intervals sampled by a discrete latin hypercube sampling wave selection method in embodiment 1;
FIG. 5 is a distribution diagram of seismic waves in a seismic wave parameter database, extracted based on a discrete Latin hypercube sampling method in embodiment 1;
FIG. 6 is a distribution diagram of a contrasted set of randomly extracted seismic waves in a seismic wave parameter database in example 1;
FIG. 7 is a three-dimensional distribution diagram of seismic waves of the seismic wave library in the seismic wave parameter database in example 2;
FIG. 8 shows Sa for seismic waves sampled based on the discrete Latin hypercube sampling wave selection method in example 2maxAnd SdmaxA sub-projection diagram of (a);
FIG. 9 shows Sa for seismic waves sampled based on the discrete Latin hypercube sampling wave selection method in example 2maxAnd SvmaxA sub-projection diagram of (a);
FIG. 10 shows the Sv of seismic waves sampled by the discrete Latin hypercube sampling method in example 2maxAnd SdmaxA sub-projection diagram of (a);
FIG. 11 is a three-dimensional distribution diagram of seismic waves in a seismic wave parameter database, extracted based on the discrete Latin hypercube sampling wave selection method in example 2;
FIG. 12 is a three-dimensional distribution diagram of the seismic waves in the seismic wave parameter database using random sampling of the comparison set in example 2.
Detailed Description
The objects of the present invention will be described in further detail with reference to the drawings and specific examples, which are not repeated herein, but the embodiments of the present invention are not limited to the following examples.
Example 1
In this embodiment, taking a structure located in a class II site (obtained by referring to national code "building earthquake-resistant design specification" according to the ground shear wave velocity and the cover layer thickness) and grouping earthquakes into a third group (obtained by referring to national code "building earthquake-resistant design specification" according to the area where a building is located) as an example, N ═ 6 seismic waves are selected from a seismic wave bank in which M ═ 60 seismic waves are analyzed based on the discrete latin hypercube sampling, and the wave selection method based on the discrete latin hypercube sampling includes the following steps:
s1, according to the site where the building is located and the earthquake grouping, selecting seismic levels meeting the fortification intensity and earthquake grouping requirements of the site where the structure is located from the K-NET strong earthquake database of the Japanese strong earthquake observation network KiK-NET>And 6-level earthquake wave bank, wherein the peak acceleration is more than or equal to 30gal, the earthquake is divided into 60 earthquake waves of a third group in a II-type field, and the earthquake waves are formed. Two seismic wave parameters Sa are selected according to research requirementsmax,SdmaxSelecting 6 seismic waves from the seismic waves, wherein 60 seismic waves are shared in the database, and the number of the seismic waves is more than NK＝6236 pieces of the bamboo strips meet the requirement.
S2, amplitude-modulating PGA (Peak Gound Accelation, ground Peak Acceleration) of each seismic wave to 1, namely adjusting the Acceleration of each seismic wave according to the following formula, wherein a (t) is the ground Acceleration value of the seismic wave changing along with time.
Figure BDA0003159760480000061
Wherein, a (t) in the formulaAfter adjustmentFor adjusted ground acceleration values, a (t)Before adjustmentFor ground acceleration values before adjustment, max (a (t)Before adjustment) Is the maximum ground acceleration value before adjustment.
Then calculating the seismic parameters Sa of each seismic wavemax，SdmaxThe values, namely, 2 seismic wave parameters are respectively calculated for 60 seismic waves, 120 seismic wave parameter values are obtained in total, and a seismic wave parameter database is constructed, as shown in fig. 2, each data point represents one seismic wave. The calculation formula is as follows:
acceleration response spectrum peak SamaxCalculating the formula:
Samax＝(Sa(Ti，ζ))max
acceleration response spectrum peak SdmaxCalculating the formula:
Sdmax＝(Sd(Ti，ζ))max
ti is a structural period, and the variation range is [ 0-6 ] s; zeta is the structure damping ratio, the reinforced concrete structure is 0.05, and the steel structure is 0.03; sa is the acceleration response spectrum value of the structure; sd is the shift response spectrum value of the structure.
S3, Sa according to earthquake wave parameter databasemaxThe sum SdmaxDistributing values to obtain 12 equal intervals; namely 60 seismic wave by seismic wave parameters SamaxIs divided by 6 equal intervals, 60 seismic waves are also divided by the seismic wave parameter SdmaxDividing the 6 equal intervals; the boundary values for the aliquotted intervals are tabulated below:
TABLE 1 Iso-interval boundary values for seismic parameters
Figure BDA0003159760480000062
Figure BDA0003159760480000071
S4, dividing the seismic wave library into 36 subintervals by 12 equal divisions, wherein the division result and the seismic wave distribution are shown in figure 2. The 36 subintervals are numbered. The numbering rules are as follows: the numbering rules are as follows: sdmaxThe aliquoting intervals of (A) are named as (A1), (A2), (A3), (A4), (A5), (A6), and (Sa)maxThe corresponding aliquoting intervals are designated B1, B2, B3, B4, B5, B6. Such as 2.5407 ≦ SamaxThe earthquake motion less than or equal to 3.0425 belongs to A1 equal interval, Sa is more than 3.0425maxSeismic oscillation less than or equal to 3.6069 belongs to A2 equal division interval because of Sa at boundary of equal division intervalmaxThe seismic oscillation of 3.0425 belongs to SamaxThe smaller interval a 1. The subinterval A1 and B1 are numbered as A1B1 (i.e. 2.5407 ≦ SamaxNot more than 3.0425 and not more than 0.0020 and not more than SdmaxSeismic events less than or equal to 0.0050 belong to this subinterval), and so on, as shown in FIG. 3.
S5, marking the 36 sub-intervals as non-empty sub-intervals and empty sub-intervals respectively according to the number of seismic waves in the 36 sub-intervals, wherein the calculation results are shown in Table 2, and A1B2, A2B5, A3B3, A4B6 and A5B6 are empty sub-intervals. As shown in fig. 3, the shaded area is an optional non-empty subinterval. In table 2, the total seismic waves in the partial regions are not 10, and due to rounding up in the calculation of the boundary value of the partial regions, the sub-regions to which the partial seismic waves at the boundary of the partial regions belong are changed, but the final result is not greatly influenced.
TABLE 2 number of seismic waves in each subinterval
Figure BDA0003159760480000072
Figure BDA0003159760480000081
S6, counting the number of non-empty subintervals of each equal partition interval, and obtaining the number through counting, as shown in Table 2, the minimum equal partition interval is the equal partition interval containing the minimum number of non-empty subintervals, the minimum equal partition interval in the embodiment is B6, the number of non-empty subintervals is only 4, and the corresponding seismic wave parameter is Samax。
S7, SamaxThe 6 equally divided intervals are arranged according to the number of the non-empty subintervals from small to large, and the arrangement order is B6, B5, B3, B2, B4 and B1.
S8, according to the 6 equally divided intervals of the arrangement sequence of the step S7, 1 non-empty subinterval is randomly extracted from each equally divided interval in sequence to obtain 6 non-empty subintervals, and then 1 seismic wave is randomly selected from the 6 non-empty subintervals to obtain 6 seismic waves. Step S8 specifically includes the following steps:
s81, according to Sa in step S7maxThe order of the 6 equal divisions, one non-null sub-division is extracted from the first equal division B6, the first non-null sub-division is obtained, and the marked equal division corresponding to the non-null sub-division is recorded. The non-empty sub-intervals selectable for the equal interval B6 are A1B6, A2B6, A3B6 and A6B6, the non-empty sub-intervals randomly extracted therefrom are A3B6, and the corresponding labeled equal intervals are recorded as A3 and B6.
S82, according to Sa in step S7maxIn a second aliquot B5, a non-empty subinterval is selected, resulting in a second non-empty subinterval. The selectable non-empty subintervals in the aliquot B5 include A1B5, A4B5, A5B5, and A6B5(A3B5 is a non-empty subinterval, but not selectable, since the A3 aliquot is the same as the labeled aliquot A3 in step S81), and one non-empty subinterval is randomly extracted therefrom, which is A4B5, and the corresponding labeled aliquots thereof are recorded as A4 and B5. If there are no selectable non-empty subintervals in the equal partition interval, the selected non-empty subintervals, such as A3B6 and marker equal partition A3 and B6, are discarded, the process returns to step S81 to restart, and the equal partition B6 performs random sampling of the non-empty subintervals again.
And S83, repeating the step S82, and randomly extracting non-empty subintervals in sequence in B3, B2, B4 and B1 respectively: A1B3, A5B2, A6B4, and A2B1, as shown in FIG. 4.
S84, randomly extracting 1 seismic wave from the obtained non-empty subinterval set [ A1B3, A2B1, A3B6, A4B5, A5B2 and A6B4], wherein the extraction result is shown in figure 5, and the shaded area is the selected non-empty subinterval.
6 seismic wave distribution diagrams in the comparison group are shown in FIG. 6, 6 seismic waves are randomly extracted from a seismic wave library, and the seismic waves are relatively concentrated; in the embodiment, a discrete latin hypercube sampling based wave selection method is adopted for sampling, as shown in fig. 5, the problem of sampling result data aggregation which may occur in random sampling can be avoided, so that the discreteness of seismic waves is better considered, and the method is more suitable for the wave selection process of structural vulnerability.
Example 2
Taking a structure which is located in a II-type field (obtained by consulting national standard building earthquake-resistant design specification according to the ground shear wave velocity and the covering layer thickness) and is divided into a third group (obtained by consulting national standard building earthquake-resistant design specification according to the area where a building is located) as an example, selecting N (6) seismic waves from a seismic wave bank with M (60) seismic waves for analysis based on discrete latin hypercube sampling, and selecting waves based on the discrete latin hypercube sampling method, the method comprises the following steps:
s1, according to the site where the building is located and the earthquake grouping, selecting seismic levels meeting the fortification intensity and earthquake grouping requirements of the site where the structure is located from the K-NET strong earthquake database of the Japanese strong earthquake observation network KiK-NET>And 6-level, the peak acceleration is more than or equal to 30gal, the class II field is divided into 60 seismic waves of a third group, and a seismic wave library is formed. Selecting a seismic wave library containing 60 seismic waves, selecting 6 seismic waves in the seismic wave library, and selecting 3 seismic wave parameters according to research requirements, wherein the parameters are Sdmax、SamaxAnd Svmax60 seismic waves less than NK＝63108, the requirement is not met, so the number of the seismic waves of the seismic wave bank is insufficient, but sampling can still be carried out, but the sampling difficulty is large.
S2, amplitude-modulating the PGA of each seismic wave to 1, and calculating the parameter Sd of each seismic wavemaxValue, SamaxValue sum SvmaxAnd respectively calculating 3 seismic wave parameters, namely 60 seismic waves to obtain 180 seismic wave parameter values, and constructing a seismic wave parameter database. As shown in FIG. 7, each data point represents 1 seismic wave. The calculation formula is as follows:
acceleration response spectrum peak SamaxCalculating the formula:
Samax＝(Sa(Ti，ζ))max
acceleration response spectrum peak SdmaxCalculating the formula:
Sdmax＝(Sd(Ti，ζ))max
acceleration response spectrum peak value SvmaxCalculating the formula:
Svmax＝(Sv(Ti，ζ))max
ti is a structural period, and the variation range is [ 0-6 ] s; zeta is the structure damping ratio, the reinforced concrete structure is 0.05, and the steel structure is 0.03; sa is the acceleration response spectrum value of the structure; sd is the displacement response spectrum value of the structure, and Sv is the velocity response spectrum value of the structure.
S3, according to the seismic wave parameter database obtained in the step S2, according to SdmaxValue, SamaxValue sum SvmaxThe distribution of values is equally divided into 18 equally divided intervals respectively, 60 seismic waves are equally divided by 6 equally divided intervals of each seismic wave parameter, 10 seismic waves exist in each equally divided interval, and the boundary values of the 18 equally divided intervals are shown in table 3:
TABLE 3 Iso-interval boundary values of seismic parameters
Figure BDA0003159760480000101
Figure BDA0003159760480000111
And S4, dividing the seismic wave bank into 108 subintervals according to the boundary values of the equally divided intervals in the table, and numbering the subintervals. The numbering rules are as follows: sdmaxAre designated as A1, A2, A3, A4, A5 and A6, SamaxThe corresponding aliquoting intervals are named B1, B2, B3, B4, B5 and B6, SvmaxThe corresponding aliquoting intervals are designated C1, C2, C3, C4, C5, and C6. While the subintervals at the bisecting intervals A1, B and C1 are numbered A1B1C1 (i.e., 0.0020. ltoreq. Sd)max≤0.0050，2.5407≤Samax≤3.0425，0.0408≤SvmaxSeismic events less than or equal to 0.0809 belong to the subinterval), other subintervals, and so on.
S5, calculating the number of seismic waves in 108 sub-intervals, and respectively marking the 108 sub-intervals as non-empty sub-intervals and empty sub-intervals. The number of seismic waves for non-empty subintervals is shown in Table 4, with the subintervals not listed being empty subintervals.
TABLE 4 number of seismic waves in each non-empty subinterval
Figure BDA0003159760480000112
Figure BDA0003159760480000121
S6, determining the equant interval containing the least number of the non-empty subintervals, also called the least equant interval, according to the number of the non-empty subintervals contained in the 18 equant intervals, wherein the least equant interval is C5, the contained non-empty subintervals are only 6, and the corresponding seismic wave parameter is Svmax. The number of non-empty subintervals for each aliquot is shown in Table 5.
TABLE 5 statistical table of the number of non-empty subintervals of each equal partition
Sdmax Number of seismic waves Samax Number of seismic waves Svmax Number of seismic waves
A1 9 B1 10 C1 10
A2 9 B2 8 C2 10
A3 8 B3 8 C3 8
A4 9 B4 9 C4 8
A5 8 B5 9 C5 6
A6 8 B6 7 C6 9
S7, mixing SvmaxThe 6 equally divided intervals are arranged in order of decreasing number of non-empty intervals, C5, C3, C4, C6, C2 and C1.
And S8, sequentially and randomly extracting 1 non-empty subinterval from the equally divided intervals respectively according to the arrangement sequence in the step S7, namely obtaining 6 non-empty subintervals, and randomly extracting 1 seismic wave from each non-empty subinterval.
Step S8 specifically includes the following steps:
s81, extracting non-null sub-intervals from the first equally divided interval C5 according to the equally divided interval arrangement sequence of the step S7, wherein the selectable non-null sub-intervals comprise: A3B4C5, A3B5C5, A4B1C5, A5B4C5, A6B1C5 and A6B2C5, from which non-empty subintervals A3B5C5 were randomly drawn, and their corresponding marker-aliquotes A3, B5 and C5 were recorded.
S82, according to the arrangement sequence of the equally divided intervals in the step S7, selecting non-empty subintervals from the second equally divided interval C3, wherein the selectable non-empty subintervals comprise: A1B3C3, A2B2C3, A2B4C3, A4B4C3 and A5B3C3, from which non-empty sub-intervals A6B4C3 were randomly drawn, and their corresponding marker-equivalent intervals A6, B4 and C3 were recorded.
S83, and one non-empty subinterval among the remaining four equally divided intervals C4, C6, C2 and C1, respectively, which are A1B6C4, A5B2C6, A2B3C2 and A4B1C1, respectively, as shown in fig. 8-10.
S84, randomly extracting 1 seismic wave in the non-empty subinterval set [ A4B1C1, A2B3C2, A6B4C3, A1B6C4, A3B5C5 and A5B2C6] respectively to obtain 6 seismic waves, wherein the extraction result is shown in FIG. 11.
The distribution results of 6 seismic waves in the comparison group are shown in fig. 12, 6 seismic waves are randomly extracted from the seismic wave library, and the distribution results of the 6 seismic waves are more concentrated; in the embodiment, sampling is performed by adopting a discrete latin hypercube sampling based wave selection method, as shown in fig. 11, under the condition of three-dimensional sampling, the problem of seismic wave sampling result aggregation which may occur can be well avoided, seismic oscillation can be uniformly extracted in a seismic wave bank, and the discreteness of seismic waves can be better considered.
Example 3
Other technical features in this embodiment are the same as those in embodiment 1, except that:
in this embodiment, the number N of seismic waves in step S1 is 10 instead of 6.
In this embodiment, 1 seismic wave parameter K in step S1 is used instead of 2 seismic wave parameters, and the seismic wave parameter is Samax。
Example 4
Other technical features in this embodiment are the same as those in embodiment 1, except that:
in this embodiment, the number N of seismic waves in step S1 is 3 instead of 6.
The above-mentioned embodiments are preferred embodiments of the present invention, and the present invention is not limited thereto, and any other modifications or equivalent substitutions that do not depart from the technical spirit of the present invention are included in the scope of the present invention.

Claims (8)

1. The discrete Latin-based hypercube sampling wave selection method is characterized by comprising the following steps of:
s1, selecting a seismic wave library containing M seismic waves, selecting N seismic waves from the seismic wave library, and selecting K seismic wave parameters, wherein M is larger than or equal to N, K is larger than or equal to 1, and both M, N and K are positive integers;
s2, calculating K seismic wave parameters of M seismic waves to obtain Q seismic wave parameter values to form a seismic wave parameter database, wherein Q is M multiplied by K;
s3, obtaining P equal intervals according to the seismic wave parameter database, wherein P is N multiplied by K;
s4, dividing M seismic waves into N seismic waves by P equal intervalsKWithin a sub-interval, and for NKNumbering each of the subintervals;
s5, according to the number of seismic waves in the subintervals, dividing NKEach subinterval is marked as a non-empty subinterval and an empty subinterval;
s6, determining the minimum equally divided intervals according to the number of the non-empty subintervals of the P equally divided intervals, and determining the corresponding seismic wave parameters according to the positions of the minimum equally divided intervals;
s7, arranging the equal intervals of the seismic wave parameters in the step S6 in sequence from less to more according to the number of non-empty subintervals to obtain N sequentially arranged equal intervals;
s8, sequentially extracting 1 non-empty subintervals from the N equal intervals in the step S7 to obtain N non-empty subintervals, and then respectively and randomly extracting 1 seismic wave from the N non-empty subintervals to obtain N seismic waves;
s81, randomly extracting one non-empty subinterval from one equally divided interval according to the equally divided interval arrangement sequence of the step S7, and recording the marked equally divided interval of the non-empty subinterval;
s82, according to the arrangement sequence of the equal division intervals in the step S7, the other equal division interval rejects the non-empty subintervals containing the marked equal division interval, then one non-empty subinterval is randomly extracted from the equal division interval, the marked equal division interval of the non-empty subinterval is recorded, when the equal division interval has no selectable non-empty subinterval, the selected non-empty subinterval and the marked equal division interval are abandoned, the step S81 is restarted, one non-empty subinterval is obtained, and the marked equal division interval of the non-empty subinterval is recorded;
s83, repeating the step S82 to obtain N-1 non-empty subintervals, wherein the N-1 non-empty subintervals and the non-empty subintervals in the step S81 form N non-empty subintervals;
s84, randomly extracting 1 seismic wave from the N non-empty subintervals in the step S83 respectively to obtain N seismic waves.
2. The discrete latin hypercube sampling based wave selection method of claim 1, wherein said K is greater than or equal to 2 in step S1.
3. The discrete latin hypercube sampling based wave selection method as claimed in claim 1, wherein M ≧ N in step S1K。
4. The discrete latin hypercube sampling based wave selection method of claim 1, wherein N is 3-7 in engineering applications.
5. The discrete latin-based hypercube sampling wave selection method according to claim 1, wherein N is greater than or equal to 10 in vulnerability studies.
6. The discrete latin hypercube sampling based wave selection method of claim 1, wherein said seismic wave parameter in step S1 is the seismic wave response spectrum maximum.
7. The discrete latin hypercube sampling and wave selecting method according to claim 1, wherein the magnitude of the seismic waves in step S1 is greater than 6, and the ground peak acceleration of the seismic waves is above 30 gal.
8. The discrete latin hypercube sampling based wave selection method of claim 1, wherein the acceleration of said seismic waves is adjusted by the formula
Figure FDA0003572639480000021
CN202110788027.1A 2021-07-13 2021-07-13 Wave selection method based on discrete latin hypercube sampling Active CN113655521B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110788027.1A CN113655521B (en) 2021-07-13 2021-07-13 Wave selection method based on discrete latin hypercube sampling

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110788027.1A CN113655521B (en) 2021-07-13 2021-07-13 Wave selection method based on discrete latin hypercube sampling

Publications (2)

Publication Number Publication Date
CN113655521A CN113655521A (en) 2021-11-16
CN113655521B true CN113655521B (en) 2022-06-14

Family

ID=78489346

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110788027.1A Active CN113655521B (en) 2021-07-13 2021-07-13 Wave selection method based on discrete latin hypercube sampling

Country Status (1)

Country Link
CN (1) CN113655521B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108416140A (en) * 2018-03-06 2018-08-17 福建工程学院 Time-history analysis seismic wave selection based on maximum instantaneous input energy and amplitude modulation method

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102341729A (en) * 2009-03-05 2012-02-01 埃克森美孚上游研究公司 Optimizing Reservoir Performance Under Uncertainty
FR2993310B1 (en) * 2012-07-13 2014-08-29 IFP Energies Nouvelles METHOD FOR OPERATING A GEOLOGICAL RESERVOIR FROM A MODEL OF A TANK RESERVOIR BY CALCULATING AN ANALYTICAL LAW FOR CONDITIONAL DISTRIBUTION OF UNCERTAIN PARAMETERS OF THE MODEL
CN104298540B (en) * 2014-10-23 2017-11-07 浙江大学 A kind of underlying model parameter correcting method of traffic simulation software
CN108804817A (en) * 2018-06-08 2018-11-13 天津大学 A kind of steel catenary riser is contacted to earth tired Uncertainty Analysis Method
GB201904340D0 (en) * 2019-03-28 2019-05-15 Univ Dublin Configurational energy calculation and crystal structure prediction
CN110361780B (en) * 2019-07-26 2021-01-12 西南交通大学 Seismic wave selection method based on conditional mean input energy spectrum
CN111368476B (en) * 2020-03-05 2022-02-11 中南大学 Method for evaluating applicability of high-speed rail bridge-track system under action of near-fault earthquake

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108416140A (en) * 2018-03-06 2018-08-17 福建工程学院 Time-history analysis seismic wave selection based on maximum instantaneous input energy and amplitude modulation method

Also Published As

Publication number Publication date
CN113655521A (en) 2021-11-16

Similar Documents

Publication Publication Date Title
CN106703776B (en) Fracturing parameter optimization method
Botet et al. Hierarchical model for irreversible kinetic cluster formation
Priezzhev et al. Formation of avalanches and critical exponents in an Abelian sandpile model
CN103276686A (en) Method of determining downstream design flood of cascade reservoirs
Hussain Application of the regional flood frequency analysis to the upper and lower basins of the Indus River, Pakistan
CN102033994A (en) Steering engine reliability simulation sampling method based on Markova chain Monte Carlo
CN106203935B (en) Technical capability evaluation and Postmatch method based on user-generated content and customer relationship
Nam et al. Efficient approach for damped window-based high utility pattern mining with list structure
Inaoka et al. Water erosion as a fractal growth process
CN113655521B (en) Wave selection method based on discrete latin hypercube sampling
Le Fractal geometry and applicability to biological simulation shapes for sustainable architecture design in Vietnam
Jia et al. A 3-D morphological approach on spatial form and cultural identity of ethnic mountain settlements: Case from Guizhou, China
Liu et al. Population distribution patterns and changes in China 1953–2010: A regionalization approach
Paté-Cornell Probability and uncertainty in nuclear safety decisions
CN102682119A (en) Deep webpage data acquiring method based on dynamic knowledge
Ueda Immigration in global historical perspective
CN101582085A (en) Set option method based on distributed information retrieval system
CN115115783B (en) Digital rock core construction method and system for simulating shale matrix nano-micro pores
MATĔJŮ et al. Transition to University under Communism and after its demise The role of socio-economic background in the transition between secondary and tertiary education in the Czech Republic 1948-1998
CN104657473A (en) Large-scale data mining method capable of guaranteeing quality monotony
CN105512726A (en) Reliability distribution method and apparatus based on immune genetic optimization
CN107944015A (en) Threedimensional model typical structure based on simulated annealing excavates and method for evaluating similarity
Nintunze et al. Hole drift velocity in the warped band model of GaAs
Thouvenot et al. Limit laws for ergodic processes
CN112818439B (en) Soft rock tunnel surrounding rock sub-level grading method

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