CN109492775A - A kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing - Google Patents
A kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN109492775A CN109492775A CN201811377579.8A CN201811377579A CN109492775A CN 109492775 A CN109492775 A CN 109492775A CN 201811377579 A CN201811377579 A CN 201811377579A CN 109492775 A CN109492775 A CN 109492775A
- Authority
- CN
- China
- Prior art keywords
- geological
- default
- learning algorithm
- variety
- sensitive
- 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
Abstract
The present invention provides a kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing, is related to technical field of geological exploration.The detection method includes: the geological information for obtaining object to be measured region;Wherein, geological information includes the information of multiple seismic properties;Sensitive information corresponding with Sensitive Attributes in geological information is extracted based on a variety of predetermined Sensitive Attributes;Sensitive information is calculated based on predetermined first learning algorithm, obtains the structure interpretation result in object to be measured region.Therefore it can be realized and automatic identification carried out to the geological structure of target area to be measured, avoid artificial subjective interference geologic structure interpretation work, the plenty of time expended during artificial subjective judgement can be reduced.
Description
Technical field
The present invention relates to technical field of geological exploration, in particular to a kind of detection method of geologic structure interpretation, inspection
Survey device and readable storage medium storing program for executing.
Background technique
Geologic structure interpretation, which is one, has multi-solution and probabilistic work.Currently, the work of geologic structure interpretation
It is mainly completed by artificial subjective judgement, and needs to take considerable time during artificial subjective judgement.
Summary of the invention
The invention reside in a kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing is provided, with solution
Certainly how geologic structure interpretation is completed by machine learning algorithm to work.
In order to solve the above-mentioned technical problem, the embodiment of the present invention is accomplished in that
In a first aspect, the embodiment of the present invention provides a kind of detection method of geologic structure interpretation, comprising: obtain object to be measured
The geological information in region;Wherein, the geological information includes the information of multiple seismic properties;Based on a variety of predetermined sensitivities
Attributes extraction goes out sensitive information corresponding with the Sensitive Attributes in the geological information;It is calculated based on predetermined first study
Method calculates the sensitive information, obtains the structure interpretation result in the object to be measured region.Therefore it can be realized and treat
The geological structure for surveying target area carries out automatic identification, avoids artificial subjective interference geologic structure interpretation work, can reduce
The plenty of time expended during artificial subjective judgement.
With reference to first aspect, it the embodiment of the invention provides the first possible embodiment of first aspect, is being based on
A variety of predetermined Sensitive Attributes extract in the geological information before sensitive information corresponding with the Sensitive Attributes, institute
State method further include: geological model is established based on the object to be measured region;Forward simulation is carried out to the geological model, is obtained
Earthquake reflective data;Determine a variety of seismic properties corresponding with the earthquake reflective data;Determine that a variety of seismic properties exist
Attribute value at default geological structure;Determine that the attribute value becomes maximum value or minimum value at the default geological structure
Seismic properties are the predetermined Sensitive Attributes.Therefore it is capable of determining that for carrying out geological structure to target area to be measured
The sensitive earthquake attribute of explanation.
With reference to first aspect or the first possible embodiment of first aspect, the embodiment of the invention provides first party
The possible embodiment of second of face determines attribute value of a variety of seismic properties at default geological structure, comprising: really
Fixed attribute value of a variety of seismic properties at karst collapse col umn geological structure and at tomography geological structure.
With reference to first aspect, it the embodiment of the invention provides the third possible embodiment of first aspect, is being based on
Predetermined first learning algorithm calculates the sensitive information, obtains the structure interpretation knot in the object to be measured region
Before fruit, the method also includes: network model is established according to a variety of default learning algorithms;By the predetermined sensitive category
Property corresponding data be input in the network model, obtain the testing result of every kind of default learning algorithm;By every kind institute
The testing result and preset reference result for stating default learning algorithm compare, and obtain the accurate of every kind of default learning algorithm
Rate;Determine that the highest default learning algorithm of the accuracy rate is predetermined first learning algorithm.It therefore can be reasonable
A kind of intelligence learning algorithm is selected, realizes and efficient, high-precision geologic structure interpretation is carried out to target area to be measured.
With reference to first aspect or the third possible embodiment of first aspect, the embodiment of the invention provides first party
The 4th kind of possible embodiment in face, after establishing network model according to a variety of default learning algorithms, the method is also wrapped
It includes: geological structure being preset at least one and assigns characteristic value;It is corresponding, by the corresponding data of the predetermined Sensitive Attributes
Be input in the network model, obtain every kind of default learning algorithm testing result, comprising: by it is each it is described defaultly
The characteristic value that texture is made data corresponding with the predetermined Sensitive Attributes are input in the network model, are obtained
Every kind of default learning algorithm described presets architectonic testing result to each;By every kind of default learning algorithm
Testing result is compared with preset reference result, obtains the accuracy rate of every kind of default learning algorithm, comprising: by each institute
It states and presets architectonic testing result preset reference result corresponding with each default geological structure and be compared, obtain
Every kind of default learning algorithm to it is described at least one preset architectonic Detection accuracy.It therefore can be according to the first
Texture, which is made, confirms the most suitable default learning algorithm in object to be measured region.
With reference to first aspect or the third possible embodiment of first aspect, the embodiment of the invention provides first party
The 5th kind of possible embodiment in face, a variety of default learning algorithms include: BP neural network algorithm, RBF neural
At least two algorithms in algorithm, algorithm of support vector machine, decision Tree algorithms and random forests algorithm.Therefore the embodiment of the present invention
The geologic structure interpretation method of offer is able to use a variety of intelligence learning algorithms.
With reference to first aspect, described more the embodiment of the invention provides the 6th kind of possible embodiment of first aspect
The predetermined Sensitive Attributes of kind include: instantaneous frequency attribute, dip angle attribute, variance attribute, chaos attribute and local flatness
At least two attributes in attribute.Therefore geologic structure interpretation method provided in an embodiment of the present invention is able to confirm that out object to be measured
The sensitive earthquake attribute in region.
Second aspect, the embodiment of the present invention provide a kind of detection device of geologic structure interpretation, comprising: acquiring unit is used
In the geological information for obtaining object to be measured region;Wherein, the geological information includes the information of multiple seismic properties;It extracts single
Member, for extracting sensitivity corresponding with the Sensitive Attributes in the geological information based on a variety of predetermined Sensitive Attributes
Information;Computing unit, for being calculated the sensitive information based on predetermined first learning algorithm, obtain it is described to
Survey the structure interpretation result of target area.
In conjunction with second aspect, the embodiment of the invention provides the first possible embodiment of second aspect, detection dresses
It sets further include: forward modeling unit, for establishing geological model based on the object to be measured region;The forward modeling unit is also used to institute
It states geological model and carries out forward simulation, obtain earthquake reflective data;Processing unit, for the determining and earthquake reflective data pair
The a variety of seismic properties answered;The processing unit is also used to determine attribute of a variety of seismic properties at default geological structure
Value;The processing unit is also used to determine that the attribute value becomes the ground of maximum value or minimum value at the default geological structure
Shake attribute is the predetermined Sensitive Attributes.
In conjunction with the possible embodiment of the first of second aspect or second aspect, the embodiment of the invention provides second party
The possible embodiment of second of face, the processing unit are also used to determine a variety of seismic properties in karst collapse col umn texture
Make the attribute value at place and tomography geological structure.
In conjunction with second aspect, the embodiment of the invention provides the third possible embodiment of second aspect, the inspections
Survey device further include: modeling unit, for establishing network model according to a variety of default learning algorithms;Analytical unit is used for institute
It states predetermined Sensitive Attributes to be input in the network model, obtains the testing result of every kind of default learning algorithm;
Comparison unit obtains every kind for comparing the testing result of every kind of default learning algorithm with preset reference result
The accuracy rate of the default learning algorithm;Determination unit, for determining that the highest default learning algorithm of the accuracy rate is described
Predetermined first learning algorithm.
In conjunction with the possible embodiment of the third of second aspect or second aspect, the embodiment of the invention provides second party
The 4th kind of possible embodiment in face, detection device further include: given unit is assigned for presetting geological structure at least one
Give characteristic value;The analytical unit be also used to by it is each it is described preset the architectonic characteristic value with it is described predetermined
The corresponding data of Sensitive Attributes are input in the network model, obtain every kind of default learning algorithm to each described default
Architectonic testing result;The comparison unit is also used to described preset architectonic testing result and each institute for each
It states the corresponding preset reference result of default geological structure to be compared, obtains every kind of default learning algorithm to described at least one
It is a to preset architectonic Detection accuracy.
The third aspect, the embodiment of the present invention provide a kind of computer readable storage medium, deposit in the readable storage medium storing program for executing
Computer program is contained, when the computer program is run on computers, so that the computer executes such as first aspect
Or in first aspect geologic structure interpretation described in any embodiment detection method.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, present pre-ferred embodiments are cited below particularly,
And cooperate appended attached drawing, it is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described.It should be appreciated that the following drawings illustrates only certain embodiments of the present invention, therefore it is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
The step of Fig. 1 is a kind of detection method of geologic structure interpretation provided in an embodiment of the present invention is schemed.
Fig. 2 is the result figure that geological model provided in an embodiment of the present invention carries out forward simulation acquisition.
Fig. 3 is the data sectional view that geological model provided in an embodiment of the present invention carries out forward simulation acquisition.
Fig. 4 is first Sensitive Attributes provided in an embodiment of the present invention for carrying out obtaining after forward simulation to geological model
Figure.
Fig. 5 is second Sensitive Attributes provided in an embodiment of the present invention for carrying out obtaining after forward simulation to geological model
Figure.
Fig. 6 is intelligence learning algorithm provided in an embodiment of the present invention to the corresponding data of Sensitive Attributes and default geological structure
Characteristic value carry out calculate acquisition first result figure.
Fig. 7 is intelligence learning algorithm provided in an embodiment of the present invention to the corresponding data of Sensitive Attributes and default geological structure
Characteristic value carry out calculate acquisition second result figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description.Obviously, described embodiment is only a part of the embodiments of the present invention, instead of all the embodiments.It is logical
The component for the embodiment of the present invention being often described and illustrated herein in the accompanying drawings can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiment of the present invention provided in the accompanying drawings is not intended to limit below claimed
The scope of the present invention, but be merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art
Member's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.In addition, term " the
One ", " second " etc. is only used for distinguishing description, is not understood to indicate or imply relative importance.
Geologic structure interpretation is not only coalfield, while being also the necessary means that oil gas field understands subsurface structure development, with
Mining environment is more severe, so the requirement for structure interpretation precision is higher and higher.Geologic structure interpretation is one with more
Solution property and probabilistic work.Currently, what the work of geologic structure interpretation was mainly completed by artificial subjective judgement, and
It needs to take considerable time during artificial subjective judgement.
In view of the above problems, present inventor proposes following embodiment to solve above-mentioned ask by the exploration that studies for a long period of time
Topic.Below in conjunction with attached drawing, elaborate to the embodiment of the present invention.In the absence of conflict, following embodiment and reality
The feature applied in example can be combined with each other.
It please refers to shown in Fig. 1, the step of Fig. 1 is a kind of detection method of geologic structure interpretation provided in an embodiment of the present invention
Figure.The detection method includes:
Step S200 obtains the geological information in object to be measured region;Wherein, geological information includes the letter of multiple seismic properties
Breath.
Step S400 is extracted corresponding with Sensitive Attributes quick in geological information based on a variety of predetermined Sensitive Attributes
Feel information.
Step S600 calculates sensitive information based on predetermined first learning algorithm, obtains object to be measured area
The structure interpretation result in domain.
Intelligible, the geological information in object to be measured region can be the geological information in a certain area in the city M, be also possible to
The geological information in a certain area in the city N.
It is intelligible, for convenience those skilled in the art understand that the embodiment of the present invention, below first to step S200, step
Noun in S400 and step S600 is illustrated.Geological information contains the much information in hyperspace, contains geometry
, kinematics, dynamics and statistics feature.Geological information also includes that there are many seismic properties.Sensitive earthquake attribute refers to
For seismic properties sensitive to known geological structure in object to be measured region.First learning algorithm is appointed for different geology
A kind of algorithm that business, different work area ranges and different geological structure objective of interpretation reasonably select.
Intelligible, step S200, step S400 and step S600 are by a kind of ground texture provided in an embodiment of the present invention
The detection method for making explanation carries out geologic structure interpretation to target area to be measured, so as to realize the ground to target area to be measured
Texture makes carry out automatic identification, avoids artificial subjective interference geologic structure interpretation work, can reduce artificial subjective judgement mistake
The plenty of time expended in journey, and then can objectively complete geologic structure interpretation work.
It will elaborate below to a kind of detection method of geologic structure interpretation provided in an embodiment of the present invention.The detection
Method carries out forward simulation by the geological information to target area to be measured, establishes geological structure and the earthquake in object to be measured region
Connection between reflectance signature obtains forward modeling seismic data.Based on forward modeling seismic data, can select to target area to be measured
The seismic properties of known geological structure sensitivity.Then the data of sensitive earthquake attribute are surveyed using a variety of intelligence learning algorithms
It tries and analyzes, and the real data for selecting the highest intelligence learning algorithm of accuracy rate to carry out object to be measured region detects, realization pair
Object to be measured region carries out efficient and high-precision geologic structure interpretation.
Optionally, geological model is established based on the object to be measured region;Forward simulation is carried out to the geological model, is obtained
Obtain earthquake reflective data;Determine a variety of seismic properties corresponding with the earthquake reflective data;Determine a variety of seismic properties
Attribute value at default geological structure;Determine that the attribute value becomes maximum value or minimum value at the default geological structure
Seismic properties be the predetermined Sensitive Attributes.Therefore it is capable of determining that for carrying out ground texture to target area to be measured
Make the sensitive earthquake attribute of explanation.
Optionally, attribute value of a variety of seismic properties at default geological structure is determined, comprising: determine described a variety of
Attribute value of the seismic properties at karst collapse col umn geological structure and at tomography geological structure.
Firstly, since forward simulation be obtain known geological structure (such as characteristics of karst collapse and fault feature) earthquake it is anti-
Penetrate the effective way of signal.Before establishing geological model, it is to be understood that the geology essential structure in object to be measured region, such as to
The geology for surveying target area geology develops scale, the type of development, the hierarchical structure of geology, if passing sunk pillars on full, tomography occur
With solution cavity etc..Then it is directed to object to be measured region and establishes geological model, forward simulation is carried out to the geological model, acquisition has mark
The earthquake reflective data of will provides theoretical value for following selections for carrying out sensitive earthquake attribute and the tentative calculation of network model
According to so that reasonably selecting the real data inspection that the highest intelligence learning algorithm of accuracy rate carries out object to be measured region in network model
It surveys, realizes that architectonic efficient and high-precision detects.
It is intelligible, for different geological tasks, different work area ranges and different geologic structure interpretation object to be measured,
Corresponding geological model can be established, these geological models are different.For example, when the geology knot in area a certain to the city M
Structure carries out geologic structure interpretation, it is to be understood that the geology essential structure in the city M is directed to the geology essential structure of the city M to establish
Geological model.For another example, when the geologic structure in a certain area to the city N carries out geologic structure interpretation, it is to be understood that the geology in the city N
Essential structure, to establish the geological model for being directed to the geology essential structure of the city N.So being directed to the geological model of the city M foundation
It may be different with the geological model for being directed to the foundation of the city N.
Referring to figure 2., shown in Fig. 3, Fig. 4 and Fig. 5, Fig. 2 is that geological model provided in an embodiment of the present invention carries out forward modeling mould
The quasi- result figure obtained.Fig. 3 is the data sectional view that geological model provided in an embodiment of the present invention carries out forward simulation acquisition.Figure
4 be first Sensitive Attributes figure provided in an embodiment of the present invention for carrying out obtaining after forward simulation to geological model.Fig. 5 is this hair
What bright embodiment provided carries out the second Sensitive Attributes figure obtained after forward simulation to geological model.Wherein, in Fig. 4 and Fig. 5
X1 indicate characteristics of karst collapse at first, X2 indicates characteristics of karst collapse at second, and F1 indicates that fault feature at first, F2 indicate the
Fault feature at two.
Next, those skilled in the art are it will already have become clear that object to be measured area from seismic properties content above-mentioned
The geological information in domain can be described with seismic properties.Therefore from having in significant earthquake reflective data and can extract extensively
A variety of seismic properties.It is intelligible, since the type and quantity of seismic properties are various, a variety of seismic properties are being extracted extensively
When, a variety of seismic properties relevant to geological model can be extracted for geological model.Then from a variety of earthquakes of extraction
One or more sensitive earthquake attributes are selected in attribute.It is intelligible, in order to meet the required precision of geologic structure interpretation, this
Inventive embodiments select a variety of sensitive earthquake attributes from a variety of seismic properties of extraction.Also, a variety of sensitivities selected
Input layer data of the data of seismic properties as following network models.
Optionally, a variety of predetermined Sensitive Attributes include: instantaneous frequency attribute, dip angle attribute, variance attribute,
At least two attributes in chaos attribute and local flatness attribute.
Specifically, by Fig. 2 and Fig. 3, can learn geological model by forward simulation abscissa be at 100 and
Abscissa is at 300, and there are characteristics of karst collapse.And can also learn the geological model abscissa be 500 at and abscissa
At 800, there are fault features.It is corresponding, in figures 4 and 5, instantaneous frequency attribute (Instantaneous frequency),
Dip angle attribute (Dip illumination), variance attribute (Variance), chaos attribute (Chaos) and local flatness attribute
(Local flatness), there are attribute values maximum or minimum to characteristics of karst collapse and fault feature for this five kinds of seismic properties
There are apparent variation tendencies for value and attribute value.Therefore, in the geological model, determine this five kinds of seismic properties for sensitivity
Seismic properties.
In other words, include characteristics of karst collapse and fault feature in the geological model of foundation, fallen by analysis seismic properties
The variation at the construction of column feature and at the construction of fault feature is fallen, the sensibility of the seismic properties can be analyzed, so that it is determined that
Whether the seismic properties are sensitive earthquake attribute.Specifically, if seismic properties are in no characteristics of karst collapse and/or fault feature
Attribute value at construction is normal, is having the attribute value at characteristics of karst collapse and/or fault feature construction to have maximum value or minimum value,
And there are apparent variation tendencies for the attribute value of the seismic properties, therefore illustrate that the seismic properties are suitble to the ground of the geological model
The detection of matter structure interpretation judges the seismic properties for sensitive earthquake attribute, vice versa.In another example an if earthquake
The substantially no variation either irregular fluctuation in the geological model of the attribute value of attribute, therefore the seismic properties can not
Characteristics of karst collapse present in the geological model and/or fault feature are identified or detect, then illustrating this seismic properties discomfort
The geological model is closed, that is, thinks that the sensibility of this seismic properties is poor, determining the seismic properties not is the sensitivity of the geological model
Seismic properties.
Next, shown please continue to refer to shown in Fig. 4 and Fig. 5, and referring to figure 6 and figure 7, Fig. 6 is the embodiment of the present invention
The intelligence learning algorithm of offer is to the corresponding data of Sensitive Attributes and presets architectonic characteristic value and carries out calculating the of acquisition
One result figure.Fig. 7 is intelligence learning algorithm provided in an embodiment of the present invention to the corresponding data of Sensitive Attributes and default geology
The characteristic value of construction calculate second result figure of acquisition.Wherein, the X1 in Fig. 6 and Fig. 7 indicates that karst collapse col umn is special at first
Sign, X2 indicate that characteristics of karst collapse at second, characteristics of karst collapse value are -5.F1 indicates that fault feature at first, F2 indicate at second
Fault feature, fault feature value are 5.
Optionally, network model is established according to a variety of default learning algorithms;The predetermined Sensitive Attributes are corresponding
Data be input in the network model, obtain every kind of default learning algorithm testing result;It is described default by every kind
The testing result of learning algorithm is compared with preset reference result, obtains the accuracy rate of every kind of default learning algorithm;Really
The fixed highest default learning algorithm of accuracy rate is predetermined first learning algorithm.
Optionally, a variety of default learning algorithms include: BP neural network algorithm, RBF neural network algorithm, support to
At least two algorithms in amount machine algorithm, decision Tree algorithms and random forests algorithm.
Intelligible, the network model of the embodiment of the present invention can include a variety of intelligence learning algorithms, and a variety of intelligence
Learning algorithm may include BP (Back Propagation, backpropagation) neural network, RBF (Radial Basis
Function, radial base) (present invention is real for neural network, support vector machine (Support Vector Machine, abbreviation SVM)
The support vector machines that example uses is applied as GA-SVM, GA therein is (Genetic Algorithm, genetic algorithm)), decision tree
(Decision Tree) and five kinds of intelligence learning algorithms of random forest (Random Forest), and this five kinds of intelligence learnings are calculated
Method is applied in geologic structure interpretation.The geological model after carrying out forward simulation is carried out by this five kinds of intelligence learning algorithms
Intelligent measurement identification, analyzes each algorithm and accurately identifies ability in geologic structure interpretation.It should be noted that the present invention is real
The network model for applying example can also include other intelligence learning algorithms, use the mode and use of other intelligence learning algorithms
The mode of above-mentioned five kinds of intelligence learning algorithms is identical.
Intelligible, BP neural network has classification accuracy high, powerful fault-tolerant ability and associative memory ability these
Advantage.RBF neural have the advantages that Computationally efficient and powerful overall situation myopia the two.Support vector machines has and can solve
Certainly higher-dimension problem improves generalization ability and without relying on these advantages of entire data.Decision tree, which has, to be easy to calculate and be easy to manage
Solve and can handle the two advantages of the sample of exact properties.Random forest have will not generate overfitting and anti-interference ability
Strong advantage.
Specifically, input layer data of the data for a variety of sensitive earthquake attributes selected as network model.Therefore,
The data of 100 groups of a variety of sensitive earthquake attributes can be selected from the data of a variety of sensitive earthquake attributes.The network model and five
Kind intelligence learning algorithm can be by that can detect this 100 groups of data, so that analysis is every using the method for crosscheck
The detection accuracy of kind intelligence learning algorithm can analyze the accuracy rate of every kind of intelligent algorithm in other words, and then reasonably select accurately
The highest intelligence learning algorithm of rate carries out the real data detection in object to be measured region, realizes architectonic efficient and high-precision
Detection.
Wherein, the method for crosscheck includes being trained using 99 groups of data to the network model, then to remaining 1
Group data are tested.The mode of inspection can be every kind of intelligence learning algorithm and is trained using 99 groups of data, every kind of intelligence
Learning algorithm obtains one group of detection data respectively, these detection datas and remaining 1 group of data are compared, these inspections are examined
Test the similarity or accuracy of data Yu remaining 1 group of data.After being detected using the method for crosscheck, obtain
The testing result of every kind of intelligence learning algorithm compares these testing results and preset reference result, to analyze every kind
The detection accuracy of intelligence learning algorithm can analyze the accuracy rate of every kind of intelligent algorithm in other words.Wherein, preset reference result packet
It includes: the data for a variety of sensitive earthquake attributes selected being input in the network model and are calculated, the network mould of acquisition
The testing result of type.
It is intelligible, due to input layer data of the data as network model for a variety of sensitive earthquake attributes selected.
And there are the corresponding data of characteristics of karst collapse and the corresponding data of fault feature in the data of a variety of sensitive earthquake attributes.
Therefore the network model can also identify characteristics of karst collapse and fault feature.Meanwhile it is special to characteristics of karst collapse and tomography
Sign is identified, can also be analyzed the detection accuracy of every kind of intelligence learning algorithm, can be analyzed every kind of intelligent algorithm in other words
Accuracy rate.
Optionally, geological structure is preset at least one and assigns characteristic value;It is corresponding, by the predetermined sensitive category
Property corresponding data be input in the network model, obtain the testing result of every kind of default learning algorithm, comprising: will be every
A architectonic characteristic value data corresponding with the predetermined Sensitive Attributes of presetting are input to the net
In network model, obtains every kind of default learning algorithm and described preset architectonic testing result to each;Described in every kind
The testing result of default learning algorithm is compared with preset reference result, obtains the accurate of every kind of default learning algorithm
Rate, comprising: described preset architectonic testing result preset reference corresponding with each default geological structure for each
As a result be compared, obtain every kind of default learning algorithm to it is described at least one preset architectonic Detection accuracy.
Specifically, first default geological structure can be characteristics of karst collapse, characteristics of karst collapse can be assigned and subside
Column characteristic value is -5.Second default geological structure can be fault feature, and can assign fault feature value to fault feature is
5.By the data for a variety of sensitive earthquake attributes selected, characteristics of karst collapse value and fault feature value are input to the network model
In, obtain the testing result of every kind of intelligence learning algorithm.By the testing result of every kind of intelligence learning algorithm and each described default
The corresponding preset reference result of geological structure (preset reference result can be the characteristic value being arranged in the network model in advance,
Such as can be characteristics of karst collapse value is -5, fault feature value is 5, and unchanged geological structure characteristic value is 0) to be compared, and is obtained
Every kind of default learning algorithm to it is described at least one preset architectonic Detection accuracy.To analyze every kind of intelligence
The detection accuracy of learning algorithm can analyze the accuracy rate of every kind of intelligent algorithm in other words.And then reasonably select accuracy rate highest
Intelligence learning algorithm carry out object to be measured region real data detection, realize it is architectonic efficiently and high-precision detect.
In conclusion by the accuracy rate of every kind of intelligence learning algorithm of global alignment, thus for different geological tasks,
The object to be measured region of different work area range and different geologic structure interpretations, can reasonably select a kind of intelligence learning algorithm,
Realize efficient, high-precision geologic structure interpretation.
It is intelligible, when the accuracy rate difference of every kind of intelligence learning algorithm is little, every kind of intelligence of global alignment can be passed through
Can learning algorithm the advantages of, thus for different geological tasks, different work area ranges and different geologic structure interpretation to
Target area is surveyed, a kind of intelligence learning algorithm can be reasonably selected, realizes efficient, high-precision geologic structure interpretation.
Second aspect of the embodiment of the present invention also provides a kind of detection device of geologic structure interpretation, comprising: acquiring unit is used
In the geological information for obtaining object to be measured region;Wherein, the geological information includes the information of multiple seismic properties;It extracts single
Member, for extracting sensitivity corresponding with the Sensitive Attributes in the geological information based on a variety of predetermined Sensitive Attributes
Information;Computing unit, for being calculated the sensitive information based on predetermined first learning algorithm, obtain it is described to
Survey the structure interpretation result of target area.
Optionally, detection device further include: forward modeling unit, for establishing geological model based on the object to be measured region;
The forward modeling unit is also used to carry out forward simulation to the geological model, obtains earthquake reflective data;Processing unit, for true
Fixed a variety of seismic properties corresponding with the earthquake reflective data;The processing unit is also used to determine a variety of seismic properties
Attribute value at default geological structure;The processing unit is also used to determine the attribute value at the default geological structure
The seismic properties for becoming maximum value or minimum value are the predetermined Sensitive Attributes.
Optionally, detection device further include: modeling unit, for establishing network model according to a variety of default learning algorithms;
Analytical unit obtains every kind of default for the predetermined Sensitive Attributes to be input in the network model
Practise the testing result of algorithm;Comparison unit, for by the testing result of every kind of default learning algorithm and preset reference result
It compares, obtains the accuracy rate of every kind of default learning algorithm;Determination unit, for determining that the accuracy rate is highest pre-
If learning algorithm is predetermined first learning algorithm.
Optionally, detection device further include: given unit, for assigning the First Eigenvalue to the first geological structure;It is described
Analytical unit is also used to for the First Eigenvalue being input in the network model;Obtaining unit, for obtaining the network
First testing result of model, the first testing result of the network model are the first reference result;The comparison unit is also used
It is compared in by the testing result of every kind of default learning algorithm and first reference result, every kind of acquisition is described default
First accuracy rate of learning algorithm;The determination unit is also used to determine that the highest default learning algorithm of first accuracy rate is
Predetermined first learning algorithm.
Optionally, detection device further include: the given unit is also used to assign Second Eigenvalue to the second geological structure;
Wherein, second geological structure is different from first geological structure, and the Second Eigenvalue and the First Eigenvalue are not
Together;The analytical unit is also used to for the Second Eigenvalue being input in the network model;The obtaining unit is also used to
The second testing result of the network model is obtained, the second testing result of the network model is the second reference result;It is described
Comparison unit is also used to compare the testing result of every kind of default learning algorithm and second reference result, obtains
Second accuracy rate of every kind of default learning algorithm;The determination unit is also used to according to first accuracy rate and described
Two accuracys rate determine predetermined first learning algorithm from a variety of default learning algorithms.
The third aspect of the embodiment of the present invention also provides a kind of computer readable storage medium, deposits in the readable storage medium storing program for executing
Computer program is contained, when the computer program is run on computers, so that the computer executes such as first aspect
Or in first aspect geologic structure interpretation described in any embodiment detection method.
Through the above description of the embodiments, those skilled in the art can be understood that the present invention can lead to
Hardware realization is crossed, the mode of necessary general hardware platform can also be added to realize by software, based on this understanding, this hair
Bright technical solution can be embodied in the form of software products, which can store in a non-volatile memories
In medium (can be CD-ROM, USB flash disk, mobile hard disk etc.), including some instructions are used so that a computer equipment (can be
Personal computer, server or network equipment etc.) execute method described in each implement scene of the present invention.
In conclusion the present invention provides a kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing.
Geological structure is carried out to target area to be measured by a variety of predetermined Sensitive Attributes and predetermined first learning algorithm
It explains, so as to realize that the geological structure to target area to be measured carries out automatic identification, avoids artificial subjective interference geology
Structure interpretation work can reduce the plenty of time expended during artificial subjective judgement, and then can objectively complete geology
Structure interpretation work.
In embodiment provided by the present invention, it should be understood that disclosed devices, systems, and methods can also lead to
Other modes are crossed to realize.Devices, systems, and methods embodiment described above is only schematical, for example, in attached drawing
Flow chart and block diagram show that the system of multiple embodiments according to the present invention, the possibility of method and computer program product are real
Existing architecture, function and operation.In this regard, each box in flowchart or block diagram can represent module, a journey
A part of sequence section or code, a part of the module, section or code include one or more for realizing defined
The executable instruction of logic function.
It should also be noted that function marked in the box can also be with difference in some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can also execute in the opposite order, this depends on the function involved.It is also noted that in block diagram and or flow chart
The combination of box in each box and block diagram and or flow chart, can function or movement as defined in executing it is dedicated
Hardware based system is realized, or can be realized using a combination of dedicated hardware and computer instructions.In addition, in the present invention
Each functional module in each embodiment can integrate one independent part of formation together, and it is independent to be also possible to modules
In the presence of an independent part can also be integrated to form with two or more modules.It can replace, it can be all or part of
It is realized by software, hardware, firmware or any combination thereof on ground.When implemented in software, can entirely or partly with
The form of computer program product is realized.The computer program product includes one or more computer instructions.In computer
Upper load and when executing the computer program instructions, entirely or partly generate according to process described in the embodiment of the present invention or
Function.
The computer can be general purpose computer, special purpose computer, computer network or other programmable devices.
The computer instruction may be stored in a computer readable storage medium, or from a computer readable storage medium to another
One computer readable storage medium transmission, for example, the computer instruction can be from web-site, computer, a service
Device or data center by wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as it is infrared, wireless,
Microwave etc.) mode transmitted to another web-site, computer, server or data center.It is described computer-readable to deposit
Storage media can be any usable medium that computer can access or include the integrated clothes of one or more usable mediums
The data storage devices such as business device, data center.The usable medium can be magnetic medium, (for example, floppy disk, hard disk, tape),
Optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State Disk (SSD)) etc..
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of detection method of geologic structure interpretation characterized by comprising
Obtain the geological information in object to be measured region;Wherein, the geological information includes the information of multiple seismic properties;
Sensitive letter corresponding with the Sensitive Attributes in the geological information is extracted based on a variety of predetermined Sensitive Attributes
Breath;
The sensitive information is calculated based on predetermined first learning algorithm, obtains the structure in the object to be measured region
Make explanation results.
2. detection method according to claim 1, which is characterized in that extracted based on a variety of predetermined Sensitive Attributes
Out in the geological information before sensitive information corresponding with the Sensitive Attributes, the method also includes:
Geological model is established based on the object to be measured region;
Forward simulation is carried out to the geological model, obtains earthquake reflective data;
Determine a variety of seismic properties corresponding with the earthquake reflective data;
Determine attribute value of a variety of seismic properties at default geological structure;
The seismic properties for determining that the attribute value becomes maximum value or minimum value at the default geological structure are described preparatory
Determining Sensitive Attributes.
3. detection method according to claim 2, which is characterized in that determine a variety of seismic properties in default ground texture
Make the attribute value at place, comprising:
Determine attribute value of a variety of seismic properties at karst collapse col umn geological structure and at tomography geological structure.
4. detection method according to claim 1, which is characterized in that be based on predetermined first learning algorithm to institute
Sensitive information is stated to be calculated, before the structure interpretation result for obtaining the object to be measured region, the method also includes:
Network model is established according to a variety of default learning algorithms;
The corresponding data of the predetermined Sensitive Attributes are input in the network model, every kind of default is obtained
Practise the testing result of algorithm;
The testing result of every kind of default learning algorithm is compared with preset reference result, obtains every kind of default
Practise the accuracy rate of algorithm;
Determine that the highest default learning algorithm of the accuracy rate is predetermined first learning algorithm.
5. detection method according to claim 4, which is characterized in that establishing network mould according to a variety of default learning algorithms
After type, the method also includes:
Geological structure is preset at least one and assigns characteristic value;
It is corresponding, the corresponding data of the predetermined Sensitive Attributes are input in the network model, every kind of institute is obtained
State the testing result of default learning algorithm, comprising:
Each architectonic characteristic value data corresponding with the predetermined Sensitive Attributes of presetting are inputted
Into the network model, obtains every kind of default learning algorithm and described preset architectonic testing result to each;
The testing result of every kind of default learning algorithm is compared with preset reference result, obtains every kind of default
Practise the accuracy rate of algorithm, comprising:
Described architectonic testing result preset reference result corresponding with each default geological structure is preset for each
Be compared, obtain every kind of default learning algorithm to it is described at least one preset architectonic Detection accuracy.
6. detection method according to claim 4, which is characterized in that a variety of default learning algorithms include:
In BP neural network algorithm, RBF neural network algorithm, algorithm of support vector machine, decision Tree algorithms and random forests algorithm
At least two algorithms.
7. detection method according to claim 1, which is characterized in that a variety of predetermined Sensitive Attributes include:
At least two attributes in instantaneous frequency attribute, dip angle attribute, variance attribute, chaos attribute and local flatness attribute.
8. a kind of detection device of geologic structure interpretation characterized by comprising
Acquiring unit, for obtaining the geological information in object to be measured region;Wherein, the geological information includes multiple seismic properties
Information;
Extraction unit, for based on a variety of predetermined Sensitive Attributes extract in the geological information with the Sensitive Attributes
Corresponding sensitive information;
Computing unit, for being calculated the sensitive information based on predetermined first learning algorithm, obtain it is described to
Survey the structure interpretation result of target area.
9. detection device according to claim 8, which is characterized in that the detection device further include:
Modeling unit, for establishing network model according to a variety of default learning algorithms;
Analytical unit, for the predetermined Sensitive Attributes to be input in the network model, every kind of acquisition is described pre-
If the testing result of learning algorithm;
Comparison unit is obtained for comparing the testing result of every kind of default learning algorithm with preset reference result
The accuracy rate of every kind of default learning algorithm;
Determination unit, for determining that the highest default learning algorithm of the accuracy rate is that predetermined first study is calculated
Method.
10. a kind of computer readable storage medium, which is characterized in that it is stored with computer program in the readable storage medium storing program for executing,
When the computer program is run on computers, so that the computer is executed such as any one of claim 1-7 institute
The detection method for the geologic structure interpretation stated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811377579.8A CN109492775B (en) | 2018-11-19 | 2018-11-19 | Geological structure interpretation detection method, geological structure interpretation detection device and readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811377579.8A CN109492775B (en) | 2018-11-19 | 2018-11-19 | Geological structure interpretation detection method, geological structure interpretation detection device and readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109492775A true CN109492775A (en) | 2019-03-19 |
CN109492775B CN109492775B (en) | 2020-05-12 |
Family
ID=65697118
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811377579.8A Active CN109492775B (en) | 2018-11-19 | 2018-11-19 | Geological structure interpretation detection method, geological structure interpretation detection device and readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109492775B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110441820A (en) * | 2019-08-21 | 2019-11-12 | 中国矿业大学(北京) | Architectonic intelligent interpretation method |
CN110515117A (en) * | 2019-07-19 | 2019-11-29 | 广州市高速公路有限公司 | A kind of underground karst cavity detection method to be impulsed based on pile monkey with decision-tree model |
CN110737021A (en) * | 2019-11-06 | 2020-01-31 | 中国矿业大学(北京) | Fault recognition method and model training method and device thereof |
CN112578446A (en) * | 2019-09-30 | 2021-03-30 | 中国石油化工股份有限公司 | Method and system for depicting formation reflection disorder degree |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130297269A1 (en) * | 2010-05-28 | 2013-11-07 | Arkex Limited | Processing geophysical data |
CN104570083A (en) * | 2013-10-29 | 2015-04-29 | 中国石油天然气集团公司 | Multi-dimensional seismic attribute-based automatic geologic body identification method |
CN105842732A (en) * | 2016-03-16 | 2016-08-10 | 中国石油大学(北京) | Inversion method of multichannel sparse reflection coefficient and system thereof |
CN108318937A (en) * | 2017-12-29 | 2018-07-24 | 中国石油天然气集团公司 | Geologic interpretation method and apparatus |
-
2018
- 2018-11-19 CN CN201811377579.8A patent/CN109492775B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130297269A1 (en) * | 2010-05-28 | 2013-11-07 | Arkex Limited | Processing geophysical data |
CN104570083A (en) * | 2013-10-29 | 2015-04-29 | 中国石油天然气集团公司 | Multi-dimensional seismic attribute-based automatic geologic body identification method |
CN105842732A (en) * | 2016-03-16 | 2016-08-10 | 中国石油大学(北京) | Inversion method of multichannel sparse reflection coefficient and system thereof |
CN108318937A (en) * | 2017-12-29 | 2018-07-24 | 中国石油天然气集团公司 | Geologic interpretation method and apparatus |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110515117A (en) * | 2019-07-19 | 2019-11-29 | 广州市高速公路有限公司 | A kind of underground karst cavity detection method to be impulsed based on pile monkey with decision-tree model |
CN110441820A (en) * | 2019-08-21 | 2019-11-12 | 中国矿业大学(北京) | Architectonic intelligent interpretation method |
CN112578446A (en) * | 2019-09-30 | 2021-03-30 | 中国石油化工股份有限公司 | Method and system for depicting formation reflection disorder degree |
CN110737021A (en) * | 2019-11-06 | 2020-01-31 | 中国矿业大学(北京) | Fault recognition method and model training method and device thereof |
Also Published As
Publication number | Publication date |
---|---|
CN109492775B (en) | 2020-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Stork et al. | Application of machine learning to microseismic event detection in distributed acoustic sensing data | |
Ross et al. | P wave arrival picking and first‐motion polarity determination with deep learning | |
CN109492775A (en) | A kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing | |
US10853893B2 (en) | System and method for automatically correlating geologic tops | |
Drew et al. | Automated microseismic event detection and location by continuous spatial mapping | |
CN111596978A (en) | Web page display method, module and system for lithofacies classification by artificial intelligence | |
EP2756335B1 (en) | Method and system of subsurface horizon assignment | |
Chamarczuk et al. | Unsupervised learning used in automatic detection and classification of ambient‐noise recordings from a large‐N array | |
Vinard et al. | Localizing microseismic events on field data using a U-Net-based convolutional neural network trained on synthetic data | |
Steinberg et al. | Estimation of seismic moment tensors using variational inference machine learning | |
Jiang et al. | 3D seismic geometry quality control and corrections by applying machine learning | |
Ma et al. | Automatic fault detection with convolutional neutral networks | |
US20210374465A1 (en) | Methodology for learning a similarity measure between geophysical objects | |
US11208886B2 (en) | Direct hydrocarbon indicators analysis informed by machine learning processes | |
CN113050191A (en) | Shale oil TOC prediction method and device based on double parameters | |
CN115128671A (en) | Three-dimensional seismic facies automatic identification method based on deep learning technology | |
Bugge et al. | Data-driven identification of stratigraphic units in 3D seismic data using hierarchical density-based clustering | |
US10114135B2 (en) | System and method for optimizing seismic data analysis | |
CN111542819A (en) | Apparatus and method for improved subsurface data processing system | |
US20220291405A1 (en) | System and method for storage and retrieval of subsurface rock physical property prediction models using seismic interpretation | |
Sinha et al. | Well-test model identification with self-organizing feature map | |
US20230141334A1 (en) | Systems and methods of modeling geological facies for well development | |
US11754737B2 (en) | System and method for quantitative quality assessment of seismic surfaces | |
EP4321905A1 (en) | System and method for storage and retrieval of subsurface rock physical property prediction models using seismic interpretation | |
Sandhya et al. | CNN based Automatic Fault Detection in 3D Seismic Images. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |