CN104199106B - Seismic data residual static correction method based on ant colony algorithm - Google Patents

Seismic data residual static correction method based on ant colony algorithm Download PDF

Info

Publication number
CN104199106B
CN104199106B CN201410460617.1A CN201410460617A CN104199106B CN 104199106 B CN104199106 B CN 104199106B CN 201410460617 A CN201410460617 A CN 201410460617A CN 104199106 B CN104199106 B CN 104199106B
Authority
CN
China
Prior art keywords
static correction
residual static
value
gather
common
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
CN201410460617.1A
Other languages
Chinese (zh)
Other versions
CN104199106A (en
Inventor
罗红明
王勤耕
何光明
李忠
王栋
杨海涛
金德刚
张华�
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.)
China National Petroleum Corp
BGP Inc
Original Assignee
Geophysical Prospecting Co of CNPC Chuanqing Drilling Engineering Co Ltd
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 Geophysical Prospecting Co of CNPC Chuanqing Drilling Engineering Co Ltd filed Critical Geophysical Prospecting Co of CNPC Chuanqing Drilling Engineering Co Ltd
Priority to CN201410460617.1A priority Critical patent/CN104199106B/en
Publication of CN104199106A publication Critical patent/CN104199106A/en
Application granted granted Critical
Publication of CN104199106B publication Critical patent/CN104199106B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention provides a seismic data residual static correction method based on the ant colony algorithm. The method combines the advantages, of entirety, high efficiency, adaption and optimizing, of the ant colony algorithm, the thought of performing residual static correction value non-linear inversion in a shoot-check domain is adopted, the high imaging resolution of an input low signal-noise ratio gather through the residual static correction estimation is guaranteed, and the efficiency of the optimal value convergence of the residual static correction is improved, the seismic data residual static correction method based on the entire optimal ant colony algorithm in the shoot-check domain is formed, and a more accurate and efficient residual static correction calculation method is provided for seismic data processing.

Description

A kind of geological data residual static corrections based on ant group algorithm
Technical field
The present invention relates to petroleum gas seismic exploration technique field, more particularly, it is related to a kind of complete based on big gun-inspection domain The geological data residual static corrections of the optimum ant group algorithm of office, are mainly used at the field seismic data of oil seismic exploration Understanding is released.
Background technology
Statics estimation be extremely important in seismic data process and and its key technological means, and complex near surface conditionss Under residual static correction problem be an extremely complex nonlinear inverse problem, difficulty is big, and research work is few, does not obtain for a long time Solve very well.Develop to the earth's surface complex area such as Gobi desert, loess tableland, mountain region with China's seismic prospecting, static correction problem becomes Task in the urgent need to address.After the static correction in field, remaining static correction value can use traditional original seismic data Method is asked for.But when the Value of residual static correction of seismic data is big, signal to noise ratio is low or Value of residual static correction is more than seismic wave waveform During 1/2 cycle, cycle skipping phenomenon occurs using traditional residual static corrections, affects the imaging effect of seismic data, Have to look for new resolving ideas.
The research starting residual static corrections from 20 century 70s is paid close attention to by numerous scholars.Successively propose point The generalized linear inversion method of shot point and receiver static correction amount, Statistic Correlation, maximum energy method etc. is calculated during solution travelling.For Value of residual static correction is substantially a nonlinear inverse problem, and by the simulated annealing of global optimization's method, heredity is calculated Method is introduced in the calculating of Value of residual static correction, and preferably combines the advantage of simulated annealing and genetic algorithm it is proposed that comprehensive Merging rows genetic algorithm.These methods are essentially all to be calculated in prestack common midpoint gather.The conventional quiet school of residue The calculating of positive quantity based on the assumption that: be 1. Value of residual static correction random distribution principle, in cmp road collection (i.e. common midpoint gather) After superposition, the impact of Value of residual static correction can eliminate.2. it is earth's surface-consistent principle, that is, shot statics and geophone station are quiet Correcting value is only relevant with landscape position.The residual static correction algorithm such as Statistic Correlation, maximum energy method and simulated annealing is all Assumed based on this 2 points.But to low signal-to-noise ratio, big Value of residual static correction, (for example, Value of residual static correction is more than seismic wave waveform 1/2 cycle) seismic data, application is this conventional cannot to solve this kind of data by the method that cmp road collection obtains model trace Static correction problem.
Content of the invention
For the deficiencies in the prior art, an object of the present invention is to solve present in above-mentioned prior art One or more problems.For example, an object of the present invention is to provide one kind to be capable of to low signal-to-noise ratio, remaining greatly quiet school The method that the seismic data of positive quantity (for example, Value of residual static correction is more than 1/2 cycle of seismic wave waveform) carries out residual static correction.
To achieve these goals, the invention provides a kind of geological data residual static correction side based on ant group algorithm Method.Methods described includes the first operation or the second operation, and wherein, described first operation is included single-shot geological data through field Common midpoint gather gather after static correction and dynamic correction is common-shot-gather, and described common-shot-gather is carried out successively with Lower process:
Step a:
A1, it is that the every random together of common-shot-gather generates one and initially remain according to the scope of default Value of residual static correction Remaining static correction value, forms the initial disaggregation of common-shot-gather Value of residual static correction;A2, according to the remaining quiet school of described common-shot-gather The initial disaggregation of positive quantity is simultaneously scanned for using ant group algorithm, obtains the optimal solution set of common-shot-gather Value of residual static correction and answers For residual static correction is carried out together to the every of common-shot-gather.
Step b:
Judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for final end Only condition, then will carry out the data after residual static correction process using the optimal solution set of described common-shot-gather Value of residual static correction Gather is common detector gather, and execution step c.
Step c:
C1, according to the scope of default Value of residual static correction be common detector gather every together random generate one initial Value of residual static correction, forms the initial disaggregation of common detector gather Value of residual static correction;C2, according to described common detector gather remain The initial disaggregation of remaining static correction value is simultaneously scanned for using ant group algorithm, obtains the optimum of common detector gather Value of residual static correction Disaggregation simultaneously is applied to carry out residual static correction together to the every of common detector gather.
Step d:
Judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for final end Only condition, then return execution step a;Described second operation is included single-shot geological data after field statics and dynamic correction Common midpoint gather gather be common detector gather, and described common detector gather is carried out successively with following process:
Step a ': operate using with the step c identical in the described first operation.
Step b ': judge whether to meet final end condition: if meeting final end condition, stop computing;If discontented The final end condition of foot, then will be carried out at residual static correction using the optimal solution set of described common detector gather Value of residual static correction Data sorting after reason is common-shot-gather, and executes described step c '.
Step c ': operate using with the step a identical in the described first operation.
Step d ': judge whether to meet final end condition: if meeting final end condition, stop computing;If discontented The final end condition of foot, then return execution step a '.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step a2 Including:
A21, the n-th solution that m ant be respectively placed in common-shot-gather Value of residual static correction concentrate corresponding road residue quiet On the node that correcting value is located, carry out the t time loop iteration (abbreviation iteration) in big gun domain, wherein, m is equal to common-shot-gather Total road number, the initial value of t and n is 1, and as n=1, the n-th disaggregation is the first of described common-shot-gather Value of residual static correction Beginning disaggregation.
A22, using n-th solution set pair common-shot-gather carry out the t time residual static correction, calculate by Equation 1 below and compare Amplitude energy e compared with the superposition road of the common-shot-gather after the t time residual static correctiontAfter the t-1 time residual static correction Common-shot-gather superposition road amplitude energy et-1, wherein, as t=1, e0Before first time residual static correction step Common-shot-gather superposition road amplitude energy,
Equation 1:
e t = σ k = 1 m σ j = 1 t d j , k t
In equation 1, m is total road number of common-shot-gather, and t is total number of samples of seismic channel,It is that the t time residue is quiet The amplitude energy of k-th sampling point in jth road of the common-shot-gather after correction.
Work as et≤et-1When, then the n-th solution is concentrated the Value of residual static correction in each road to carry out random perturbation and calculated according to ant colony The probability branch mode of method, to select the Value of residual static correction after every one disturbance in common-shot-gather, forms common-shot-gather and remains (n+1)th disaggregation of remaining static correction value, wherein, the probability branch mode of described ant group algorithm is:
First, the kth ant positioned at node i during the t time iteration is calculated according to equation 2 below and transfer to next node The transition probability of jWherein, node i is the node corresponding to Value of residual static correction that the n-th solution concentrates kth road, and node j is N-th solution concentrates the node corresponding to Value of residual static correction that the Value of residual static correction in kth road obtains after random perturbation.
Equation 2:
In equation 2, τijT () represents that all ants are retained in the letter to j node path for the i-node in the t time iteration Breath element amount, τisT () represents that all ants are retained in the pheromones amount to s node path for the i-node, i-node in the t time iteration Represent other all optional paths, η in addition to i-node to j node path to s node pathijT () is heuristic function, be i-node To the inverse of j node Euclidean distance, ηisT () is the inverse to s node Euclidean distance for the i-node, α represents transfer track Importance, β represents the relative importance of visibility.
Then, compareWith default acceptance probability pacp, judge whether that accepting ant k transfers to jth from the i-th node Node:
WhenWhen, then accept the residue that the n-th solution concentration kth road Value of residual static correction is obtained after disturbance Static correction value concentrates the Value of residual static correction in kth road as the (n+1)th solution;WhenWhen, then refuse to concentrate the n-th solution The Value of residual static correction that kth road Value of residual static correction obtains after disturbance concentrates the residual static correction in kth road as the (n+1)th solution Amount, and again concentrate the Value of residual static correction in kth road to carry out disturbance the n-th solution, then turned according to the probability of described ant group algorithm Shifting mode judge whether to accept to concentrate the n-th solution the kth road Value of residual static correction Value of residual static correction after disturbance again as (n+1)th solution concentrates the Value of residual static correction in kth road, until
Finally, after all ants all complete the selecting and form the (n+1)th disaggregation of next node, fresh information element, make t =t+1, n=n+1 simultaneously return execution step a21.
Work as et> et-1When, then using the n-th disaggregation as common-shot-gather Value of residual static correction current optimal solution set;a23、 Determine whether to meet big gun domain end condition: if being unsatisfactory for big gun domain end condition, the n-th solution is concentrated with the residual static correction in each road The disaggregation that amount carries out random perturbation and constituted disturbance Hou Ge road Value of residual static correction is remaining directly as common-shot-gather (n+1)th disaggregation of static correction value, makes t=t+1, and n=n+1 simultaneously returns execution step a21;If meeting big gun domain end condition, by N disaggregation is as the optimal solution set of common-shot-gather Value of residual static correction.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step The big gun domain end condition of a23 is: after determining whether t reaches big gun domain maximum iteration time and determine the t time residual static correction The amplitude energy e in the superposition road of common-shot-gather datatWith the common-shot-gather data after the t-1 time residual static correction step Superposition road amplitude energy et-1Difference whether reach minimum, work as et-et-1It is not reaching to minimum and t is not up to big gun domain maximum During iterations, judge not meeting big gun domain end condition;Work as et-et-1Reach minimum or t reaches big gun domain maximum iteration time When, judge to meet big gun domain end condition;Wherein, described step a21 and a22 are repeated several times and cannot obtain and compare et-et-1Less Difference when judge et-et-1Reach minimum.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step A22 carries out Pheromone update according to equation 3 below.
Equation 3:
τij(t+1)=(1- ζ) τij(t)+δτij(t)
In equation 3,ζ is pheromones attenuation coefficient, ζ=ρt, and ρ It is 0~1 constant, t is iterations.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, its feature exists In described step c2 includes:
C21, the n-th ' solution that m ' ant be respectively placed in common detector gather concentrate corresponding road Value of residual static correction institute Node on, m ' be equal to common detector gather total road number, carry out the secondary loop iteration of t ' in industry disturbance in geophone domain, wherein, t ' It is 1 with the initial value of n ', and when n '=1, the n-th ' disaggregation of common detector gather is that described common detector gather is remaining The initial disaggregation of static correction value.
C22, using the n-th ' solution set pair common detector gather carry out the secondary residual static correction of t ', calculated using described equation 1 And compare the amplitude energy e in the superposition road of the common detector gather after the secondary residual static correction of t 't' quiet with -1 residue of t ' The amplitude energy e in the superposition road of the common detector gather after correctiont-1', wherein, e0' for first time residual static correction step it The amplitude energy in the superposition road of front common detector gather, works as et′≤et-1' when, then concentrate the residue in each road quiet the n-th ' solution Correcting value carries out random perturbation and is selected every one in common detector gather according to the probability branch mode of described ant group algorithm Value of residual static correction after disturbance, forms the n-th '+1 disaggregation of common detector gather Value of residual static correction, and fresh information element, order T '=t '+1, n '=n '+1 simultaneously returns execution step c21;Work as et' > et-1' when, then using the n-th ' disaggregation as common detector gather The current optimal solution set of Value of residual static correction.
C23, determine whether to meet industry disturbance in geophone domain end condition: if being unsatisfactory for industry disturbance in geophone domain end condition, solve to the n-th ' Every Value of residual static correction together is concentrated to carry out random perturbation the disaggregation being constituted disturbance Hou Ge road Value of residual static correction Directly as the n-th '+1 disaggregation of common detector gather Value of residual static correction, make t '=t '+1, n '=n '+1 simultaneously returns and executes step Rapid c21;If meeting industry disturbance in geophone domain end condition, using the n-th ' disaggregation as the Value of residual static correction of common detector gather optimum Disaggregation.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step The industry disturbance in geophone domain end condition of c23 is: determines whether t ' reaches industry disturbance in geophone domain maximum iteration time and determine that the secondary residue of t ' is quiet The amplitude energy e in the superposition road of the common detector gather data after correctiont' and -1 residual static correction step of t ' after The amplitude energy e in the superposition road of common detector gather datat-1' difference whether reach minimum, work as et′-et-1' be not reaching to Minimum and when t ' is not up to industry disturbance in geophone domain maximum iteration time, judge not meeting industry disturbance in geophone domain end condition;Work as et′-et-1' reach To minimum or t ' reach industry disturbance in geophone domain maximum iteration time when, judge to meet industry disturbance in geophone domain end condition;Wherein, it is repeated several times Described step c21 and c22 and cannot obtain and compare et′-et-1E is judged during ' less differencet′-et-1' reach minimum.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step b With d ' inclusion: determine whether to reach default run time and determine the common-shot-gather data after the t time residual static correction Superposition road amplitude energy etAmplitude with the superposition road of the common-shot-gather data after the t-1 time residual static correction step Energy et-1Difference whether reach pre-established threshold, wherein, work as et-et-1More than pre-established threshold and not up to default run time When, judge to be unsatisfactory for final end condition;Work as et-et-1Less than or equal to pre-established threshold or when reaching default run time, Judge to meet final end condition.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described step d With b ' inclusion: determine whether to reach default run time and determine the common detector gather number after the secondary residual static correction of t ' According to superposition road amplitude energy et' and -1 residual static correction step of t ' after common detector gather data superposition road Amplitude energy et-1' difference whether reach pre-established threshold, wherein, work as et′-et-1' more than pre-established threshold and not up to default Run time when, judge be unsatisfactory for final end condition;Work as et′-et-1' be less than or equal to pre-established threshold or reach default Run time when, judge meet final end condition.
According to an embodiment based on the geological data residual static corrections of ant group algorithm for the present invention, described process is moved The Value of residual static correction of the geological data after correction and field statics is more than 1/2 cycle or the signal to noise ratio of seismic wave waveform Low.Here, low signal-to-noise ratio is exactly the continuous reflection lineups that the geological data gathering does not compare, and is usually expressed as mixed and disorderly ripple Shape.
Compared with prior art, the beneficial effect comprise that the present invention fully absorbs based on overall efficient self-adapted The advantage of the ant group algorithm of optimizing, using the thinking carrying out Value of residual static correction non-linear inversion in big gun-inspection domain, by big gun- Self-adaptive genetic operator of taking turns are carried out on the dynamic correction road collection in inspection domain calculate more, obtain optimum big gun-cautious Value of residual static correction, and By determination strategy and the pheromone update strategy of adaptive adjustment searching route select probability, not only ensure that remaining quiet school Just estimate the high imaging resolution to input low signal-to-noise ratio road collection, highlight the efficiency of residual static correction optimal value convergence again, Form the geological data residual static corrections based on big gun-Jian Yu global optimum ant group algorithm, provide more for seismic data process Fine and efficient residual static correction computational methods.
Brief description
By the description carrying out below in conjunction with the accompanying drawings, the above and other purpose of the present invention and feature will become more clear Chu, wherein:
Fig. 1 shows the geological data residual static corrections based on ant group algorithm according to an exemplary embodiment of the present invention Flow chart;
Fig. 2 show according to an exemplary embodiment of the present invention in common-shot-gather utilize ant group algorithm according to common-source point The flow chart that the initial disaggregation of the Value of residual static correction of road collection obtains the optimal solution set of Value of residual static correction of common-shot-gather;
Fig. 3 shows and according to an exemplary embodiment of the present invention utilizes ant group algorithm in common detector gather according to altogether inspection The initial disaggregation of the Value of residual static correction of wave point road collection obtains the stream of the optimal solution set of Value of residual static correction of common detector gather Cheng Tu.
Specific embodiment
Hereinafter, accompanying drawing will be combined and exemplary embodiment describes in detail according to the present invention based on ant group algorithm Geological data residual static corrections.
First, involved technical term in the simple elaboration present invention.
Common-shot-gather: same shot point excites, it is being total to of this shot point that all roads that different geophone stations receive form road collection Shot point road collection.
Common detector gather (is also called common receiving point gather): different shot points excite, all roads that same geophone station receives Form the common detector gather that road collection is this geophone station.
Residual static correction: because technical reason or some human factors are so that fieldwork data is often inadequate Accurately, therefore after having carried out field statics still remaining Value of residual static correction.Extract Value of residual static correction the mistake being corrected for Journey is residual static correction.
Ant group algorithm: be a kind of bionical quasi-nonlinear optimized algorithm, there is concurrency, positive feedback and global minimal search The features such as ability is strong.The mechanism of ant group algorithm is: the ant in living nature when search of food source, on the path that can pass by it Discharge a kind of ant distinctive secretion pheromones so that other ants a range of can be perceived and affect its row For.When the ant passed by some paths gets more and more, this pheromones track staying is also more, so that pheromones intensity Increasing (certainly, passage in time can gradually weaken), making ant later select the probability in this path also higher, thus more increasing The pheromones intensity in this path.The biological synergetic mechanism component that ant colony is just located against this inside forms their things Front and incognizant shortest path.
In search procedure, every human oasis exploited independently works ant group algorithm in the following way: from starting point, Select next node according to certain probability transition rule, after all ants all complete the selection of next node, to all Pheromones intensity on path is updated.
According to earth's surface condition for consistence, after dynamic correction, total static correction value all decomposable asymmetric choice net of any one road earthquake record For shot point, geophone station, CMP and construction item, totally 4 sums, that is, meet equation below (1):
t abe = s a + r b + c e h ab 2 + y e - - - ( 1 )
In equation (1), tabeIt is total static correction value, saFor the Value of residual static correction of a position shot point, rbFor the detection of b position The Value of residual static correction of point, ceFor the Value of residual static correction of CMP e position, geophone offset habExamine for a position shot point to b position The horizontal range of wave point, yeIt is construction item Value of residual static correction.
Select computation window that the impact of construction item can be tried one's best before carrying out residual static correction on section to eliminate.Due to ground The non-double curve factor of layer and hyperbola formula move correction, casting out of its Taylor expansion higher order term, are just causing dynamic correction residue Often the time difference increases with the increase of geophone offset, and in little geophone offset, residual NMO is very little, negligible.And, root The CMP domain calculating Value of residual static correction can be converted into common big gun by the common big gun-cautious residual static corrections according to the present invention Point domain and common receiver domain.After conversion, the assumed condition of Value of residual static correction and noise random distribution is easier to meet, Preferable effect is achieved in low signal-to-noise ratio, the calculating of big Value of residual static correction data.Therefore, the application is by asking for each shot point Value of residual static correction and the Value of residual static correction of each geophone station be obtained with total static correction value of any one road earthquake record
Therefore, according to the present invention, based on the premise that
1. assume ripple in weathering zone perpendicular to overland propagation, the i.e. residue of same shot point or same geophone station (receiving point) Static correction value is identical.
2. the Value of residual static correction of certain recorded trace includes shot point Value of residual static correction and geophone station Value of residual static correction.
1. and 2. according to, in same common-shot-gather, the shot point Value of residual static correction in each road is identical, and geophone station remains Remaining static correction value is different;And in same geophone station road collection, the shot point Value of residual static correction in each road is different, geophone station is remaining Static correction value is identical.
Fig. 1 shows the geological data residual static corrections based on ant group algorithm according to an exemplary embodiment of the present invention Flow chart.
As shown in figure 1, first by acquired original to single-shot data carry out field statics process and dynamic correction, wherein wild Outer static correction and dynamic correction are the routine operations of this area, will not be described here, because dynamic correction is typically in concentrically road collection Middle process, therefore, the geological data after dynamic correction is carried out road collection extraction process by the application first, after will moving correction The gather of concentrically road collection be common-shot-gather or common detector gather.
According to the present invention it is possible to adopt two sets of flow processs, shown in the first operating process that is, in Fig. 1, left column is listed and right row Second operating process, wherein, the first operating process is first to carry out calculating and then in the calculating of common receiver domain in common-source point domain, and the Two operating processes are first to carry out calculating in common receiver domain and then calculate in common-source point domain, and the processing means of two flow processs are Identical, difference is the sequencing of each step, therefore, in order to save length, below only to first shown in left column Operating process is described in detail.
In a step 101, calculated in the common-shot-gather taking out road collection formation.Specifically, using ant group algorithm root Obtain the optimal solution of the Value of residual static correction of common-shot-gather according to the initial disaggregation search of the Value of residual static correction of common-shot-gather Collection.
Fig. 2 show according to an exemplary embodiment of the present invention in common-shot-gather utilize ant group algorithm according to common-source point The flow chart that the initial disaggregation of the Value of residual static correction of road collection obtains the optimal solution set of Value of residual static correction of common-shot-gather.
In one embodiment, can be using the method shown in Fig. 2 in execution step 101.
As shown in Fig. 2 in step 301, the scope according to default Value of residual static correction is each of common-shot-gather Road generates an initial residual static correction value at random, forms initial disaggregation (i.e. n-th solution of common-shot-gather Value of residual static correction Collection, n=1).
In step 302, the n-th solution m ant being respectively placed in common-shot-gather Value of residual static correction is concentrated and is corresponded to On the node that road Value of residual static correction is located, m is equal to total road number of common-shot-gather, and the loop iteration number of times in setting big gun domain is T=1, starts the t time loop iteration (abbreviation iteration) in common-shot-gather, the wherein initial value of t is 1.
In step 303, quiet by applying the n-th solution set pair common-shot-gather to carry out the t time residue in common-shot-gather Correction.
In step 304, calculate the common-shot-gather after the t time residual static correction according to equation below (2) Superposition road (i.e. model trace) amplitude energy etAmplitude energy with the superposition road of the common-shot-gather before the t time residual correction Amount et-1.It should be appreciated that as t=1, the amplitude energy e in the superposition road of the common-shot-gather after the t-1 time residual correction0For Do not apply the original common-shot-gather that the initial disaggregation of common-shot-gather is carried out before residual static correction to pass through the superposition of big gun domain to obtain Superposition road amplitude energy.
e t = σ k = 1 m σ j = 1 t d j , k t - - - ( 2 )
In equation (2), m is total road number of common-shot-gather, and t is total number of samples of seismic channel,The the one t time remaining quiet The amplitude energy of k-th sampling point in jth road of the common-shot-gather after correction.
In step 305, determine the amplitude energy e in the superposition road of common-shot-gather after the t time residual static correctiontIt is The amplitude energy e in the no superposition road being more than the common-shot-gather after the t-1 time residual static correctiont-1.
When determining e in step 305t> et-1When, then it is believed that the Value of residual static correction of this common-shot-gather is to optimum Direction convergence, the Value of residual static correction that at this moment big gun collection can be unconditionally accepted as current optimal residual static correction value, that is, in step In rapid 306 using the n-th disaggregation as common-shot-gather Value of residual static correction current optimal solution set.
In step 307, judge whether to meet sufficient big gun domain end condition, big gun domain end condition could be arranged to: determine this Whether secondary iterations t reaches default big gun domain maximum iteration time and whether finds least residual, if find least residual Determine the amplitude energy e in the superposition road of common-shot-gather data after the t time residual static correctiontRemaining with the t-1 time quiet The amplitude energy e in the superposition road of the common-shot-gather data after aligning stept-1Difference whether reach minimum.
When determining e in step 307t-et-1When being not reaching to minimum and t and being not up to big gun domain maximum iteration time, that is, less than During sufficient big gun domain end condition, turn to step 308.
When determining e in step 307t-et-1Reach minimum or when t reaches big gun domain maximum iteration time, then solve n-th Collection is as the optimal solution set of the Value of residual static correction of common-shot-gather;Wherein, described step 302 is repeated several times to 307 and cannot E is compared in acquisitiont-et-1E is judged during less differencet-et-1Reach minimum.Here, the number of times of repetition can be carried out according to actual conditions Select, for example, 10~100 times.
In step 308, current one group optimum Value of residual static correction (i.e. current optimal solution set) is disturbed at random Dynamic, i.e. one random disturbance quantity of plus/minus on the basis of the n-th solution set Zhong Ge road Value of residual static correction currency, and by disturbance The disaggregation that Hou Ge road Value of residual static correction is constituted is directly as the (n+1)th disaggregation, defeated as the calculating of loop iteration next time Enter Value of residual static correction.Meanwhile, make t=t+1, n=n+1 simultaneously returns execution step 302.It should be appreciated that first time loop iteration The input Value of residual static correction calculating is the Value of residual static correction that common-shot-gather initial solution is concentrated.
When determining e in step 305t≤et-1When, judge that now Value of residual static correction is not to convergence direction approximation, though So now Value of residual static correction is not to convergence direction approximation, but due to it is considered that the convergence of this Value of residual static correction is not one Individual linear trend, but one is nonlinear, has a lot of local minimums, thus can be received (i.e. following by probability Equation (3)) mode allow it to jump in another convergent tendency locally or globally.Specifically, the n-th solution is concentrated often Value of residual static correction together carries out random perturbation and selects common detector gather according to the probability branch mode of ant group algorithm In Value of residual static correction after every one disturbance, form the (n+1)th disaggregation.
As shown in Fig. 2 in the step 310, the Value of residual static correction in each road is concentrated to carry out random perturbation the n-th solution, that is, One random disturbance quantity of plus/minus on the basis of the n-th solution set Zhong Ge road Value of residual static correction currency.
In step 311, calculate ant k using equation (3) to concentrate corresponding to the Value of residual static correction in kth road from the n-th solution Node i transfer to the n-th solution and concentrate corresponding to the Value of residual static correction that obtains after random perturbation of Value of residual static correction in kth road Node j acceptance probability pij kT (), wherein, k gets m from 1.
In equation (3), τijT () represents that all ants are retained in i-node to j node path in the t time iteration Pheromones amount, τisT () represents that all ants are retained in the pheromones amount to s node path for the i-node in the t time iteration, i saves Point represents other all optional paths, η in addition to i-node to j node path to s node pathijT () is heuristic function, be i section Point arrives the inverse of j node Euclidean distance, ηisT () is the inverse to s node Euclidean distance for the i-node, α represents transfer rail The importance of mark, β represents the relative importance of visibility.
In step 312, judgeWhether it is more than default acceptance probability pacp, with judge whether to accept ant k from I-th node transfers to jth node.
When judging in step 312When, then accept in a step 314 to concentrate kth road remaining n-th solution The Value of residual static correction that static correction value obtains after disturbance concentrates kth as the (n+1)th solution of common-shot-gather Value of residual static correction The Value of residual static correction in road, completes the process that the n-th solution concentrates kth road.
WhenWhen, then refuse to concentrate the residue that kth road Value of residual static correction obtains after disturbance quiet n-th solution Correcting value solves, as (n+1)th, the Value of residual static correction concentrating kth road, and concentrates undone process to the n-th solution in step 313 The Value of residual static correction in kth road carries out disturbance again, and return to step 311, repeat the above steps, until
In step 315, judge whether all roads that the n-th solution is concentrated are completed renewal and process, that is, judge whether own Ant all completes the selection of next node, if certain one or multi-channel do not complete, return to step 313;If sentencing in step 315 Ding Ge road has all completed renewal and has processed, then form the (n+1)th disaggregation in step 316, and fresh information element, then makes t=t+1, n =n+1 simultaneously returns execution step 302.In this embodiment, it is preferred that, adopt to complete pheromones according to equation below (4) Update.
τij(t+1)=(1- ζ) τij(t)+δτij(t) (4)
In equation (4),ζ is pheromones attenuation coefficient, ζ=ρt, and And ρ is 0~1 constant, t is iterations.
Certainly the invention is not restricted to this, pheromones attenuation coefficient can also be using conventional constant coefficient decay.But, at this In embodiment, pheromones attenuation coefficient is optimized, exponential function decay is changed to by conventional constant coefficient decay, improves The optimizing ability of this algorithm.
In the step 102 of Fig. 1, the optimum of the Value of residual static correction of the common-shot-gather being obtained using the step 309 of Fig. 2 Solution set pair common-shot-gather carries out residual static correction.
In step 103, judge whether to meet final end condition.It should be noted that the requirement of final end condition Stricter than big gun domain end condition and industry disturbance in geophone domain end condition, can by determine whether to reach default run time and Determine the amplitude energy e in the superposition road of common-shot-gather data after the t time residual static correctiontQuiet school remaining with the t-1 time The amplitude energy e in the superposition road of the common-shot-gather data after positive stept-1Difference whether reach pre-established threshold.
When e is worked as in judgement in step 103t-et-1Less than or equal to pre-established threshold or when reaching default run time, Terminate computing (end).
When judging e in step 103t-et-1More than pre-established threshold and not up to default run time when, then execute step Rapid 104.
At step 104, the common detector gather carrying out next attribute group calculates, will be after step 102 is processed Common-shot-gather is converted to common detector gather, and first according to the Value of residual static correction of common detector gather using ant group algorithm Beginning disaggregation calculates the optimal solution set of the Value of residual static correction of common detector gather.
Fig. 3 shows and according to an exemplary embodiment of the present invention utilizes ant group algorithm in common detector gather according to altogether inspection The initial disaggregation of the Value of residual static correction of wave point road collection obtains the stream of the optimal solution set of Value of residual static correction of common detector gather Cheng Tu.
In one embodiment, can be using the method shown in Fig. 3 in execution step 104.
From the figure 3, it may be seen that utilizing ant group algorithm in common detector gather according to the Value of residual static correction of common detector gather Initial disaggregation obtain the optimal solution set of Value of residual static correction of common-shot-gather and utilize ant group algorithm in common-shot-gather The initial disaggregation of the Value of residual static correction according to common-shot-gather obtains the optimal solution set of the Value of residual static correction of common-shot-gather Mode of operation the same, difference is that the two is the place that carries out in common detector gather and in common-shot-gather respectively Reason.Thus, it is briefly described below in conjunction with Fig. 3 and utilize ant group algorithm surplus according to common detector gather in common detector gather The initial disaggregation of remaining static correction value obtains the flow process of the optimal solution set of Value of residual static correction of common-shot-gather.
As shown in figure 3, in step 401, the scope according to default Value of residual static correction is every in common detector gather Generate together at random an initial residual static correction value, formed common detector gather initial disaggregation (i.e. the n-th ' disaggregation, n '= 1).
In step 402, the n-th ' solution m ' ant being respectively placed in common detector gather concentrates corresponding road residue quiet On the node that correcting value is located, m ' is equal to total road number of common detector gather, and the loop iteration number of times in setting industry disturbance in geophone domain is T '=1, starts the secondary loop iteration of t ' in common detector gather, the wherein initial value of t ' is 1.
In step 403, pass through to apply the initial disaggregation of described common detector gather in common detector gather, that is, utilize The n-th ' disaggregation (n '=1) carries out the secondary residual static correction of t ' to common detector gather.
In step 404, the common detector gather after the secondary residual static correction of t ' is superimposed acquisition one by industry disturbance in geophone domain Individual superposition road (i.e. model trace), the amplitude energy in this superposition road is et′;Correspondingly, the common inspection after -1 residual correction of t ' The amplitude energy e in the superposition road of wave point road collectiont-1′;It should be appreciated that the common receiver when t '=1, after the 0th residual correction The amplitude energy e in the superposition road of road collection0' former before residual static correction for not applying the initial disaggregation of common detector gather to carry out The amplitude energy in the superposition road that beginning common detector gather is obtained by industry disturbance in geophone domain superposition.
In step 405, determine the amplitude energy in the superposition road of common detector gather after the secondary residual static correction of t ' etWhether it is more than the amplitude energy e in the superposition road of common detector gather after -1 residual static correction of t 't-1.
When determining e in step 405t' > et-1' when, then it is believed that the Value of residual static correction of this common detector gather to Optimum direction convergence, the Value of residual static correction that can unconditionally accept detection point set at this moment is as current optimal residual static correction Amount, that is, in a step 406 using the n-th ' disaggregation as current optimal solution set.
In step 407, judge whether to meet sufficient industry disturbance in geophone domain end condition, industry disturbance in geophone domain end condition can be arranged For: determine whether the current iteration number of times t ' of common detector gather reaches default industry disturbance in geophone domain maximum iteration time and whether look for To least residual, if find the superposition that least residual determines the common detector gather data after the secondary residual static correction of t ' The amplitude energy e in roadt' and -1 residual static correction step of t ' after common detector gather data superposition road amplitude energy Amount et-1' difference whether reach minimum.
When determining e in step 407t′-et-1' it is not reaching to minimum and t ' not up to industry disturbance in geophone domain maximum iteration time When, that is, when not meeting big gun domain end condition, turn to step 308.
When determining e in step 407t′-et-1' reach minimum or when t reaches industry disturbance in geophone domain maximum iteration time, then will The n-th ' disaggregation is as the optimal solution set of the Value of residual static correction of common detector gather.Wherein, be repeated several times described step 402 to 407 and cannot obtain and compare et′-et-1E is judged during ' less differencet′-et-1' reach minimum.Here, the number of times of repetition can root Selected according to actual conditions, for example, 10~100 times.
In a step 408, (the i.e. current optimal solution of the Value of residual static correction to current one group optimum common detector gather Collection) carry out random perturbation, that is, on the basis of the n-th ' solution set Zhong Ge road Value of residual static correction currency of common detector gather One random disturbance quantity of plus/minus, and the disaggregation that disturbance Hou Ge road Value of residual static correction is constituted is directly as the n-th '+1 solution Collection, the input Value of residual static correction calculating as loop iteration next time.Meanwhile, make t '=t '+1, n '=n '+1 returning holds Row step 402.It should be appreciated that the input Value of residual static correction that first time loop iteration calculates is common detector gather initial solution The Value of residual static correction concentrated.
When determining e in step 405t′≤et-1' when, concentrate the residue in each road quiet the n-th ' solution of common detector gather Correcting value carries out random perturbation and selects every one disturbance in common detector gather according to the probability branch mode of ant group algorithm Value of residual static correction afterwards, forms the n-th '+1 disaggregation.
As shown in figure 3, in step 410, the Value of residual static correction concentrating each road is solved to the n-th ' of common detector gather Carry out random perturbation, i.e. plus/minus on the basis of the n-th solution set Zhong Ge road Value of residual static correction currency of common detector gather One random disturbance quantity.
In step 411, calculate ant k ' from the n-th ' of common detector gather Value of residual static correction using above equation (3) Solution concentrates the node i ' corresponding to Value of residual static correction in kth ' road to transfer to the n-th ' of common detector gather Value of residual static correction The acceptance of the node j ' corresponding to Value of residual static correction that the Value of residual static correction in solution concentration kth ' road obtains after random perturbation ProbabilityWherein, k ' gets m ' from 1.
In step 412, judgeWhether it is more than default acceptance probability pacp', to judge whether to accept ant k ' Transfer to jth ' node from the i-th ' node.
When judging in step 412When, then accepting in step 414 will be quiet for common detector gather residue The Value of residual static correction that the n-th ' solution concentration kth ' road Value of residual static correction of correcting value obtains after disturbance is as common receiver road The n-th '+1 solution of collection Value of residual static correction concentrates the Value of residual static correction in kth ' road, completes the process that the n-th ' solution concentrates kth ' road.
WhenWhen, then refuse for the n-th ' solution of common detector gather to concentrate kth ' road Value of residual static correction warp The Value of residual static correction obtaining after disturbance solves, as the n-th '+1 of common detector gather, the Value of residual static correction concentrating kth ' road, and The Value of residual static correction not completing the kth ' road processing is concentrated to carry out weight the n-th ' solution of common detector gather in step 413 New disturbance, and return to step 411, until
In step 415, judge whether all roads that the n-th ' solution of common detector gather is concentrated are completed renewal and process, Judge whether that all ants all complete the selection of next node, if certain one or multi-channel do not complete, return to step 413; If judging in step 415, each road has all completed renewal and has processed, and forms the n-th '+1 solution of common detector gather in step 416 Collection, and then fresh information element makes t '=t '+1, n '=n '+1 simultaneously returns execution step 402..In this embodiment, it is preferred that, Adopt the renewal to complete pheromones according to above equation (4), wherein, in this step, the t in equation (4) is industry disturbance in geophone domain Interior iterations t '.
In step 105, the optimal solution set of the Value of residual static correction of the common detector gather being obtained using step 409 is to altogether Geophone station road collection carries out residual static correction.
In step 106, judge whether to meet final end condition.When e is worked as in judgement in step 106t′-et-1' be less than Or when being equal to pre-established threshold or reaching default run time, terminate computing (end);When judging e in step 106t′- et-1' more than pre-established threshold and not up to default run time when, then return execution step 101, adjustment Value of residual static correction Preset range and regenerate common-shot-gather Value of residual static correction initial disaggregation, restart calculate.
In addition, it is necessary to explanation, above only single-shot geological data is processed, but the invention is not restricted to this, right In many big guns geological data, its processing mode is the same, can using said method, many big guns geological data be processed simultaneously.
Above-mentioned example embodiment for a better understanding of the present invention, is carried out to it with reference to specific example further Explanation.
Example:
(1) the common-shot-gather geological data of the Value of residual static correction to be calculated after dynamic correction, this common-shot-gather are inputted Geological data Zhong You 500 road.
(2) (big gun domain residual static correction calculating) initialization: set the iterations t=1 in big gun domain, (m is ant sum for m For solving the road number in the common-shot-gather of Value of residual static correction, i.e. m=500)
(3) (big gun domain residual static correction calculating) assumes the initial residual static correction value difference in each road of this common-shot-gather For:I.e. initial disaggregation is By m ant It is respectively placed in (solution of the n as solution space of solution residual static correction, example on n static correction value node by certain initial rules Then have 61 values as between -60ms~+60ms at interval of for 2ms value, be respectively as follows: -60ms, -58ms, - 56ms ..., -2ms, 0ms, 2ms ..., 56ms, 58ms, 60ms), the node that these place ants is randomly generated. Due to m (m=500) the > n (n=61) enumerating in this example, thus, the initial residual static correction value that there are some roads is located Value node is identical, for example,It is -2ms, if the node number of -2ms is i, the 1st ant, the 5th ant Ant and the 30th ant are both placed in node i.
(4) carry out ant group algorithm search, application message element more new regulation changes pheromones value.In construction solution, ant n walks The every paths crossed are adjusted and are tried to achieve by equation (1) and (2), specifically:
First, apply initial residual static correction value, that is, utilize the initial solution set pair of common-shot-gather Value of residual static correction former Beginning common-shot-gather carries out first time residual static correction, then the common-shot-gather after first time residual static correction is folded Plus obtain the amplitude energy e of superposition road (i.e. model trace)1, with the amplitude energy being superimposed before application initial residual static correction value Amount e0It is compared:
Wherein, m is the road number of every big gun, and t is the number of samples of seismic channel,It is the jth of initial disaggregation K-th sample value in road.
Wherein, m is the road number of every big gun, and t is the number of samples of seismic channel,It is that first round residual is quiet K-th sample value in this Baoj road after correcting value renewal.
If e1>e0, accept current residual static correction value as current optimal value, carry out Value of residual static correction random perturbation, Wherein, initial solution is concentrated the residual static correction value that every one Value of residual static correction obtain after random perturbation directly to form the N+1 disaggregation, the initial value of n is 1.
If e1≤e0, carry out the Value of residual static correction random perturbation by road, and according to the probability branch mode of ant group algorithm Lai Select the Value of residual static correction after every one disturbance in common detector gather, form the (n+1)th disaggregation.
For example, as n=1, the (n+1)th disaggregation is
Wherein, initial solution is concentrated every Value of residual static correction together carry out obtaining after first time random perturbation:
v 1 1 = v 1 0 + ξ 1 0 · δv v 2 1 = v 2 0 + ξ 2 0 · δv v 3 1 = v 3 0 + ξ 3 0 · δv . . . . . . v m 1 = v m 0 + ξ m 0 · δv - - - ( 5 )
In equation (5), δ v is random disturbance quantity, for example, be taken as 5ms in this example, For coefficient of disturbance, for example in this exampleIt is taken as the arbitrary value in -2~2.
Calculate every transition probability together successively, for example, initial solution concentrates first Value of residual static correctionPlace node For i,The Value of residual static correction obtaining after first time random perturbation isAndPlace node is j, then according to equation (3) calculate the first track data Value of residual static correction fromDisturbance is arrivedAcceptance probability (transition probability), that is, calculate first Ant is from i-node disturbance to the transition probability of j node
Then pass throughTo determine whether to accept the Value of residual static correction in current seismic road Value is updated to the Value of residual static correction after disturbance, for example,When, accept the first track data Value of residual static correction fromDisturbance is arrivedInitial solution will concentrate firstIt is updated toOtherwise refusal updates, and again to first Value of residual static correction carries out disturbance, makes until searching suitable disturbance quantity
After completing the disturbance of Value of residual static correction of all seismic channels of this common-shot-gather and selecting, that is, m ant is equal After finding next node, the Value of residual static correction after the disturbance of each road forms the (n+1)th disaggregation, enters row information using equation (4) Plain intensity updates, and makes t=t+1, n=n+1, carries out the amplitude energy being superimposed road before and after the Value of residual static correction application of next round Amount contrast.
(5) if m ant all adjusts and complete after t wheel, and min (e)=et-et-1, that is, after the t time residual static correction Common-shot-gather data superposition road amplitude energy etWith the common-shot-gather number after the t-1 time residual static correction step According to superposition road amplitude energy et-1Residual error reach minimum, and repeatedly revise and still cannot obtain more Small object value min (e), Or t reaches maximum cycle, then the n-th disaggregation as the optimal solution set of common-shot-gather Value of residual static correction and is applied to Data in common-shot-gather is carried out go to step (6) after residual static correction, otherwise return execution step (4).
(6) if final end condition meets, terminate;Otherwise, go to step the common receiver road that (2) carry out next attribute group Collection calculates, and will select through the geological data after the optimal solution set of common-shot-gather Value of residual static correction carries out residual static correction Arrange as common detector gather, and carry out in industry disturbance in geophone domain common detector gather Value of residual static correction the search of optimal solution set and And complete residual static correction.Wherein, final end condition can set the length lower limit of shortest path, such as min (e) < 10-5Then tie Bundle, also can limit run time, for example, run time time > 1000000 seconds then terminate calculating.
In sum, the invention provides a kind of big gun altogether-cautious residual static corrections, Value of residual static correction will be calculated CMP domain is converted into common-source point domain and common receiver domain.After conversion, Value of residual static correction and noise random distribution Assumed condition is easier to meet, and achieves preferable effect in low signal-to-noise ratio, the calculating of big Value of residual static correction data.Have The features such as calculation procedure simplicity, computational efficiency height and high treating effect.
Although above by describing the present invention with reference to exemplary embodiment, those skilled in the art should be clear Chu, in the case of the spirit and scope being limited without departing from claim, can be carried out respectively to the exemplary embodiment of the present invention Plant modifications and changes.

Claims (8)

1. a kind of geological data residual static corrections based on ant group algorithm are it is characterised in that methods described includes the first behaviour Make or the second operation, wherein,
Described first operation includes the common midpoint gather choosing by single-shot geological data after field statics and dynamic correction Arrange as common-shot-gather, and described common-shot-gather carried out successively with following process:
Step a:
A1, it is that to generate an initial residual together at random quiet for common-shot-gather every according to the scope of default Value of residual static correction Correcting value, forms the initial disaggregation of common-shot-gather Value of residual static correction;
A2, the initial disaggregation according to described common-shot-gather Value of residual static correction are simultaneously scanned for using ant group algorithm, are total to The optimal solution set of shot point road collection Value of residual static correction simultaneously is applied to carry out residual static correction together to the every of common-shot-gather;
Step b:
Judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for finally terminating bar Part, then will carry out the data sorting after residual static correction process using the optimal solution set of described common-shot-gather Value of residual static correction For common detector gather, and execution step c;
Step c:
C1, it is that the every of common detector gather generates an initial residual together at random according to the scope of default Value of residual static correction Static correction value, forms the initial disaggregation of common detector gather Value of residual static correction;
C2, the initial disaggregation according to described common detector gather Value of residual static correction are simultaneously scanned for using ant group algorithm, obtain The optimal solution set of common detector gather Value of residual static correction simultaneously is applied to carry out remaining quiet school together to the every of common detector gather Just;
Step d:
Judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for finally terminating bar Part, then return execution step a;
Described second operation includes the common midpoint gather gather by single-shot geological data after field statics and dynamic correction For common detector gather, and described common detector gather is carried out successively with following process:
Step a ': operate using with the step c identical in the described first operation;
Step b ': judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for Whole end condition, then will be carried out after residual static correction process using the optimal solution set of described common detector gather Value of residual static correction Data sorting be common-shot-gather, and execution step c ';
Step c ': operate using with the step a identical in the described first operation;
Step d ': judge whether to meet final end condition: if meeting final end condition, stop computing;If being unsatisfactory for Whole end condition, then return execution step a '.
2. the geological data residual static corrections based on ant group algorithm according to claim 1 are it is characterised in that described Step a2 includes:
A21, the n-th solution that m ant be respectively placed in common-shot-gather Value of residual static correction concentrate corresponding road residual static correction On the node that amount is located, carry out the t time loop iteration in big gun domain, wherein, m is equal to total road number of common-shot-gather, t's and n Initial value is 1, and as n=1, the n-th disaggregation is the initial disaggregation of described common-shot-gather Value of residual static correction;
A22, using n-th solution set pair common-shot-gather carry out the t time residual static correction, calculate by Equation 1 below and compare t The amplitude energy e in the superposition road of the common-shot-gather after secondary residual static correctiontWith the common big gun after the t-1 time residual static correction The amplitude energy e in the superposition road of point road collectiont-1, wherein, as t=1, e0For the common big gun before first time residual static correction step The amplitude energy in the superposition road of point road collection,
Equation 1:
e t = &sigma; k = 1 m &sigma; j = 1 t d j , k t
In equation 1, m is total road number of common-shot-gather, and t is total number of samples of seismic channel,For the t time residual static correction it The amplitude energy of k-th sampling point in jth road of common-shot-gather afterwards;
Work as et≤et-1When, then concentrate the Value of residual static correction in each road to carry out random perturbation and according to ant group algorithm the n-th solution Probability branch mode, to select the Value of residual static correction after every one disturbance in common-shot-gather, forms common-shot-gather residue quiet (n+1)th disaggregation of correcting value, wherein, the probability branch mode of described ant group algorithm is:
First, the kth ant positioned at node i during the t time loop iteration is calculated according to equation 2 below and transfer to next node The transition probability of jWherein, node i is the node corresponding to Value of residual static correction that the n-th solution concentrates kth road, and node j is N-th solution concentrates the node corresponding to Value of residual static correction that the Value of residual static correction in kth road obtains after random perturbation,
Equation 2:
In equation 2, τijT () represents that all ants are retained in the pheromones to j node path for the i-node in the t time iteration Amount, τisT () represents that all ants are retained in the pheromones amount to s node path for the i-node, i-node to s in the t time iteration Node path represents other all optional paths, η in addition to i-node to j node pathijT () is heuristic function, be i-node to j The inverse of node Euclidean distance, ηisT () is the inverse to s node Euclidean distance for the i-node, α represents transfer track Importance, β represents the relative importance of visibility;
Then, compareWith default acceptance probability pacp, judge whether to accept ant k and transfer to jth node from the i-th node:
WhenWhen, then accept the residual static correction that the n-th solution concentration kth road Value of residual static correction is obtained after disturbance Amount concentrates the Value of residual static correction in kth road as the (n+1)th solution;
WhenWhen, then refuse the residual static correction that the n-th solution concentration kth road Value of residual static correction is obtained after disturbance Amount concentrates the Value of residual static correction in kth road as the (n+1)th solution, and again concentrates the Value of residual static correction in kth road to carry out the n-th solution Disturbance, then judges whether to accept the n-th solution is concentrated the residual static correction of kth road according to the probability branch mode of described ant group algorithm Value of residual static correction after disturbance again for the amount concentrates the Value of residual static correction in kth road as the (n+1)th solution, until
Finally, after all ants all complete the selecting and form the (n+1)th disaggregation of next node, fresh information element, make t=t+ 1, n=n+1 and return execution step a21;
Work as et> et-1When, then using the n-th disaggregation as common-shot-gather Value of residual static correction current optimal solution set;
A23, determine whether to meet big gun domain end condition: if being unsatisfactory for big gun domain end condition, concentrate each road to remain the n-th solution Remaining static correction value carries out random perturbation and the disaggregation that constituted disturbance Hou Ge road Value of residual static correction is directly as common-source point (n+1)th disaggregation of road collection Value of residual static correction, makes t=t+1, and n=n+1 simultaneously returns execution step a21;If meeting big gun domain to terminate bar Part then using the n-th disaggregation as common-shot-gather Value of residual static correction optimal solution set.
3. the geological data residual static corrections based on ant group algorithm according to claim 2 are it is characterised in that described The big gun domain end condition of step a23 is:
Determine whether t reaches big gun domain maximum iteration time and determine the common-shot-gather data after the t time residual static correction The amplitude energy e in superposition roadtAmplitude energy with the superposition road of the common-shot-gather data after the t-1 time residual static correction step Amount et-1Difference whether reach minimum,
Work as et-et-1When being not reaching to minimum and t and being not up to big gun domain maximum iteration time, judge not meeting big gun domain end condition;
Work as et-et-1Reach minimum or when t reaches big gun domain maximum iteration time, judge to meet big gun domain end condition;
Wherein, described step a21 and a22 are repeated several times and cannot obtain and compare et-et-1E is judged during less differencet-et-1Reach Minimum.
4. the geological data residual static corrections based on ant group algorithm according to claim 2 are it is characterised in that described Step a22 carries out Pheromone update according to equation 3 below,
Equation 3:
τij(t+1)=(1- ζ) τij(t)+△τij(t)
In equation 3,ζ is pheromones attenuation coefficient, ζ=ρt, and ρ is 0 ~1 constant, t is iterations.
5. the geological data residual static corrections based on ant group algorithm according to any one in claim 2 or 4, its It is characterised by, described step c2 includes:
C21, the n-th ' solution that m ' ant be respectively placed in common detector gather concentrate corresponding road Value of residual static correction to be located On node, m ' is equal to total road number of common detector gather, carries out the secondary loop iteration of t ' in industry disturbance in geophone domain, wherein, t ' and n ' Initial value be 1, and when n '=1, the n-th ' disaggregation of common detector gather is the remaining quiet school of described common detector gather The initial disaggregation of positive quantity;
C22, using the n-th ' solution set pair common detector gather carry out the secondary residual static correction of t ', calculated using described equation 1 and compare Amplitude energy e compared with the superposition road of the common detector gather after the secondary residual static correction of t 't' and -1 residual static correction of t ' The amplitude energy e in the superposition road of common detector gather afterwardst-1', wherein, e0' for before first time residual static correction step The amplitude energy in the superposition road of common detector gather,
Work as et′≤et-1' when, then the Value of residual static correction in each road is concentrated to carry out random perturbation and according to described ant colony the n-th ' solution The probability branch mode of algorithm, to select the Value of residual static correction after every one disturbance in common detector gather, forms common receiver The n-th '+1 disaggregation of road collection Value of residual static correction, and fresh information element, make t '=t '+1, n '=n '+1 simultaneously returns execution step c21;
Work as et' > et-1' when, then using the n-th ' disaggregation as common detector gather Value of residual static correction current optimal solution set;
C23, determine whether to meet industry disturbance in geophone domain end condition: if being unsatisfactory for industry disturbance in geophone domain end condition, the n-th ' solution is concentrated The disaggregation that every Value of residual static correction together carries out random perturbation and constituted disturbance Hou Ge road Value of residual static correction is direct As the n-th '+1 disaggregation of common detector gather Value of residual static correction, make t '=t '+1, n '=n '+1 simultaneously returns execution step c21;If meeting industry disturbance in geophone domain end condition, using the n-th ' disaggregation as the Value of residual static correction of common detector gather optimal solution Collection.
6. the geological data residual static corrections based on ant group algorithm according to claim 5 are it is characterised in that described The industry disturbance in geophone domain end condition of step c23 is: determines whether t ' reaches industry disturbance in geophone domain maximum iteration time and determine that t ' is secondary surplus The amplitude energy e in the superposition road of the common detector gather data after remaining static correctiont' and -1 residual static correction step of t ' it The amplitude energy e in the superposition road of common detector gather data afterwardst-1' difference whether reach minimum,
Work as et′-et-1' when being not reaching to minimum and t ' and being not up to industry disturbance in geophone domain maximum iteration time, judge not meeting industry disturbance in geophone domain End condition;
Work as et′-et-1' reach minimum or when t ' reaches industry disturbance in geophone domain maximum iteration time, judge that meeting industry disturbance in geophone domain terminates bar Part;
Wherein, described step c21 and c22 are repeated several times and cannot obtain and compare et′-et-1E is judged during ' less differencet′-et-1′ Reach minimum.
7. the geological data residual static corrections based on ant group algorithm according to claim 1 are it is characterised in that described Step b and d ' inclusion: determine whether to reach default run time and determine the common-shot-gather after the t time residual static correction The amplitude energy e in the superposition road of datatWith the superposition road of the common-shot-gather data after the t-1 time residual static correction step Amplitude energy et-1Difference whether reach pre-established threshold, wherein,
Work as et-et-1More than pre-established threshold and not up to default run time when, judge to be unsatisfactory for final end condition;
Work as et-et-1Less than or equal to pre-established threshold or when reaching default run time, judge to meet final end condition.
8. the geological data residual static corrections based on ant group algorithm according to claim 1 are it is characterised in that described Step d and b ' inclusion: determine whether to reach default run time and determine the common receiver after the secondary residual static correction of t ' The amplitude energy e in the superposition road of road collection datat' and t ' -1 residual static correction step after common detector gather data The amplitude energy e in superposition roadt-1' difference whether reach pre-established threshold, wherein,
Work as et′-et-1' more than pre-established threshold and not up to default run time when, judge be unsatisfactory for final end condition;
Work as et′-et-1' less than or equal to pre-established threshold or when reaching default run time, judge to meet final end condition.
CN201410460617.1A 2014-09-11 2014-09-11 Seismic data residual static correction method based on ant colony algorithm Active CN104199106B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410460617.1A CN104199106B (en) 2014-09-11 2014-09-11 Seismic data residual static correction method based on ant colony algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410460617.1A CN104199106B (en) 2014-09-11 2014-09-11 Seismic data residual static correction method based on ant colony algorithm

Publications (2)

Publication Number Publication Date
CN104199106A CN104199106A (en) 2014-12-10
CN104199106B true CN104199106B (en) 2017-01-18

Family

ID=52084418

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410460617.1A Active CN104199106B (en) 2014-09-11 2014-09-11 Seismic data residual static correction method based on ant colony algorithm

Country Status (1)

Country Link
CN (1) CN104199106B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106547024B (en) * 2015-09-16 2018-08-07 中国石油化工股份有限公司 For the residual static correction amount estimation method and device of microseism perforation data
CN105425300B (en) * 2015-11-17 2017-12-19 中国石油天然气集团公司 A kind of residual static corrections
CN112649860B (en) * 2019-10-12 2022-10-14 中国石油化工股份有限公司 Layer velocity inversion method and system based on continuous ant colony algorithm
CN113917537A (en) * 2020-07-10 2022-01-11 中国石油化工股份有限公司 Earth surface consistency residual static correction calculation method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6393366B1 (en) * 2000-07-07 2002-05-21 Saudi Arabian Oil Company Deconvolution of seismic data based on fractionally integrated noise
CN103728665A (en) * 2012-10-12 2014-04-16 中国石油化工股份有限公司 Fine residual static correction method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6393366B1 (en) * 2000-07-07 2002-05-21 Saudi Arabian Oil Company Deconvolution of seismic data based on fractionally integrated noise
CN103728665A (en) * 2012-10-12 2014-04-16 中国石油化工股份有限公司 Fine residual static correction method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
曹中林 等.改进的郭涛算法在剩余静校正中的应用.《石油天然气学报》.2014,第36卷(第2期),第62-66页. *
潘树林 等.共炮(检)点剩余静校正方法.《石油地球物理勘探》.2011,第46卷(第1期),第83-88页. *
王书明 等.地球物理资料非线性反演方法讲座(九)——蚁群算法.《工程地球物理学报》.2009,第6卷(第2期),第131-136页. *

Also Published As

Publication number Publication date
CN104199106A (en) 2014-12-10

Similar Documents

Publication Publication Date Title
CN104199106B (en) Seismic data residual static correction method based on ant colony algorithm
CN104134351B (en) A kind of Short-time Traffic Flow Forecasting Methods
CN108805269B (en) Method for picking seismic facies arrival time based on LSTM recurrent neural network
CN105388518B (en) A kind of centroid frequency and earthquake inversion of quality factor method in the united well of Frequency spectrum ratio
CN104133245B (en) The static correcting method and system of a kind of seismic data
CN104181596B (en) Geologic horizon automatic tracking method and device
CN102313901B (en) Iteration acquisition method for first arrival wave
CN104091081B (en) Traffic data make-up method
CN106226818A (en) Seismic data processing method and device
CN109544916A (en) A kind of road network vehicle OD estimation method based on sample path data
CN102901985A (en) Depth domain layer speed correcting method suitable for undulating surface
CN115327616B (en) Automatic positioning method for mine microseism focus driven by massive data
CN106154319A (en) A kind of method for separating of imaging road collection
CN103995873B (en) A kind of data digging method and data digging system
CN103628816B (en) The method of seismic properties Guidance Levels well guiding and device thereof
CN105607119B (en) Near-surface model construction method and static correction value acquiring method
CN109558898A (en) A kind of more options learning method of the high confidence level based on deep neural network
CN105319590A (en) Anisotropic single parameter inversion method based on HTI media
Cameron Optimization and monitoring of geological carbon storage operations
CN103852789B (en) Nonlinear chromatography method and device for seismic data
CN106443777A (en) Converted wave anisotropy parameter spectrum obtaining method and converted wave velocity analysis method
CN110118994B (en) Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning
CN104181588B (en) A kind of method building high-quality model trace
CN107193044A (en) A kind of pre-stack seismic Multi-parameters conversion method of hybrid global optimization
CN106530109A (en) Oilfield development appraisal well decision method based on information value

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20180208

Address after: 072751 Zhuozhou, Baoding, Fan Yang Road West, No. 189

Patentee after: BGP INC., CHINA NATIONAL PETROLEUM Corp.

Address before: No. 216, No. 216, Huayang Avenue, Huayang Town, Shuangliu County, Shuangliu County, Sichuan

Patentee before: CNPC CHUANQING DRILLING ENGINEERING Co.,Ltd.

TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20200918

Address after: 100007 Beijing, Dongzhimen, North Street, No. 9, No.

Patentee after: CHINA NATIONAL PETROLEUM Corp.

Patentee after: BGP Inc., China National Petroleum Corp.

Address before: 072751 Zhuozhou, Baoding, Fan Yang Road West, No. 189

Patentee before: BGP Inc., China National Petroleum Corp.

TR01 Transfer of patent right