CN104866653A - Method for acquiring underground three-dimensional density structure - Google Patents
Method for acquiring underground three-dimensional density structure Download PDFInfo
- Publication number
- CN104866653A CN104866653A CN201510213820.3A CN201510213820A CN104866653A CN 104866653 A CN104866653 A CN 104866653A CN 201510213820 A CN201510213820 A CN 201510213820A CN 104866653 A CN104866653 A CN 104866653A
- Authority
- CN
- China
- Prior art keywords
- model
- geologic
- information
- subsurface
- initial
- 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.)
- Granted
Links
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method for acquiring an underground three-dimensional density structure, relates to the field of geophysical exploration, and solves the technical problem that the underground three-dimensional density structure cannot be accurately acquired in the prior art. The main technical scheme of the invention is as follows: determining the ordinate, the abscissa, the depth and the two-dimensional 2D section space of the modeling area; according to the prior information and the modeling area, performing gravity inversion modeling by using a naive Bayes classification method to obtain an initial geological model; constructing the initial geological model into a 2D geological model by adopting a discrete body simulation method; splicing and fitting the 2D geological model into a 3D geological model; and carrying out visualization and structural interpretation processing on the 3D geological model to obtain an underground three-dimensional density structure. The method is mainly based on the Bayesian classification inversion principle, and combines the discrete body simulation technology to realize the accurate establishment of the gravity density model and improve the accuracy of the acquired underground three-dimensional density structure.
Description
Technical field
The present invention relates to field of geophysical exploration, particularly relate to a kind of method obtaining subsurface three-dimensional density structure.
Background technology
Obtaining Accurate subsurface three-dimensional density structure, extremely important for the distributional pattern understanding and grasping Earth geologic body.As a kind of important means of geophysical survey, gravity dimensional Modeling Technology can calculate subsurface three-dimensional density structure by the gravity field information of measured zone, and then the distributional pattern of plastid speculatively.
At present, gravity dimensional Modeling Technology mainly utilizes priori geological information to set up subsurface three-dimensional model in conjunction with gravimetric data, and then obtains subsurface three-dimensional density structure.Take a broad view of these priori geology existing information constrained under gravity dimensional Modeling Technology, usually three parts are all comprised: 3D display and geologic interpretation three steps of (2D refers to two dimension, and 3D refers to three-dimensional) and model are simulated in the foundation of initial model, 2D/3D gravitational inversion.
Inventor finds that in gravity dimensional Modeling Technology, the accurate foundation of initial model can lay a good foundation for follow-up gravitational inversion step; But prior art does not pay attention to this step, only simply set up a gimbal region according to prior imformation, accuracy is low, not only adds many workloads to follow-up work, also impact obtain the accuracy of subsurface three-dimensional density structure.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of method obtaining subsurface three-dimensional density structure, and fundamental purpose is by accurately setting up initial model, to improve the accuracy of obtained subsurface three-dimensional density structure.
For achieving the above object, the present invention mainly provides following technical scheme:
Embodiments provide a kind of method obtaining subsurface three-dimensional density structure, the method comprises the steps:
Determine the ordinate in modeling region, horizontal ordinate, the degree of depth and two-dimentional 2D profile spacing;
According to prior imformation and modeling region, utilize Naive Bayes Classification method to carry out gravitational inversion modeling, obtain initial geologic model;
Adopt discrete bodies analogy method that described initial geology model construction is become 2D geologic model;
Described 2D geologic model splicing is fitted to 3D geologic model;
Visual and structure decipher process is carried out to described 3D geologic model, obtains subsurface three-dimensional density structure.
The method of aforesaid acquisition subsurface three-dimensional density structure, described according to prior imformation and modeling region, utilize Naive Bayes Classification method to carry out gravitational inversion, obtain the step of initial geologic model, be specially:
Model parameter vectors is determined according to described prior imformation; Wherein, described model parameter vectors comprises: the buried depth of geologic body, thickness, density and wave velocity;
According to described modeling region and the funtcional relationship between prior imformation determination observation data and physical subsurface model parameter, namely participate in the kernel function of inverting;
Observation vector data are calculated, described observation vector data composing training sample according to described model parameter and kernel function;
Be different classifications by described observation vector Data Placement, and calculate the frequency of observation vector data in described training sample of each classification;
According to the frequency of observation vector data in described training sample of described kernel function, each classification, calculate the maximum likelihood solution of geologic body in model area;
Sorter is used to differentiate to meet the model parameter of described geologic body maximum likelihood solution;
Identification result according to described sorter generates geologic model, and calculates the GRAVITY ANOMALIES of described geologic model and the matching variance of A/W abnormal observation;
Wherein, as the positive number ε that described matching variance < sets, the geologic model parameter be finally inversed by by the identification result of sorter, as inversion result, obtains initial geologic model;
As the positive number ε of described matching variance >=setting, correction model parameter vector, repeats above-mentioned steps.
The method of aforesaid acquisition subsurface three-dimensional density structure, in described model area, the computing formula of the maximum likelihood solution of geologic body is:
Wherein, L (m) the maximum likelihood solution that is geologic body in model area; G (m) for just to calculate son, i.e. above-mentioned kernel function; D is observation data vector, and m is model parameter vectors, and T represents transposition, and subscript G represents gravimetric data, and subscript S represents other the known geologic datas except gravimetric data.
The method of aforesaid acquisition subsurface three-dimensional density structure, described sorter is hidden Markov chain sorter.
The method of aforesaid acquisition subsurface three-dimensional density structure, the positive number ε of described setting is 0.001-0.03.
The method of aforesaid acquisition subsurface three-dimensional density structure, described prior imformation comprises geophysical information and geological information;
Described geophysical information comprises gravity field information, electromagnetic information, earthquake information and rock physics information;
Described geological information comprises chronostratigraphic information, borehole data information and geology background information.
The method of aforesaid acquisition subsurface three-dimensional density structure, described initial geology model construction is become the step of 2D geologic model by described employing discrete bodies analogy method, is specially:
Adopt discrete bodies analogy method, in conjunction with prior imformation, gravitational inversion is carried out to described initial model, obtain 2D geologic model.
The method of aforesaid acquisition subsurface three-dimensional density structure, fits to the step of 3D geologic model, is specially by described 2D geologic model splicing:
Adopt each 2D section of man-machine interaction method to described 2D geologic model to modify, stacking splicing simulates 3D geologic model.
The method of aforesaid acquisition subsurface three-dimensional density structure, adopts GOCAD D visualized simulation software to carry out visual and structure decipher process to 3D geologic model, obtains subsurface three-dimensional density structure; Or
Employing Encom PA geological data analyses, simulation softward carry out visual and structure decipher process to 3D geologic model, obtain subsurface three-dimensional density structure.
Compared with prior art, the acquisition subsurface three-dimensional density structure that the embodiment of the present invention proposes at least has following beneficial effect:
(1) method of the acquisition subsurface three-dimensional density structure of embodiment of the present invention proposition carries out gravitational inversion by bayes classification method, provides density value and the distribution range of each geologic body in model exactly, obtains initial geologic model accurately; And the accurate foundation of initial model, not only make follow-up gravity potential field matching more time saving and energy saving, key is the accuracy improving the subsurface three-dimensional density structure obtained.
(2) method of the acquisition subsurface three-dimensional density structure of embodiment of the present invention proposition, it is the gravity 3D Inverse modeling technology based on discrete bodies simulation, improve traditional discrete bodies modeling method, absorb the advantage of Bayes's inversion theory, the prior imformation of comprehensive utilization model, the existing information such as measured data and physics law carrys out inverse model parameter, take maximum likelihood function Algorithm for Solving, obtain the Posterior probability distribution of model and corresponding characteristic quantity thereof, in computing, hidden Markov chain is applied in the discriminating of gravitational inversion sorter, calculate density and the distribution range of geologic body exactly, realize the meticulous depiction of Gravity Models structure, improve the accuracy of the subsurface three-dimensional density structure obtained.
Accompanying drawing explanation
The process flow diagram of the acquisition subsurface three-dimensional density structure that Fig. 1 provides for the embodiment of the present invention 1;
The gravitational inversion process flow diagram based on Bayesian Classification Principle that Fig. 2 provides for the embodiment of the present invention 1;
Any form geologic body triangulation schematic diagram that Fig. 3 provides for the embodiment of the present invention 2;
Fig. 4 is the gravity forward modelling result schematic diagram of any form geologic body in the embodiment of the present invention 2;
Fig. 5 is the 3D illustraton of model on Ni He mining area ore body and the stratum utilizing the method for the embodiment of the present invention 1 to obtain.
Embodiment
For further setting forth the present invention for the technological means reaching predetermined goal of the invention and take and effect, below in conjunction with preferred embodiment, to a kind of method embodiment, feature and the effect thereof obtaining subsurface three-dimensional density structure proposed according to the present invention, be described in detail as follows.
Gravity dimensional Modeling Technology by the gravity field information of measured zone, can extrapolate subsurface three-dimensional density structure, and then infers the distributional pattern geologic body.More effective gravity dimensional Modeling Technology mainly utilizes gravimetric data to set up subsurface three-dimensional model in conjunction with priori geological information now, and then obtains subsurface three-dimensional density structure.Take a broad view of these priori geology existing information constrained under gravity dimensional Modeling Technology, usually three parts are all comprised: 3D display and geologic interpretation three steps of (2D refers to two dimension, and 3D refers to three-dimensional) and model are simulated in the foundation of initial model, 2D/3D gravitational inversion.Wherein, the establishment step of initial model, 2D/3D gravitational inversion simulation steps play vital effect in modeling process.What inventor found initial model sets up this step, not yet draws attention at present, also more extensive.Mainly simply set up a gimbal region according to prior imformation at present, accuracy is low, not only adds many workloads to follow-up work, makes the accuracy of obtained subsurface three-dimensional density structure low, more affects the supposition effect of geologic body distributional pattern.The simulation of 2D/3D gravitational inversion is the further optimization to initial model, and provides the structure physics of model and the space distribution of geometric parameter.At present, the simulation of 2D/3D gravitational inversion mainly contains surperficial n-body simulation n, Structure inversion simulation and discrete bodies and simulates three kinds of methods.Inventor finds, surperficial n-body simulation n method can the tectonic information of simulate geologic body, but gravity field error of fitting is larger; Structure inversion analogy method easily obtains the physical property distribution of model, the abnormal good contrast abnormal with actual measurement of the theory that model produces, but Structure inversion is difficult to the constraint adding priori geological information, analog result reflection geologic body macroscopic view distribution still can, local detail is often larger with actual gap; Discrete bodies analogy method then has the comprehensive advantage of above-mentioned two kinds of methods, both geological information can be reflected well, the error of fitting of observation field and theoretical field can be made again less, it fully can absorb priori geological information (as dip direction, tomography and mineralization body etc.), can also play the experience of sand smeller and the understanding to areal geology to greatest extent.Therefore, in 2D/3D gravitational inversion simulation steps, inventor thinks that discrete bodies analogy method is practical.
Based on the problems referred to above, the present inventor proposes the method for the gravity 3D Inverse modeling technical limit spacing subsurface three-dimensional density structure utilizing discrete bodies to simulate, the core of the method is according to geological knowledge, the two-dimentional geologic section building several assigning density values in modeling region, i.e. initial model, man-machine interaction method is adopted to revise these 2D model sections one by one again, until obtain satisfied data fitting and rational geologic model.Wherein, the accurate foundation of initial model is improvement emphasis of the present invention; The present inventor is based on Bayes's inversion theory, in conjunction with discrete bodies analogue technique, adopt hidden Markov chain and maximum likelihood function algorithm, meticulous density and the distribution range judging each geologic body, realize the accurate foundation of gravimetric density model, make follow-up 2.5D/3D gravity potential field matching more time saving and energy saving.
Below in conjunction with embodiment, the present invention is described in further detail.
Embodiment 1
The present embodiment provides a kind of method obtaining subsurface three-dimensional density structure, as shown in Figure 1, specifically comprises the steps:
1, modeling region is defined: determine the ordinate in modeling region, horizontal ordinate, the degree of depth and two-dimentional 2D profile spacing.
This step mainly determines modeling region according to the target geological region that will study.
2, the pre-service of prior imformation:
As shown in Figure 1, the prior imformation in this step comprises geophysical information and geological information.Geophysical information comprises gravity field information, electromagnetic information, earthquake information and rock physics information.Geological information comprises chronostratigraphic information, borehole data information and areal geology information.
This step mainly collects the existing information in target geological region, simplifies geologic unit, the physical properties of rock in measurement target geologic province, the editor of gravimetric data and process, the explanation of electromagnetism or seismic section, and the decipher etc. of drilling well well-log information.
3, the foundation of initial model: according to prior imformation, modeling region, utilizes Naive Bayes Classification method to carry out gravitational inversion modeling, obtains initial geologic model.
This step Main Basis Bayesian Classification Arithmetic, accurately provides density value and the distribution range of each geologic body in model.Particularly, adopt Naive Bayes Classification method to carry out gravitational inversion, as shown in Figure 2, key step is as follows for design cycle:
The first step, determines characteristic attribute and division: determine model parameter vectors m according to prior imformation.Wherein, the model parameter vectors physical parameter such as buried depth, thickness, density, wave velocity that comprises each geologic body in model and the scope that may change thereof.
Second step, obtains training sample: determine the funtcional relationship between observation data and physical subsurface model parameter according to priori geology, geophysical information and modeling region, namely participate in the kernel function g of inverting
gand g
s.Again by formula d=gm, go out to characterize the observation vector data d of institute's established model body by kernel function g and model parameter vectors m forward modelling.
Here, the funtcional relationship between GRAVITY ANOMALIES and model has Newton formula.When true origin is positioned at ground, Z-direction vertical is downward, and X, Y-axis are positioned at surface level, and residual mass is the gravity anomaly that the geologic body of σ produces at observation station place, and according to the law of universal gravitation, computing formula is:
In formula, (ξ, η, ζ) is the coordinate of geologic body volume element, and the coordinate that (x, y, z) is observation station, G is constant, equals 6.67 × 10
-11m
3/ (Kgs
2).
3rd step, the frequency of each classification in calculation training sample: the observation vector Data Placement generated by model is different classification.
Classification in this step is summed up as the variation range of each geologic body residual mass in model.That is, different region of variation sections is divided into according to the distribution range of model geologic body.This step is specially: for the geologic body that coordinate position in model is fixing, when volume is determined, the relative density variation range of geologic body and region country rock is divided by segmentation, according to " quality=density × volume ", be equal to the variation range dividing residual mass, again according to above-mentioned Formula of Universal Gravitation, also just calculate the gravity anomaly scope of modeling geologic body.Meanwhile, calculate the possible probability of the relative density of the geologic body that this coordinate position is fixed, namely calculate the frequency of each classification in training sample.
4th step, under calculating each class conddition, each characteristic attribute divides frequency.That is, maximum likelihood solution L (m) of each geologic body in modeling region is calculated according to prior imformation.
In this step, in model area, the computing formula of the maximum likelihood solution of each geologic body is:
Wherein, L (m) the maximum likelihood solution that is model geologic body the maximum likelihood solution of geologic body (that is, in model area); G (m) is for just to calculate son, and d is observation data vector, and m is model parameter vectors, and T represents transposition, and subscript G represents gravimetric data, and subscript S represents other the known geologic datas except gravimetric data.
5th step, uses sorter to differentiate the maximum likelihood solution of described model geologic body, and is classified.
According to above-mentioned maximum likelihood solution computing formula, require that model parameter meets observed gravimetric data and this principle of classification of priori geological information to greatest extent, each geologic body (rock mass) in accurate Modling model.
Preferably, this step uses hidden Markov chain sorter to carry out gravitational inversion modeling.Namely, in computing, hidden Markov chain is applied in the discriminating of gravitational inversion sorter, does not need the state procedure understanding the generation of model geologic body, only need control the random function of conversion, to design program thus Inversion Calculation, effectively to improve gravity three-dimensional modeling mode.
6th step, the classification results determination geologic model according to the 5th step: the identification result according to described sorter generates geologic model, and calculates matching variance.
According to Formula of Universal Gravitation, by above-mentioned kernel function, the GRAVITY ANOMALIES of forward modelling geologic model, then the variance (that is, above-mentioned matching variance) of A/W abnormal observation calculating it and priori.When this variance yields is less than some positive number ε of setting, then using this model parameter as inversion result, and then obtain initial model.In gravity modeling, variance scope should be determined according to modeling area size, general 10km
2surface area modeling region, variance yields ε is set to 0.001-0.03, and now gravity anomaly matching is better.Otherwise the characteristic attribute revising modeling geologic body divides, as: relative density scope or coordinate position, then repeat above-mentioned steps, until meet the requirements.
The enforcement of above-mentioned six steps can reference flowchart Fig. 2, according to the data prepared, the model space determined, known prior imformation, select initial model, according to selected initial model determination gravity anomaly training sample (that is: obtaining observation vector data by initial model and kernel function forward modelling), the frequency (being summed up as the variation range of computation model geologic body residual mass) of each classification in calculation training sample, Bayes classifier is adopted to carry out taxonomic history (the maximum likelihood solution by solving model geologic body) to observation vector data, what calculate original model parameter approaches interval, if this approaches interval meet the requirement of matching variance, then output recover result, obtain initial geologic model.If this approaches interval and does not meet the requirement of matching variance, then according to prior imformation, reselect initial model.
By above-mentioned six steps, density value and the distribution range of each geologic body in model can be provided exactly, obtain initial geologic model accurately, so not only make follow-up gravity potential field matching more time saving and energy saving, key is the accuracy that improve obtained subsurface three-dimensional density structure.
4, adopt discrete bodies analogy method that described initial geology model construction is become 2D geologic model.
The structure of this step mainly 2D model, reality is exactly the gravitational inversion of band prior imformation, to optimize further initial model, and provides and builds the physics of model, the space distribution of geometric parameter.
This step can adopt the computing of Tikhonov regularization method, specific as follows:
Owing to there is linear functional relation between gravimetric observation data and the model of foundation, and first kind Fredholm's equations can be expressed as
Here E
kexpression is a kth observation data, G
kx () is the kernel function of Formula of Universal Gravitation relation, the model continuous function to be asked of m (x) to be field of definition be a≤x≤b, N is the number of observation data.With matrix representation be:
Γα＝E
Wherein, α=(α
1, α
2, L, α
n)
t, E=(E
1, E
2, L, E
n)
t, Γ is the inner product symmetric positive definite matrix be made up of kernel function, Γ
knrepresent this entry of a matrix element, can α be calculated thus.
By introducing positive weighting function ω (x), department pattern m (x) being retrained according to prior imformation, making m (x) everywhere have equal power; Body plan model, makes the Euclid length of solution get minimum, obtains solution to model m (x)
The α that matrix equation solves above is substituted into, can solution to model m (x) be obtained.
5,2D geologic model is fitted to 3D geologic model.
The matching of this step mainly 2.5D/3D gravity potential field, gives the model of initial density to each, adopt each 2D section of man-machine interaction method to 2D geologic model to modify, until obtain rational geologic model and satisfied data fitting.
Particularly, as shown in Figure 1, this step is specially, first the gravity simulation of all 2D sections in modeling region is completed, according to Modelvision Pro gravity and magnetic data processing, Inversion Software, the model strike length Y-direction of every bar 2D section is shortened to profile spacing, makes initial model become 2.5D geologic model; Again Gravity Curves matching, model modification process are carried out to 2.5D model, obtain serial 2.5D geologic model; Finally, successively 2.5D model merging is become 3D model according to the spatial order of section, the theory calculating 3D model is abnormal, and with actual Anomalies contrast, the place that error of fitting is larger, turn back to 2D section to modify, so 2.5D geologic model carried out in 3D environment matching, integrate process, obtain 3D geologic model.
6, visual and structure decipher process is carried out to described 3D geologic model, obtain subsurface three-dimensional density structure.
This step mainly adopts GOCAD D visualized simulation software (or Encom PA geological data analyses, simulation softward) to carry out visual and structure decipher process to 3D geologic model, and the data after visual and structure decipher process can obtain subsurface three-dimensional density structure intuitively visually.
Embodiment 2
The present embodiment is mainly used in verifying that Bayes's inversion method sets up the feasibility of initial model.Concrete grammar is as follows: selecting with the relative density σ of country rock is 30kgm
-3any form geologic body carry out forward simulation experiment.
For convenience of calculating, for any form geologic body, triangulation methodology is adopted to be divided into the tetrahedron of several minimum unit, subdivision is regarded as unit mass spheroid to final each tetrahedron is approximate, according to the coordinate (x of these minimum unit tetrahedron bottom center's points, y, z) with triangle area calculate each tetrahedral volume, then with density product.The computing method of each tetrahedron gravity value are equivalent to spheroid and calculate, and according to Newton's law of gravitation, gravity value computing formula is:
Wherein, G=6.67 × 10
-11m
3/ (Kgs
2), M is field source quality, and h is field source central depths or buried depth, R is radius of sphericity, and X is that on plane of vision, measuring point, to the horizontal range at field source center, refers to survey line distance here, the unit of h, R, X is m, and σ is density difference or the relative density of the relative country rock of model, and unit is kgm
-3, Δ g unit is μ Gal; 1 μ Gal=10
-9ms
-2.Finally to all tetrahedral gravity value summations, obtain the gravity value of any form geologic body thus.
For closing to reality, geologic body is defined as the phacoid that major axis a equals 300m, minor axis b equals 30m, geologic body after triangulation as shown in Figure 3.Plastid is positioned at measuring point coordinate X=1000m place hypothetically, and buried depth h is 800m, and survey line overall length is 2000m, and measuring point spacing is 10m.Add the noise of normal distribution, the observed gravimetric data that forward modelling obtains as shown in Figure 4.First adopt naive Bayesian inversion program to calculate, through 10 interative computations, error is less than 0.02, meets accuracy requirement, and obtaining relative density values is 31.7578125 ± 5.2734375Kgm
-3.Visible, adopt the complicated form geologic body of triangulation methodology superposition calculation for the impact of Bayes's complementary operation and not quite, also illustrate that Bayes's inversion solution has certain stability, to gradient direction not requirement simultaneously.Adopt hidden Markov chain sorter to differentiate to carry out Inversion Calculation again, obtaining residual density value is 30.0626173207599Kgm
-3, matching variance is 1.72748273715972 × 10
-5, result and the given relative density σ of reality are 30kgm
-3closely, illustrate that Bayes's inversion method of design is effective equally to complex object.
Embodiment 3
The present embodiment is application example, and the method for the acquisition subsurface three-dimensional density structure specifically embodiment 1 provided is applied in actual geologic prospect.
Mud river iron ore is geologic examination institute of Anhui Province under Porphyrite iron ore ore_forming model and large ore deposit cluster metallogenic theory instruct, utilize probing to verify gravity-magnetic anomaly, in May, 2007 Late Cambrian.Mud river iron ore-deposit is positioned at Anhui Province's hut fir Volcanic Basin northwest edge, is in sieve river-Huang village east northeast on metallogenic belt.South westing sieve river iron ore-deposit 3km, east northeast, apart from imperial bridge iron ore-deposit 13km, has identical minerogenetic conditions with sieve river iron ore.In Ni He mining area, stratum is mainly Cretacic brick bridge group (K
1and the two mausoleum group (K of Cretacic zh)
1sh) volcanics, Cretacic Yang Wan group (K
1y) sandstone and Quaternary system (Q) gravel soil.Mining area structure is comparatively simple, is mainly the northwest (NW) inclination uniclinal structure of Stratum of Volcanic Rocks and the shallow faults after the metallogenic period.Attitude of stratum is mild, fold agensis, toward deep formation occurrence slightly fluctuations.According to forefathers' research data, mud river iron ore is mainly composed in the volcanoclastic rock of top and the brick bridge group hypomere that there is diorite-porphyrite (subvolcano rock) body, and the volcanoclastic rock of brick bridge group hypomere is the stratigraphic marker finding iron ore.The favorable structure composing ore deposit is the domal uplift position of diorite-porphyrite body, and genetic type of ore deposit is broad sense porphyrite sections ore deposit.
The discovery of mud river iron ore-deposit by no means fortuitous, its architectural feature specifically how, mineralization machanism and look for ore deposit clue what is again, what kind of has look for ore deposit to enlighten to other Kuang Ji district? except the iron ore body of the 700-1200m degree of depth that mining area finds except probing, new ore body whether is also had to exist in its deep and outer rim? these queries also existed at present, trace it to its cause, be that mining area structural research is meticulous not enough.In order to solve above-mentioned query, the present embodiment is based on 1:1 ten thousand for gravity planar survey data, and the method adopting embodiment 1 to provide carries out gravity three-dimensional modeling to Ni He mining area, portraying this mining area subsurface three-dimensional density structure, providing solid foundation for solving above-mentioned query.The modeling result that obtains of method proposed according to embodiment 1 as shown in Figure 5.
Ni He mining area 3D illustraton of model in Fig. 5 clearly depicts three-dimensional configuration and the Distribution Characteristics of rock ore body.As shown in Figure 5, the main ore body in Ni He mining area is magnetic iron ore, pyrite and Gypsum Mine; 3D model as shown in Figure 5 it can also be seen that: ore body entirety in east northeast to spread, ore body slightly lifting when extending to northeast; Magnetic iron ore and pyritous content higher, Gypsum Mine content is less; Magnetic iron ore is mainly positioned at the west and south of study area, in lensing; Pyrite mainly concentrates on the northeast in mining area, the Gypsum Mine that middle part is a small amount of; Pyrite and Gypsum Mine buried depth more shallow relative to magnetic iron ore, ore body covered depth is roughly within the scope of the 600m-1100m of underground.Northeast, study area ore body is based on the two-layer pyrite in vertical direction, and upper strata ore body small volume, in layered distribution, ore body west and south width is less, and northeast is comparatively large, and mean breadth is about 245m, and buried depth is about 600m, average thickness 40m.Lower floor's ore body volume is comparatively large, and in lensing, buried depth is at 800m-1050m, and breadth extreme is about 680m.
3D geologic model shown in Fig. 5 also discloses the orecontrolling feature in mining area, mud extraction river: in study area, stratum is mainly Quaternary system cap rock (Q), Cretacic Yang Wan group (K from top to bottom
1y), the two mausoleum group (K of Cretacic
1and Cretacic brick bridge group (K sh)
1zh).Compared with the pyrite of shallow-layer and Gypsum Mine multidigit in brick bridge group epimere stratum, small volume, in stratiform like layered distribution; Main ore body concentrates in brick bridge group hypomere irruptive rock, and irruptive rock is mainly pyroxene diorite-porphyrite and dyke rock.Intrusive body end face forms protuberance, and the western protuberance in south, mining area rises steeply, and east northeast portion, mining area protuberance is wide slow.
These are consistent with other method (metallogenic geochronology, geochemistry etc.) inference result, illustrate that the method acquisition subsurface three-dimensional density structure that the present embodiment 1 proposes is effective.
To sum up, the method of the acquisition subsurface three-dimensional density structure that the embodiment of the present invention provides, it is the gravity 3D Inverse modeling technology based on discrete bodies simulation, improve traditional discrete bodies modeling method, absorb the advantage of Bayes's inversion theory, the prior imformation of comprehensive utilization model, the existing information such as measured data and physics law carrys out inverse model parameter, take maximum likelihood function Algorithm for Solving, obtain the Posterior probability distribution of model and corresponding characteristic quantity thereof, in computing, hidden Markov chain is applied in the discriminating of gravitational inversion sorter, calculate density and the distribution range of geologic body exactly, realize the meticulous depiction of Gravity Models structure, improve the accuracy of the subsurface three-dimensional density structure obtained.
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; change can be expected easily or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of described claim.
Claims (9)
1. obtain a method for subsurface three-dimensional density structure, it is characterized in that, comprise the steps:
Determine the ordinate in modeling region, horizontal ordinate, the degree of depth and two-dimentional 2D profile spacing;
According to prior imformation and modeling region, utilize Naive Bayes Classification method to carry out gravitational inversion modeling, obtain initial geologic model;
Adopt discrete bodies analogy method that described initial geology model construction is become 2D geologic model;
Described 2D geologic model splicing is fitted to 3D geologic model;
Visual and structure decipher process is carried out to described 3D geologic model, obtains subsurface three-dimensional density structure.
2. method according to claim 1, is characterized in that, described according to prior imformation and modeling region, utilizes Naive Bayes Classification method to carry out gravitational inversion modeling, obtains the step of initial geologic model, be specially:
Model parameter vectors is determined according to described prior imformation; Wherein, described model parameter vectors comprises: the buried depth of geologic body, thickness, density and wave velocity;
Kernel function is determined according to described modeling region, prior imformation; Wherein, described kernel function represents the funtcional relationship between observation data and model parameter;
Observation vector data are determined, described observation vector data composing training sample according to described model parameter vectors and kernel function;
Be different classifications by described observation vector Data Placement, and calculate the frequency of observation vector data in described training sample of each classification;
According to the frequency of observation vector data in described training sample of described kernel function, each classification, calculate the maximum likelihood solution of geologic body in model area;
Sorter is used to differentiate to meet the model parameter of described geologic body maximum likelihood solution;
Identification result according to described sorter generates geologic model, and calculates the GRAVITY ANOMALIES of described geologic model and the matching variance of A/W abnormal observation;
Wherein, as the positive number ε that described matching variance < sets, the geologic model parameter be finally inversed by by the identification result of sorter, as inversion result, obtains initial geologic model;
As the positive number ε of described matching variance >=setting, correction model parameter vector, repeats above-mentioned steps.
3. method according to claim 2, is characterized in that, in described model area, the computing formula of the maximum likelihood solution of geologic body is:
Wherein, L (m) the maximum likelihood solution that is geologic body in model area; G (m) for just to calculate son, i.e. kernel function; D is observation data vector, and m is model parameter vectors, and T represents transposition, and subscript G represents gravimetric data, and subscript S represents other the known geologic datas except gravimetric data.
4. method according to claim 2, is characterized in that, described sorter is hidden Markov chain sorter.
5. method according to claim 2, is characterized in that, the positive number ε of described setting is 0.001-0.03.
6. method according to claim 1, is characterized in that, described prior imformation comprises geophysical information and geological information;
Described geophysical information comprises gravity field information, electromagnetic information, earthquake information and rock physics information;
Described geological information comprises chronostratigraphic information, borehole data information and geology background information.
7. method according to claim 1, is characterized in that, described initial geology model construction is become the step of 2D geologic model by described employing discrete bodies analogy method, is specially:
Adopt discrete bodies analogy method, in conjunction with prior imformation, gravitational inversion is carried out to described initial model, obtain 2D geologic model.
8. method according to claim 1, is characterized in that, described 2D geologic model is fitted to the step of 3D geologic model, is specially:
Adopt each 2D section of man-machine interaction method to described 2D geologic model to modify, superposition, splicing simulate 3D geologic model.
9. method according to claim 1, is characterized in that,
Adopt GOCAD D visualized simulation software to carry out visual and structure decipher process to 3D geologic model, obtain subsurface three-dimensional density structure; Or
Employing Encom PA geological data analyses, simulation softward carry out visual and structure decipher process to 3D geologic model, obtain subsurface three-dimensional density structure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510213820.3A CN104866653B (en) | 2015-04-29 | 2015-04-29 | Method for acquiring underground three-dimensional density structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510213820.3A CN104866653B (en) | 2015-04-29 | 2015-04-29 | Method for acquiring underground three-dimensional density structure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104866653A true CN104866653A (en) | 2015-08-26 |
| CN104866653B CN104866653B (en) | 2017-10-24 |
Family
ID=53912478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510213820.3A Active CN104866653B (en) | 2015-04-29 | 2015-04-29 | Method for acquiring underground three-dimensional density structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104866653B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105549099A (en) * | 2015-12-11 | 2016-05-04 | 中国石油大学(华东) | Apparent magnetization intensity three-dimensional inversion method based on full-space regularization downward continuation data |
| CN108229759A (en) * | 2018-02-05 | 2018-06-29 | 中国地质大学（北京） | Nonlinear discrete presumption model exploration prediction method |
| CN108828662A (en) * | 2018-05-24 | 2018-11-16 | 深圳大学 | A kind of earthquake source inverting visual analysis method, storage medium and server |
| CN109445670A (en) * | 2018-11-09 | 2019-03-08 | 深圳市有限元科技有限公司 | A kind of access method, intelligent terminal and the storage medium of simplation visualizing result |
| CN110208860A (en) * | 2019-07-02 | 2019-09-06 | 中国煤炭地质总局地球物理勘探研究院 | A kind of prediction technique and device of Igneous rock invasion range |
| CN110221344A (en) * | 2019-06-17 | 2019-09-10 | 中国地质大学（北京） | A kind of seismic full-field shape of earth's crust three-dimensional density structure and gravity Inversion method |
| CN110287620A (en) * | 2019-06-28 | 2019-09-27 | 中国地震局地球物理研究所 | Spherical coordinate system density interface forward modeling method and system suitable for surface observation face |
| CN110286416A (en) * | 2019-05-13 | 2019-09-27 | 吉林大学 | A kind of fast two-dimensional inversion of Density method based on physical property function |
| CN110376660A (en) * | 2019-08-01 | 2019-10-25 | 山东大学 | Underground engineering geological disaster slip casting effect method of real-time |
| CN110888176A (en) * | 2019-10-25 | 2020-03-17 | 东华理工大学 | Ore finding method by utilizing ground high-precision gravity measurement |
| CN111696208A (en) * | 2020-06-03 | 2020-09-22 | 中国地质调查局西安地质调查中心(西北地质科技创新中心) | Geological-geophysical three-dimensional modeling method based on multi-data fusion |
| CN111859709A (en) * | 2020-07-31 | 2020-10-30 | 河北地质大学 | Geological statistical simulation method and device for variation and transition probability of aquifer structure |
| CN112051624A (en) * | 2020-06-23 | 2020-12-08 | 江苏华勘环保科技有限公司 | Three-dimensional visualization method for determining geological survey of solid waste site |
| CN112053438A (en) * | 2020-09-18 | 2020-12-08 | 中南大学 | Ore formation deep inference modeling method based on level set |
| CN112147707A (en) * | 2020-08-06 | 2020-12-29 | 中煤科工集团西安研究院有限公司 | Borehole transient electromagnetic one-dimensional inversion electrical information stereo imaging method |
| CN113159321A (en) * | 2021-04-07 | 2021-07-23 | 中南大学 | Bayes inference method for fracture surface morphology under gravity constraint |
| CN114254566A (en) * | 2021-12-27 | 2022-03-29 | 厦门大学 | Neural network geological detection inversion method based on 2.5-dimensional mixed spectral element method |
| CN117372639A (en) * | 2023-10-10 | 2024-01-09 | 中国地质科学院 | Three-dimensional geological-geophysical modeling method combining sparse diaphysis section with auxiliary surface at any position |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101661115A (en) * | 2008-08-29 | 2010-03-03 | 中国石油集团东方地球物理勘探有限责任公司 | Standard framework-based method for quick three-dimensional gravity and magnetic inversion for physical properties |
| CN102901989A (en) * | 2011-07-29 | 2013-01-30 | 中国国土资源航空物探遥感中心 | Gravity field or magnetic field data based geologic body three-dimensional visualized modeling and interpretation method |
| CN104535391A (en) * | 2014-12-30 | 2015-04-22 | 中国科学院地质与地球物理研究所 | Physical geography data processing method based on layered geography models |
-
2015
- 2015-04-29 CN CN201510213820.3A patent/CN104866653B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101661115A (en) * | 2008-08-29 | 2010-03-03 | 中国石油集团东方地球物理勘探有限责任公司 | Standard framework-based method for quick three-dimensional gravity and magnetic inversion for physical properties |
| CN102901989A (en) * | 2011-07-29 | 2013-01-30 | 中国国土资源航空物探遥感中心 | Gravity field or magnetic field data based geologic body three-dimensional visualized modeling and interpretation method |
| CN104535391A (en) * | 2014-12-30 | 2015-04-22 | 中国科学院地质与地球物理研究所 | Physical geography data processing method based on layered geography models |
Non-Patent Citations (2)
| Title |
|---|
| 刘彦 等: "基于离散体模拟的重力3D建模方法研究-以庐枞盆地泥河矿区为例", 《地质论评》 * |
| 祁光 等: "先验地质信息约束下的三维重磁反演建模研究-以安徽泥河铁矿为例", 《地球物理学报》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105549099A (en) * | 2015-12-11 | 2016-05-04 | 中国石油大学(华东) | Apparent magnetization intensity three-dimensional inversion method based on full-space regularization downward continuation data |
| CN108229759A (en) * | 2018-02-05 | 2018-06-29 | 中国地质大学（北京） | Nonlinear discrete presumption model exploration prediction method |
| CN108828662B (en) * | 2018-05-24 | 2020-04-07 | 深圳大学 | Seismic source inversion visual analysis method, storage medium and server |
| CN108828662A (en) * | 2018-05-24 | 2018-11-16 | 深圳大学 | A kind of earthquake source inverting visual analysis method, storage medium and server |
| CN109445670A (en) * | 2018-11-09 | 2019-03-08 | 深圳市有限元科技有限公司 | A kind of access method, intelligent terminal and the storage medium of simplation visualizing result |
| CN109445670B (en) * | 2018-11-09 | 2021-10-08 | 深圳市有限元科技有限公司 | Access method of simulation visualization result, intelligent terminal and storage medium |
| CN110286416A (en) * | 2019-05-13 | 2019-09-27 | 吉林大学 | A kind of fast two-dimensional inversion of Density method based on physical property function |
| CN110221344A (en) * | 2019-06-17 | 2019-09-10 | 中国地质大学（北京） | A kind of seismic full-field shape of earth's crust three-dimensional density structure and gravity Inversion method |
| CN110287620B (en) * | 2019-06-28 | 2022-05-27 | 中国地震局地球物理研究所 | Spherical coordinate system density interface forward modeling method and system suitable for earth surface observation surface |
| CN110287620A (en) * | 2019-06-28 | 2019-09-27 | 中国地震局地球物理研究所 | Spherical coordinate system density interface forward modeling method and system suitable for surface observation face |
| CN110208860A (en) * | 2019-07-02 | 2019-09-06 | 中国煤炭地质总局地球物理勘探研究院 | A kind of prediction technique and device of Igneous rock invasion range |
| CN110208860B (en) * | 2019-07-02 | 2021-03-26 | 中国煤炭地质总局地球物理勘探研究院 | Method and device for predicting intrusion range of igneous rock |
| CN110376660A (en) * | 2019-08-01 | 2019-10-25 | 山东大学 | Underground engineering geological disaster slip casting effect method of real-time |
| CN110888176A (en) * | 2019-10-25 | 2020-03-17 | 东华理工大学 | Ore finding method by utilizing ground high-precision gravity measurement |
| CN111696208A (en) * | 2020-06-03 | 2020-09-22 | 中国地质调查局西安地质调查中心(西北地质科技创新中心) | Geological-geophysical three-dimensional modeling method based on multi-data fusion |
| CN111696208B (en) * | 2020-06-03 | 2023-04-07 | 中国地质调查局西安地质调查中心(西北地质科技创新中心) | Geological-geophysical three-dimensional modeling method based on multi-data fusion |
| CN112051624A (en) * | 2020-06-23 | 2020-12-08 | 江苏华勘环保科技有限公司 | Three-dimensional visualization method for determining geological survey of solid waste site |
| CN112051624B (en) * | 2020-06-23 | 2023-05-23 | 江苏华勘环保科技有限公司 | Three-dimensional visualization method for determining geological survey of solid waste site |
| CN111859709A (en) * | 2020-07-31 | 2020-10-30 | 河北地质大学 | Geological statistical simulation method and device for variation and transition probability of aquifer structure |
| CN111859709B (en) * | 2020-07-31 | 2022-10-18 | 河北地质大学 | Geologic statistics simulation method and device for aquifer structure variation transition probability |
| CN112147707A (en) * | 2020-08-06 | 2020-12-29 | 中煤科工集团西安研究院有限公司 | Borehole transient electromagnetic one-dimensional inversion electrical information stereo imaging method |
| CN112147707B (en) * | 2020-08-06 | 2023-01-24 | 中煤科工集团西安研究院有限公司 | Borehole transient electromagnetic one-dimensional inversion electrical information stereo imaging method |
| CN112053438A (en) * | 2020-09-18 | 2020-12-08 | 中南大学 | Ore formation deep inference modeling method based on level set |
| CN112053438B (en) * | 2020-09-18 | 2022-07-15 | 中南大学 | Ore formation deep inference modeling method based on level set |
| CN113159321B (en) * | 2021-04-07 | 2022-05-20 | 中南大学 | Bayes inference method for fracture surface morphology under gravity constraint |
| CN113159321A (en) * | 2021-04-07 | 2021-07-23 | 中南大学 | Bayes inference method for fracture surface morphology under gravity constraint |
| CN114254566A (en) * | 2021-12-27 | 2022-03-29 | 厦门大学 | Neural network geological detection inversion method based on 2.5-dimensional mixed spectral element method |
| CN114254566B (en) * | 2021-12-27 | 2024-06-04 | 厦门大学 | Neural network geological exploration inversion method based on 2.5-dimensional mixed spectral element method |
| CN117372639A (en) * | 2023-10-10 | 2024-01-09 | 中国地质科学院 | Three-dimensional geological-geophysical modeling method combining sparse diaphysis section with auxiliary surface at any position |
| CN117372639B (en) * | 2023-10-10 | 2024-03-26 | 中国地质科学院 | Three-dimensional geological-geophysical modeling method combining sparse diaphysis section with auxiliary surface at any position |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104866653B (en) | 2017-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104866653A (en) | Method for acquiring underground three-dimensional density structure | |
| Wang et al. | Mineral potential targeting and resource assessment based on 3D geological modeling in Luanchuan region, China | |
| CN101906965B (en) | Technology for identifying and analyzing stratums in VR (Virtual Reality) geological environment | |
| CN105510993B (en) | The buried extrusion pressing type complexity gypseous-salt rock identification of foreland basin and distribution forecasting method | |
| CN106934858A (en) | Three-dimensional geological modeling method and system for scale region of mining area | |
| CN107748399B (en) | Method for identifying deep tectonic layer of mountain front zone by utilizing gravity interface inversion | |
| CN105005097A (en) | Method for comprehensive recognition of igneous rocks by employing gravity, magnetism, electromagnetism, and earthquake data | |
| CN110031896A (en) | Earthquake stochastic inversion methods and device based on Multiple-Point Geostatistics prior information | |
| CN108802851A (en) | A kind of metamorphosed sedimentary type iron-graphite mine exploitation method | |
| CN106920176B (en) | Mining area scale mineral resource estimation method and system | |
| Li et al. | Part II: A demonstration of integrating multiple-scale 3D modelling into GIS-based prospectivity analysis: A case study of the Huayuan-Malichang district, China | |
| Wang | Uncertainty quantification and reduction in the characterization of subsurface stratigraphy using limited geotechnical investigation data | |
| CN103410504A (en) | Method and device for determining true resistivity of horizontal well/highly-deviated well | |
| CN105678399A (en) | Regional mineral resource quantity estimation analysis method and system | |
| CN104536054A (en) | Stripe-shaped iron mine assessment method and device | |
| Hillier et al. | 3D form line construction by structural field interpolation (SFI) of geologic strike and dip observations | |
| CN112652066B (en) | Three-dimensional geological model-based geological characterization condition evaluation method and system | |
| KR20180055518A (en) | System and method of 3d mineral prospectivity | |
| Wang et al. | Recognition of significant surface soil geochemical anomalies via weighted 3D shortest-distance field of subsurface orebodies: a case study in the Hongtoushan copper mine, NE China | |
| Wang et al. | Multi-source data integration and multi-scale modeling framework for progressive prediction of complex geological interfaces in tunneling | |
| CN110705168A (en) | Simulation method of structural stress field | |
| Bertoncello | Conditioning surface-based models to well and thickness data | |
| Jiskani et al. | Fault orientation modeling of Sonda-Jherruck coalfield, Pakistan | |
| Lyu et al. | Classification, modeling and characterization of marine carbonate paleokarst reservoirs in Tahe Oilfield, Tarim Basin, China | |
| CN107797148A (en) | A kind of aeromagnetic anomaly field separation method and system based on three-dimensional geological modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181221 Address after: 100037 26 million village street, Xicheng District, Beijing Patentee after: Mining Resources Inst., Chinese Academy of Geology Sciences Address before: 100037 Mineral Resources Research Institute, Chinese Academy of Geological Sciences, 26, million Zhuang Street, Xicheng District, Beijing Co-patentee before: Liu Yan Patentee before: Mining Resources Inst., Chinese Academy of Geology Sciences |
|
| TR01 | Transfer of patent right |