CN102867330B - Region-division-based spatial complex horizon reconstruction method - Google Patents
Region-division-based spatial complex horizon reconstruction method Download PDFInfo
- Publication number
- CN102867330B CN102867330B CN201210311567.1A CN201210311567A CN102867330B CN 102867330 B CN102867330 B CN 102867330B CN 201210311567 A CN201210311567 A CN 201210311567A CN 102867330 B CN102867330 B CN 102867330B
- Authority
- CN
- China
- Prior art keywords
- interpolation
- tomography
- section
- point
- macrolattice
- 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.)
- Expired - Fee Related
Links
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses a region-division-based spatial complex horizon reconstruction method. The horizon reconstruction in three-dimensional geological modeling under the complex topography conditions such as multi-fault overlapping can be realized without the needing of a fault polygon; simultaneously, the dynamic control of specified grid spacing is realized by a hybrid grid interpolation method, so that a geologic structure can be expressed finely; the reconstructed geologic horizon surface and the fault are closed strictly; and the processing, such as blocking processing, in later geological modeling is facilitated.
Description
Technical field
The invention belongs to Review of geologic model building techniques field, particularly a kind of spatial complex layer position reconstructing method of dividing based on region.
Background technology
Geologic modeling is by certain technological means, limited geologic data to be processed on computers, reproduces geologic structure model.Essence is to change limited discrete space sample points certificate into continuous, visual geology face or geologic body, and wherein the reconstruct of aspect is crucial.In the method for aspect reconstruct, the most important thing is the model structure of interpolation method and aspect.
Interpolation is to utilize the method for surface fitting, sets up a continuous function by discrete input sample point, obtains the functional value of optional position with the function of this reconstruction.And layer position reconstruct is exactly the method for utilizing the limited discrete reconstructed surface of three dimensions.
Conventional spatial interpolation methods has several as follows at present:
(1) method is taken advantage of in inverse distance: it is a weighted mean method of interpolation that method is taken advantage of in inverse distance.It thinks that interpolation puts several nearest known points and treat the contribution maximum of interpolation point, and its contribution is inversely proportional to distance.The control of degree parameter weight coefficient how along with leave a net point distance and decline.The given higher weight of nearer data point, the given less weight of data point far away.As follows:
In formula, z is estimated value, z
ii known point value, d
ibe the distance of i known point and interpolation point, p is weighting coefficient, and its value is higher, shows that the influence factor of distance is larger, and interpolation result just more has smooth effect.Have computing velocity faster apart from inverse ratio weighting algorithm, but be easily subject to the impact of data point cluster, the normal situation of a kind of isolated point higher than its ambient data point that occur in practical application, therefore needs to limit according to the different situations inverse ratio weighted method of adjusting the distance.
(2) natural neighbor interpolation method: natural neighbor interpolation method is the gridding new method that Surfer7.0 just has.Its ultimate principle is for more shape changeable of one group of Tyson, when add a new data point in data centralization, will revise these Thiessen polygon, and uses the weighted mean of adjoint point will determine the weight of interpolation.It is homogeneous and unconverted, and evenly spaced data are carried out to interpolation of great use, but for the defectiveness of interpolation of inhomogeneous data, and inhomogeneous original layers bit data is very common.
(3) Ke Lijin (Kriging) algorithm: claim again space autocovariance optimal interpolation method, it is a kind of optimal interpolation method with the naming of South Africa Mineral Engineering teacher D.G.Krige is a kind of geological statistics gridding method of great use.The method is based upon on variation function theory and structure analysis basis, calculates weights by introducing taking distance as the variation function of independent variable, can reflect the space structure characteristic of variable, random distribution nature that again can response variable.Say from the statistical significance, it is from correlation of variables and variability, in limited area, the value of regionalized variable is carried out without a kind of method inclined to one side, optimal estimation; From interpolation angle, it is a kind of method that the data of space distribution are asked linear optimal, estimated without inclined to one side interpolation.Because Kriging method has optimum efficiency in the interpolation of space layer position, therefore this programme taking Kriging method as basis, solves the layer position interpolation problem of complex-terrain.
Above interpolation method is all confined to solve the aspect matching of two dimensional surface, can not solve three-dimensional aspect fitting problems.In actual geologic structure, geological condition complex, the aspect matching of common two dimensional surface can not solve the fitting problems of this type of geological stratum.So three dimensions problem need to be converted into two-dimensional problems, the above interpolation method of recycling carries out aspect matching.
The structure of FEM layer model is exactly the organizational form of aspect data, and different organizational forms is set up and had different features for model.In the solution of existing proposition, main deck structure has level line, regular grid (GRID), TIN (TIN).Wherein regular grid and TIN apply FEM layer model structure more widely:
(1) regular grid (GRID)
Regular grid is to carry out presentation layer bit model with regularly arranged rectangular node.Data structure is simple, is easy to network forming, and memory data output is little, can also compress storage, and various analysis and calculations are very convenient effectively.But because mesh spacing is certain, exist the shortcoming that is difficult to determine suitable size of mesh opening expression true to nature for complicated topography and geomorphology.
(2) irregular grid (TIN)
The space aspects sampled point of collection is connected into and covers whole region and mutual nonoverlapping many triangles by certain rule, just formed TIN.TIN can express structure and the thin portion of landform relatively accurately, takes preferably aspect unique point, characteristic curve into account, represents that complex-terrain is more accurate than rectangular node.But data storage and the complicated operation of TIN.The triangle form point of TIN and the topological relation more complicated on limit, be difficult to TIN to carry out terrain analysis, and storage space is larger, is generally only applicable to large-scale high precision landform modeling among a small circle.
Prior art related to the present invention comprises:
On the reconstruction of three-dimensional geological aspect, there are many implementations at present both at home and abroad.Whether existing scheme is according to needing tomography polygon to be divided into two kinds: a kind of is the aspect reconfiguration scheme retraining based on tomography polygon, and another kind is the aspect reconfiguration scheme of dividing based on region.
(1) the aspect reconfiguration scheme based on the constraint of tomography polygon
Tomography polygon is exactly the intersection of three-dimension disclocation face and layer plane.The comprehensive tomography polygon of analyzing on multiple interfaces of aspect reconstruct need of work, according to tomography, the form of three-dimensional space interbed position is inferred in polygonal constraint.Many interpre(ta)tive system algorithms are also to complete on the basis of each cross-sectional image, but not directly reconstruct in three-dimensional raw data.In the combination of tomography polygon, generally adopt manual trace mode at present, adopt manual breakpoint method of attachment, this mode not only error is larger, and impact is configured to the precision of figure, and efficiency is extremely low.Wang Zhaohu, Liu Cai etc. propose a kind of tomography polygon Automatic Combined technology of explaining based on layer position, but this technology only can layer position very continuously, data meticulousr in the situation that automatic tracing go out tomography polygon, higher for data demand, inapplicable common situation.
(2) the aspect reconfiguration scheme of dividing based on region
The existing scheme of three-dimensional geological aspect reconstruct of dividing based on region is less, the grid cutting algorithm of the at present domestic a kind of overlapping subregion that only has Cai Qiang and Yang Qin proposition
[6].This algorithm is introduced the concept of bridge limit and crossing loop wire, realizes the aspect reconstruct under overlapping region by interlock subdivision algorithm.This scheme Shortcomings part: one, this scheme is by asking intersection to divide overlapping subregion in three dimensions, in three dimensions intersection solve more complicated and also speed is slower.Two, this scheme does not propose to solve multiple trap-up and layer position pinching, layer position and goes out the method for the three-dimensional geological aspect reconstruct in particular cases such as the end, ground, earth's surface.
Existing scheme, according to the institutional framework of reconstruct aspect, is divided into two classes: the aspect reconstruct based on specification grid (GRID) and the aspect reconstruct based on triangular mesh (TIN).Regular grid is to carry out presentation layer bit model with regularly arranged rectangular node.Because mesh spacing is certain, for complicated topography and geomorphology, be difficult to determine suitable size of mesh opening expression true to nature.The triangle form point of triangular mesh and the topological relation more complicated on limit, be difficult to TIN to carry out terrain analysis, and storage space is larger, is generally only applicable to large-scale high precision landform modeling among a small circle.
The outer scheme for the three-dimensional aspect reconstruct of geology and interpolation of Present Domestic has a lot, but all has following limitation:
(1) the three-dimensional aspect reconfiguration scheme of present most geology all needs to obtain in advance the changeable row of tomography.Huge by human-edited's tomography polygon workload, and accuracy is difficult to guarantee, inefficiency.Also there is no good solution and automatically generate tomography polygon.
(2) the three-dimensional aspect reconfiguration scheme of most of geology is not considered the situation of tomography, although some schemes have been considered tomography, does not consider for the situation of multiple trap-up.
(3) mesh spacing of the three-dimensional aspect reconfiguration scheme of the geology of existing rule-based grid processing is determined, can not dynamically be suitable for the needs of geologic structure, is difficult to meticulous expression geologic structure.
Summary of the invention
In order to overcome the above-mentioned shortcoming of prior art, the invention provides a kind of spatial complex layer position reconstructing method of dividing based on region, by dividing overlapping subregion, the aspect reconstruction in the three-dimensional geological modeling that does not need can to solve in the polygonal situation of tomography under the MODEL OVER COMPLEX TOPOGRAPHY such as multiple tomography is overlapping; By the interpolation method of hybrid grid, realized the dynamic control to specification mesh spacing simultaneously, thereby can meticulously give expression to geologic structure, the geological stratification plane reconstructing and tomography are strict sealings, are convenient to the processing in later stage Geologic modeling, as become piece processing.
The technical solution adopted for the present invention to solve the technical problems is: a kind of spatial complex layer position reconstructing method of dividing based on region, comprises the steps:
Step 1, hybrid grid interpolation method: at the little grid of macrolattice nested inside, when interpolation, first macrolattice is carried out to interpolation, and then utilize the new interpolation point on existing macroreticular Seed Points and macrolattice to carry out interpolation to little grid;
Step 2, definition layer bit field data structure;
Step 3, fault surface matching: get the interpolation envelope scope of the two-dimentional envelope of tomography kind subsegment decryption on grid plan as cross section correct, then adopt the interpolation method of hybrid grid to carry out interpolation to regmatic network lattice point within the scope of tomography envelope, described decryption specifically, the data structure of definition phase decryption:
Seg＝{ID,Direct,UpOrDown,Start,End,Points,LeftInfo,RightInfo},
Wherein, the call number of the layer position of the ID section of being ownership,
The Direct section of being direction: X-direction section be X_Direct, Y direction be Y_Direct,
The relation of the UpOrDown section of being and tomography: being positioned at hanging wall is Up, and being positioned at footwall is Down,
The Start section of being starting position,
The End section of being end position,
The related information Info of the left paragraph header of the LeftInfo section of being,
The related information Info of right section of tail of the RightInfo section of being,
The paragraph header of described section and section tail related information Info structure are: Info={ID, and UpOrDown},
The discrete point set of the Points section of being;
Step 4, layer position raw data editor: the data of deletion error; Definition layer bit boundary and tomography are restrained boundary, if in a two dimensional cross-section, have multiple layers of position section between two restrained boundaries, need these layer of position section to be combined into a layer position section;
Step 5, initialization original segment data: the incidence relation of initialization segments and restrained boundary;
Step 6, subregion are divided;
Step 7, the matching of tomography subregion layer plane:
(1) initialization upper lower burrs Seed Points;
(2) tomography upper lower burrs subregion face interpolation;
(3) intercepting of the upper and lower actual card of tomography;
The layer plane matching of step 8, non-tomography subregion:
(1) initialization seed point: both using the data point of non-fault region as Seed Points, again using the borderline region data point of tomography subregion as Seed Points; And for the tomography subregion after interpolation, using all on two dimensional surface not heavily the data points of value also as Seed Points;
(2) interpolation of non-tomography subregion: for the Seed Points having generated, adopt each net point of the subregion scope of hybrid grid interpolation method to non-tomography to carry out interpolation;
The earth's surface processing at the end, ground is gone out in step 9, layer position pinching, layer position: remove the illegal pinching part in current layer position and go out the earth's surface part at the end, ground.
Compared with prior art, good effect of the present invention is: solved the layer position surface reconstruction problem under complex-terrain in three-dimensional geological modeling, provide new thinking for isoplethes drawing, geology become piece etc.The inventive method tool has the following advantages:
(1) only need original layer position sampled data and tomography sampled data, do not need the tomography polygon can be to reconstructing aspect.Without the tomography polygon of editing manually, save numerous and diverse human-edited.
(2) support all kinds of tomographies, comprise trap-down, trap-up, can process the situation of multiple tomography simultaneously, be applicable to the topographic structure of various complexity, can process the layer position interpolation under various complex situations, there is good adaptability.
(3) be applicable to the multiple application in three-dimensional geological modeling and two-dimensional layer iso-surface patch, there is good versatility.
(4) the layer position interpolation method that adopts hybrid network to format, had both ensured the quick demonstration of data, had ensured again the precision of matching.
(5) processing that ground, earth's surface end complex situations are gone out in supporting layer position pinching, layer position.
(6) propose to utilize section envelope to carry out the method for subregion division, and provided concrete realization.
(7), by boundary constraint, realized the seamless matching of aspect and tomography.Geological stratum and fault surface after reconstruct strictly seal.
Brief description of the drawings
Examples of the present invention will be described by way of reference to the accompanying drawings, wherein:
Fig. 1 is the process flow diagram of the inventive method;
Fig. 2 is the schematic diagram of deletion error data;
Fig. 3 is the schematic diagram of initialization original segment data.
Embodiment
First some basic geologic structures and scheme term are defined:
Layer position: refer to a certain ad-hoc location in succession of strata.
Tomography: after the stressed effect fracture of rock mass, the rift structure of remarkable displacement occurs along rupture surface both sides sillar.
Segment data: layer position or tomography on a section by one group of continuous data acquisition that discrete point forms.
Hanging wall: be positioned at fault surface top.
Footwall: be positioned at fault surface below.
Stratigraphic pitch-out: refer to that sedimentary deposit is towards sedimentary basin edge, its thickness gradually attenuation until deposition not.
Gridding: the region that discrete points data is carried out is in logic divided, with the logical nets of formation rule, is convenient to layer position interpolation.
Interpolation: utilize known point to calculate the process of unknown point.
Matching: the data after utilization layer position interpolation completes form a process of aspect.
A spatial complex layer position reconstructing method of dividing based on region, as shown in Figure 1, comprises the steps:
Step 1, hybrid grid interpolation method
In existing conventional grid interpolation method, if the too sparse meeting of grid causes precision inadequate, the too close interpolation speed that can affect of grid.This programme proposes a kind of hybrid grid interpolation method of secondary Grid Nesting, and one-level grid has ensured the speed of interpolation, and secondary grid has ensured the precision of interpolation.
Hybrid grid is secondary grid, i.e. the little grid of macrolattice nested inside.When interpolation, first the point on macrolattice is carried out to interpolation, and then utilize the new interpolation point on existing macroreticular Seed Points and macrolattice to carry out interpolation to little grid.Due to the ratio of big or small grid, search the scope of Seed Points and the quantity of Seed Points can be inputted manually, make interpolation carry out different parameter settings for different situations.So both solve seed that direct Kriging supports very few and cause the problem that interpolation precision is inadequate, improved again interpolation speed.
(1) macrolattice interpolation
For the macrolattice point of interpolation, put legal Seed Points around by search value interposition this interpolation point is carried out to interpolation calculation.In the sparse layer position of some Seed Points, only relying on Seed Points to carry out interpolation cannot all calculate complete by all interpolation points.In this case, reservation cannot be calculated the interpolation point of Z value, and after all interpolation points that can rely on Seed Points to carry out interpolation calculation are disposed, the point that interpolation is completed is used as new Seed Points and is carried out layer position compensation interpolation, until all interpolation points are all disposed.The concrete grammar of macrolattice interpolation is as follows:
1) scan one by one macrolattice point, determine whether macrolattice point needs to carry out interpolation, if this macrolattice point has existed data point, do not need to carry out interpolation, otherwise need interpolation.
2), to needing the macrolattice point of interpolation, according to artificial inputted search scope, the macrolattice number of searching for again, searches for current interpolation point macroreticular Seed Points data around.
3) according to the sector of artificial input, Seed Points is carried out to sector classification.
4), to the Seed Points finding in each sector, arrange from small to large apart from current macrolattice point distance according to it.
5) according to each sector Seed Points quantity selected threshold of input, and select the Seed Points data in the nearest threshold range of the interior distance interpolation net point of this Sector Range to join in the Seed Points queue of current macrolattice interpolation.
6) if current macrolattice point does not find effective Seed Points data to carry out interpolation, not interpolation of current macrolattice, the next macrolattice point of diversion treatments, otherwise turn to the 7th) step.
7) with the Seed Points in the queue of current macrolattice interpolation Seed Points, current macrolattice point is carried out to Kriging interpolation.
8) if all macrolattice points processed once, check and whether have macrolattice point interpolation failure, if there is the failure of macrolattice point interpolation, using the successful macrolattice point of interpolation also as Seed Points, return to the 1st) step is carried out interpolation again, until the success of all net point interpolation.
(2) little net point interpolation
After macrolattice interpolation, on the grid of layer position, all macrolattice points have been plugged corresponding data.Recycle existing macrolattice point data and carry out the subdivision interpolation of little grid.In order to improve interpolation speed, the interpolation of little net point is processed taking macrolattice as unit.For certain macrolattice, search for Seed Points data in the Seed Points in current macrolattice and eight macrolattices around as primordial seed point, then utilize Kriging interpolation that these primordial seed points are fitted to a curved surface.
Can directly carry out interpolation by the value of getting on this curved surface for all little net points in macrolattice, avoid repeatedly Kriging interpolation.By a piece of data, a Kriging interpolation fitting, realizes the interpolation to little net point in whole macrolattice, greatly improves the efficiency of interpolation.For certain interpolation point I
xif the legal seed point set finding is S'={S
1, S
2..., S
k, and meet k>=2, can utilize Ke Lijin to carry out interpolation.The point set coordinate that S' is corresponding is SP={p
1, p
2..., p
k, wherein p
i(1≤i≤k) is three-dimensional coordinate point (x
i, y
i, z
i), I
xcoordinate be (x
0, y
0, z), z the unknown.In gram, gold layer position interpolation is exactly to utilize the xy coordinate (x of known point set SP and interpolation point
0, y
0) calculate the process of interpolation point z value.
Step 2, definition layer bit field data structure
Layer bit field data is the set of a string continuous discrete point.Because section is the set of discrete point, so the attribute of section is all unified for all Seed Points on this section, and the logic of relations of tomography is strong.
The data structure of definition phase decryption in this programme:
Seg＝{ID,Direct,UpOrDown,Start,End,Points,LeftInfo,RightInfo}
ID: the call number of the layer position of section ownership.
Direct: section direction: X-direction section be X_Direct, Y direction be Y_Direct.
UpOrDown: the relation of section and tomography: being positioned at hanging wall is Up, and being positioned at footwall is Down.
Start: section starting position.
End: section end position.
The paragraph header of definition phase and section tail related information structure are: Info={ID, UpOrDown}
LeftInfo: the related information Info of the left paragraph header of section.
RightInfo: the related information Info of right section of tail of section.
Points: the discrete point set of section.
The set of definition layer position is: S={S
1, S
2... S
i, S
ibe i layer position set.Layer position S
idata structure be S
i=ID, and XSegments, YSegments}, ID is layer position S
ilayer item, XSegments is for belonging to a layer S
iall segment data set of X-direction, all segment data set that YSegments is Y direction.
Step 3, fault surface matching
Tomography is earth's crust rock stratum because the stressed some strength that reaches breaks, and has along the plane of fracture structure obviously relatively moving.The face that fault rupture staggers is called fault surface.The sillar of fault surface both sides is called fault wall, is positioned at the dish that is called on fault surface, under fault surface, is called lower wall.Unique nature according to tomography is divided into: the trap-up of the trap-down rising relative to upper dish that upper dish declines relatively.
Cross section correct need to first be determined the interpolation scope of cross section correct.Because this programme is based on grid, so get the interpolation envelope scope of the two-dimentional envelope of tomography kind subsegment decryption on grid plan as cross section correct.Then adopt the interpolation method of hybrid grid to carry out interpolation to regmatic network lattice point within the scope of tomography envelope.
Step 4, layer position raw data editor
For some layer of bit data, possible its raw data has part to violate actual tectonic structure, is mainly following situation: as shown in Figure 2, at upper dish or the lower wall of tomography, there is and can only have a layer plane, and in explanation of seismic data, may occur misdata.Therefore need automatic cutting and manual editing's layer position raw data, the data of deletion error are to carry out a layer bit space interpolation, thereby simulate correct aspect.
Definition layer bit boundary and tomography are restrained boundary, and all layers position section between two restrained boundaries has same alike result.If in a two dimensional cross-section, between two restrained boundaries, there are multiple layers of position section, need these layer of position section to be combined into a layer position section, concrete combined method is as follows:
(1) the layer position section on current section is pressed to a section starting position sort ascending.
(2) add up the tomography on current section.
(3) each tomography is processed.Upper dish processing: find the layer position section Seg of the current hanging wall of next-door neighbour, then taking this layer of position section as initial segment, searching the next one can merge with the layer position section of its merging.Seg upgrades to point to and merges back segment, and recurrence merges next section, until certain restrained boundary; Lower wall processing: the layer position section Seg that finds the current footwall of next-door neighbour, then taking this layer of position section as initial segment, searching the next one can merge with the layer position section of its merging, and Seg upgrades to point to and merges back segment, recurrence merges next section, until certain restrained boundary.
(4), if there is no tomography, only all layer position sections need be merged into one section by initial rest position.
Step 5, initialization original segment data
The restrained boundary that is tegillum bit boundary and tomography definition due to layer position section retrains, so paragraph header and section tail are inevitable and some restrained boundaries are associated.Initialization original segment data are the incidence relation of initialization segments and restrained boundary, as shown in Figure 3.
Step 6, subregion are divided
The object that subregion is divided is to make whole layer position be divided into several little subregions, and these subregions are not all heavily worth, thereby makes the heavy tomography of more complicated and the interpolation of trap-up become not the heavily interpolation of value.It is to determine according to the envelope scope of section that subregion is divided, each tomography can corresponding two sub regions, go up plate region and lower wall subregion.
The envelope scope of definition section Fi is Pi, and Pi is the two-dimensional projection envelope scope of section Fi on grid.Definition upper plate region corresponding to tomography Fi is Ui, and lower wall subregion is Di.Ui and Di are the set of the envelope scope of a series of tomographies of being associated of current tomography envelope scope and current tomography Fi.
The concrete grammar that subregion is divided is as follows:
(1) if tomography Fi envelope scope Pi does not have and other tomography envelope scope overlaids, the upper lower burrs subregion of tomography Fi is the envelope scope Si of tomography Fi, i.e. Di={Pi}, Ui={Pi}.
(2) if having multiple tomographies the envelope scope of Fs...Ft} the envelope scope overlaid of Ps...Pt} and tomography Fi and and the upper (lower) dish direct neighbor of tomography Fi, the set of the envelope scope that the upper (lower) plate region of tomography Fi is all relevant tomographies, be Ui (Di)={ Pi, Ps...Pt}.
(3), after the upper lower burrs subregion of all tomographies has been divided, region, remaining non-tomography layer position is as an independent non-tomography subregion LH.
The whole region of layer position H can be expressed as: H={U1, D1...Ui, Di, LH}.Ui is respectively and layer upper plate region and a lower wall subregion of the tomography Fi that position H is associated with Di, and LH is non-fault region.
Step 7, the matching of tomography subregion layer plane
(1) initialization upper lower burrs Seed Points
To, to a certain subregion interpolation, first to determine the Seed Points data of this sub regions interpolation.These Seed Points data are not heavily worth on equatorial projection.
If the data segment structure of i layer position is S
i=ID, and XSegments, YSegments}, ID is the layer item of layer position, all segment data set that XSegments is X-direction, all segment data set that YSegments is Y direction.Definition UpSegs is hanging wall kind subsegment, and DownSegs is footwall kind subsegment.The set form of UpSegs or DownSegs is
UpSegs＝{seg1,seg2...segi}
DownSegs＝{seg1,seg2...segj}
Its stage casing segi or segj are layer position S
isection set in a certain section, and have following attribute: segi or segj paragraph header association be current hanging wall, the section tail association of segi or segj be current footwall.If the set of hanging wall subregion Seed Points is UpPoints, the set of lower wall subregion Seed Points is DownPoints.The expression formula form of UpPoints and DownPoints is:
UpPoints＝{p1,p2...pi}
DownPoints＝{p1,p2...pj}
In above formula, pi, pj are respectively the points in Ui, Di section, and the two two-dimensional projection on mesh coordinate is respectively in the upper and lower plate region of tomography (UpPoints, DownPoints).
(2) tomography upper lower burrs subregion face interpolation
With the interpolation method of hybrid grid, the Seed Points taking UpPoints and DownPoints as interpolation respectively, each net point of the upper lower burrs subregion scope to tomography carries out interpolation.
(3) intercepting of the upper and lower actual card of tomography
The layer bitplane data of the tomography upper lower burrs subregion obtaining after definition interpolation is initial upper and lower card, and initial upper and lower card data are within the scope of whole tomography upper lower burrs subregion, larger than actual upper lower burrs layer bitplane data, passed through fault surface.Need to intercept initial upper and lower card data according to the constraint of tomography.For upper and lower card and tomography are strictly sealed, we are in cross section, in fact descending card to seal processing.
If line does not pass through tomography between defining, claim to be communicated with at these 2 at 2.We taking one must be in the data of card up and down of reality point as initial source point, go out actual card up and down by connective recursive search.In fact the recursive search method of card is as follows:
1) travel through initial upper disk surface grid, find one must be on fault surface grid some p, putting p must be on actual upper disk surface, therefore taking a p as initial source point.
2) set up a data base STACK, be used for depositing source point.First source point p is pressed in data base STACK.
3) judge whether data station STACK is empty, and empty if, recursive search finishes.Otherwise from data base STACK, take out 1 q.Up and down four the some pis (i=1,2,3,4) of traversal point q on mesh coordinate.
4) if pi can be communicated with q, pi is joined in fact in card, pi is pressed in data base STACK simultaneously.If pi can not be communicated with q, illustrate that q is the frontier point of card in fact, changes to q point z value the z value of the fault surface of corresponding net point.
Also be the same for the recursive search of card under reality and the processing of upper disk surface.
The layer plane matching of step 8, non-tomography subregion
(1) initialization seed point
Carry out layer position interpolation fitting iff the data point of selecting to be positioned at non-fault region as Seed Points, can make between the layer plane of non-tomography subregion and layer plane of tomography subregion transition unsmooth.The present invention, both using the data point of non-fault region as Seed Points, again using the borderline region data point of tomography subregion as Seed Points, thereby has solved the rough problem of transition between subregion.For the tomography subregion after interpolation, using all on two dimensional surface not heavily the data points of value also as Seed Points.
(2) interpolation of non-tomography subregion
For the Seed Points having generated, with the interpolation method of hybrid grid, each net point of the subregion scope to non-tomography carries out interpolation.After interpolation completes, whole layer position surface fitting completes.
The earth's surface processing at the end, ground is gone out in step 9, layer position pinching, layer position
Layer position pinching is the same with the situation that the end, ground, earth's surface is gone out in layer position in essence, is all to go current aspect to retrain with pinching layer position and bottom surface, ground, earth's surface.So only need to and go out the part at the end, ground, earth's surface and remove just passable pinching part illegal current layer position.
Defining current interpolated layer position is main stor(e)y position S, and to have the layer position of pinching be H in current layer position
i, ground surface is UH, is DH at the bottom of ground.Definition H
i, UH, DH be restraint layer position, between restraint layer position and main stor(e)y position, have relativeness be which layer position upper, which under.The disposal route that layer position pinching, layer are gone out the end, ground, earth's surface is as follows:
(1), if current restraint layer position is earth's surface, all points of going out earth's surface (being the Z value that on net point, Z value is less than corresponding net point on ground surface) are intercepted.
(2), if current restraint layer position is the end, ground, the point of all going out at the end, ground (being the Z value that on net point, Z value is greater than corresponding net point on bottom surface, ground) is intercepted.
(3) if current restraint layer position is pinching layer position.If on the S of main stor(e)y position, being less than Z value on net point the point of the Z value of the corresponding net point in restraint layer position, this pinching layer position intercepts.If under the S of main stor(e)y position, being greater than Z value on net point the point of the Z value of the corresponding net point in restraint layer position, this pinching layer position intercepts.
Claims (3)
1. a spatial complex layer position reconstructing method of dividing based on region, is characterized in that: comprise the steps:
Step 1, hybrid grid interpolation method: at the little grid of macrolattice nested inside, when interpolation, first macrolattice is carried out to interpolation, and then utilize the new interpolation point on existing macroreticular Seed Points and macrolattice to carry out interpolation to little grid,
The described method that macrolattice is carried out to interpolation is:
1) scan one by one macrolattice point, determine whether macrolattice point needs to carry out interpolation, if macrolattice point has existed data point, do not need to carry out interpolation, otherwise need interpolation;
2), to needing the macrolattice point of interpolation, according to artificial inputted search scope, search for current interpolation point macroreticular Seed Points data around;
3) according to the sector of artificial input, Seed Points is carried out to sector classification;
4), to the Seed Points finding in each sector, arrange from small to large apart from current macrolattice point distance according to it;
5) according to each sector Seed Points quantity selected threshold of input, and select the Seed Points data in the nearest threshold range of the interior distance interpolation net point of this Sector Range to join in the Seed Points queue of current macrolattice interpolation;
6) if current macrolattice point does not find effective Seed Points data to carry out interpolation, not interpolation of current macrolattice, the next macrolattice point of diversion treatments, otherwise turn to the 7th) step;
7) with the Seed Points in the queue of current macrolattice interpolation Seed Points, current macrolattice point is carried out to Kriging interpolation;
8) if all macrolattice points processed once, check and whether have macrolattice point interpolation failure, if there is the failure of macrolattice point interpolation, using the successful macrolattice point of interpolation also as Seed Points, return to the 1st) step is carried out interpolation again, until the success of all net point interpolation;
Step 2, definition layer bit field data structure;
Step 3, fault surface matching: get the interpolation envelope scope of the two-dimentional envelope of tomography kind subsegment decryption on grid plan as cross section correct, then adopt the interpolation method of hybrid grid to carry out interpolation to regmatic network lattice point within the scope of tomography envelope, described decryption specifically, the data structure of definition phase decryption:
Seg＝{ID,Direct,UpOrDown,Start,End,Points,LeftInfo,RightInfo},
Wherein, the call number of the layer position of the ID section of being ownership,
The Direct section of being direction: X-direction section be X_Direct, Y direction be Y_Direct,
The relation of the UpOrDown section of being and tomography: being positioned at hanging wall is Up, and being positioned at footwall is Down,
The Start section of being starting position,
The End section of being end position,
The related information Info of the left paragraph header of the LeftInfo section of being,
The related information Info of right section of tail of the RightInfo section of being,
The paragraph header of described section and section tail related information Info structure are: Info={ID, and UpOrDown},
The discrete point set of the Points section of being;
Step 4, layer position raw data editor: the data of deletion error; Definition layer bit boundary and tomography are restrained boundary, if in a two dimensional cross-section, have multiple layers of position section between two restrained boundaries, need these layer of position section to be combined into a layer position section;
Step 5, initialization original segment data: the incidence relation of initialization segments and restrained boundary;
Step 6, subregion are divided;
Step 7, the matching of tomography subregion layer plane:
(1) initialization upper lower burrs Seed Points;
(2) tomography upper lower burrs subregion face interpolation;
(3) intercepting of the upper and lower actual card of tomography;
The layer plane matching of step 8, non-tomography subregion:
(1) initialization seed point: both using the data point of non-fault region as Seed Points, again using the borderline region data point of tomography subregion as Seed Points; And for the tomography subregion after interpolation, on two dimensional surface, heavily the data points of value are not as Seed Points yet using all, and specifically, between tomography, envelope scope is overlapped for described heavy value;
(2) interpolation of non-tomography subregion: for the Seed Points having generated, adopt each net point of the subregion scope of hybrid grid interpolation method to non-tomography to carry out interpolation;
The earth's surface processing at the end, ground is gone out in step 9, layer position pinching, layer position: remove the illegal pinching part in current layer position and go out the earth's surface part at the end, ground.
2. the spatial complex layer position reconstructing method of dividing based on region according to claim 1, is characterized in that: the described method that multiple layers of position section are combined into a layer position section is:
(1) the layer position section on current section is pressed to a section starting position sort ascending;
(2) add up the tomography on current section;
(3) each tomography is processed, described processes specifically each tomography, upper dish processing: the layer position section Seg that finds the current hanging wall of next-door neighbour, then taking this layer of position section as initial segment, searching the next one can merge with the layer position section of its merging, Seg upgrades to point to and merges back segment, recurrence merges next section, until certain restrained boundary, lower wall processing: the layer position section Seg that finds the current footwall of next-door neighbour, then taking this layer of position section as initial segment, searching the next one can merge with the layer position section of its merging, Seg upgrades to point to and merges back segment, recurrence merges next section, until certain restrained boundary,
(4), if there is no tomography, only all layer position sections need be merged into one section by initial rest position.
3. the spatial complex layer position reconstructing method of dividing based on region according to claim 1, is characterized in that: the method that described subregion is divided is:
(1) if certain tomography envelope scope does not have and other tomography envelope scope overlaids, the upper lower burrs subregion of this tomography is the envelope scope of tomography;
(2) if having the envelope scope of multiple tomographies and the envelope scope overlaid of a certain tomography and and upper dish or the lower wall direct neighbor of this tomography, the set of the envelope scope that the upper dish of this tomography or lower wall subregion are all relevant tomographies;
(3), after the upper lower burrs subregion of all tomographies has been divided, region, remaining non-tomography layer position is as an independent non-tomography subregion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210311567.1A CN102867330B (en) | 2012-08-29 | 2012-08-29 | Region-division-based spatial complex horizon reconstruction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210311567.1A CN102867330B (en) | 2012-08-29 | 2012-08-29 | Region-division-based spatial complex horizon reconstruction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102867330A CN102867330A (en) | 2013-01-09 |
| CN102867330B true CN102867330B (en) | 2014-10-01 |
Family
ID=47446186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210311567.1A Expired - Fee Related CN102867330B (en) | 2012-08-29 | 2012-08-29 | Region-division-based spatial complex horizon reconstruction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102867330B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| MX2016000640A (en) * | 2013-08-16 | 2016-09-22 | Landmark Graphics Corp | Identifying and extracting stratigraphic layers in one or more bodies representing a geological structure. |
| CN103489222B (en) * | 2013-09-06 | 2016-06-22 | 电子科技大学 | Target body surface reconstruction method in 3-D view |
| CN104656131A (en) * | 2013-11-22 | 2015-05-27 | 中国石油天然气集团公司 | Method for improving seismic horizon automatic tracking precision based on waveform recovery technique |
| CN104182571B (en) * | 2014-08-12 | 2018-05-04 | 电子科技大学 | Kriging interpolation methods based on Delaunay and GPU |
| CN104166163B (en) * | 2014-08-27 | 2016-08-24 | 电子科技大学 | Tomography curved surface extraction method based on three-dimensional big data quantity seismic data cube |
| CN104240301B (en) * | 2014-09-11 | 2017-03-15 | 中国石油集团川庆钻探工程有限公司地球物理勘探公司 | Geological surface reconstructing method and equipment |
| CN104199099B (en) * | 2014-09-19 | 2016-08-24 | 东北石油大学 | A kind of method utilizing multiwindow serial section efficiently to carry out seismic interpretation |
| CN104613945B (en) * | 2015-02-11 | 2017-01-18 | 国家***第二海洋研究所 | Reconstruction method for terrain of shallow-sea large-sized complicated sand wave area |
| CN113379907A (en) * | 2020-03-10 | 2021-09-10 | 中国石油天然气股份有限公司 | Method and device for constructing fault block geological model |
| CN112634450A (en) * | 2021-01-06 | 2021-04-09 | 安徽理工大学 | Method for integrally constructing local complex fault through corner grid |
| CN113674296A (en) * | 2021-09-03 | 2021-11-19 | 广东三维家信息科技有限公司 | Region cutting method and device, electronic equipment and storage medium |
| CN115438604B (en) * | 2022-11-08 | 2023-03-24 | 中国空气动力研究与发展中心计算空气动力研究所 | Grid identification method based on prime number system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101303414A (en) * | 2008-05-22 | 2008-11-12 | 北京航空航天大学 | Method for generating ground layer surface and geologic body based on level set |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010082969A1 (en) * | 2009-01-13 | 2010-07-22 | Exxonmobil Upstream Research Company | Methods and systems to volumetrically conceptualize hydrocarbon plays |
-
2012
- 2012-08-29 CN CN201210311567.1A patent/CN102867330B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101303414A (en) * | 2008-05-22 | 2008-11-12 | 北京航空航天大学 | Method for generating ground layer surface and geologic body based on level set |
Non-Patent Citations (2)
| Title |
|---|
| 地质断层三维可视化模型的构建方法与实现技术;朱良峰等;《软件学报》;20080815;第19卷(第8期);第2004-2017页 * |
| 朱良峰等.地质断层三维可视化模型的构建方法与实现技术.《软件学报》.2008,第19卷(第8期),第2004-2017页. |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102867330A (en) | 2013-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102867330B (en) | Region-division-based spatial complex horizon reconstruction method | |
| CN102222365B (en) | Method for reconstructing curved surface of complex space | |
| CN101582173B (en) | Block model building method for complex geological structure | |
| CN103454686B (en) | Small scale sedimentary facies based on strata slicing carries out the method and system of reservoir prediction | |
| CN101906965B (en) | Technology for identifying and analyzing stratums in VR (Virtual Reality) geological environment | |
| Ming et al. | GSIS: A 3D geological multi-body modeling system from netty cross-sections with topology | |
| US9378312B2 (en) | Faulted geological structures having unconformities | |
| US10995592B2 (en) | Method and system for analyzing the uncertainty of subsurface model | |
| CN103901478A (en) | Method for determining deposition characteristics and distribution of reservoirs by combining logging and seismic information | |
| CN102609982B (en) | Topology discovery method of space geological data based on unstructured mode | |
| CN102495427B (en) | Interface perception ray tracing method based on implicit model expression | |
| Wu et al. | Three-dimensional geological modeling and its application in digital mine | |
| CN102867332B (en) | Based on the multistage subdivided meshes curved surface fitting method of complex boundary constraint | |
| CN102194252A (en) | Geological-stratum-structure-based method for generating triangular lattice grids | |
| Wu et al. | A 3D modeling approach to complex faults with multi-source data | |
| CN111983691A (en) | Multi-model fusion reservoir prediction method and software system | |
| CN109102564A (en) | A kind of coupling modeling method of complex geologic body numerical model | |
| CN106355640A (en) | Ground-underground three-dimensional integrated space processing method and system | |
| CN104240300A (en) | Large-scale point cloud complex space curved face reconstruction method based on distributed parallelism | |
| CN102222366A (en) | Method for fitting complex space curved surfaces | |
| Zhang et al. | 3D parametric modeling of complex geological structures for geotechnical engineering of dam foundation based on T‐splines | |
| Zu et al. | Overview of three-dimensional geological modeling technology | |
| CN108646294B (en) | Deep water turbidite sandstone reservoir high-precision inversion method under a kind of complicated fault system | |
| CN102903149A (en) | Block forming method and device for geologic model | |
| CN102750738A (en) | Three-dimensional geologic body self-generating technology based on plane drilling |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141001 Termination date: 20170829 |