CN116595706A - Method, electronic equipment and storage medium for inverting underground structure based on width learning - Google Patents
Method, electronic equipment and storage medium for inverting underground structure based on width learning Download PDFInfo
- Publication number
- CN116595706A CN116595706A CN202310178893.8A CN202310178893A CN116595706A CN 116595706 A CN116595706 A CN 116595706A CN 202310178893 A CN202310178893 A CN 202310178893A CN 116595706 A CN116595706 A CN 116595706A
- Authority
- CN
- China
- Prior art keywords
- underground structure
- inverting
- width learning
- subsurface
- simulated earth
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000002159 abnormal effect Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000704 physical effect Effects 0.000 claims description 4
- 238000003384 imaging method Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 9
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 239000006260 foam Substances 0.000 description 5
- 229910052742 iron Inorganic materials 0.000 description 4
- 239000010410 layer Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 229910052751 metal Inorganic materials 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000000053 physical method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000002356 single layer Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 210000003462 vein Anatomy 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 239000003673 groundwater Substances 0.000 description 1
- 239000010985 leather Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The application relates to a method, electronic equipment and storage medium for inverting an underground structure based on width learning, wherein the method comprises the following steps of S1, collecting apparent resistivity data of a region to be detected; s2, inputting the apparent resistivity data into a pre-trained underground structure prediction model to generate an initial underground structure of the region to be detected; s3, combining the apparent resistivity data with the initial underground structure, and inverting by utilizing a pre-constructed inversion rule to obtain the underground structure of the region to be detected; the underground structure prediction model is obtained by inputting a pre-constructed training sample into a pre-defined width learning network for training; the training sample comprises: and simulating an earth sample and observing data corresponding to the simulated earth sample. The method has the advantages of accurate prediction of the initial underground structure, high inversion speed, high result precision and small deep imaging deviation.
Description
Technical Field
The application relates to the technical field of geophysical inversion, in particular to a method, electronic equipment and a storage medium for inverting an underground structure based on width learning.
Background
Geophysics are a natural discipline for studying the earth by physical methods, and have wide application in the production and living of society. For example, it can be used in exploration of oil, gas, metal vein, etc., locating groundwater layers, finding archaeological remains, determining thickness of glaciers, soil, etc., as well as monitoring natural disasters, environmental monitoring, etc. The geophysical observation data is utilized to directly image the underground structure, the imaging precision is low, and the underground complex structure is difficult to explain.
In order to better understand the subsurface structure, in the prior art, geophysical inversion often uses an optimization algorithm to invert the observed data to obtain the actual subsurface structure. Common algorithms include genetic algorithms, particle swarm optimization algorithms, simulated annealing algorithms, conjugate gradient algorithms, nonlinear conjugate gradient algorithms, gaussian Newton's algorithm, finite memory Newton's algorithm, and the like. In the prior art, the obtained underground structure information can be directly observed, accurate priori information cannot be obtained, and the common uniform half space is the initial underground structure, so that the technical problems of low inversion speed, low result precision and large imaging deviation along with the increase of the exploration depth are caused by larger deviation from the actual underground structure.
Disclosure of Invention
First, the technical problem to be solved
In view of the above-mentioned shortcomings and disadvantages of the prior art, the application provides a method, electronic equipment and storage medium for inverting an underground structure based on width learning, which solve the technical problems of inaccurate initial underground structure, low inversion speed, low result precision and large deep imaging deviation.
(II) technical scheme
To achieve the above object, in a first aspect, the present application provides a method for inverting a subsurface structure based on width learning, comprising:
s1, acquiring apparent resistivity data of a region to be detected;
s2, inputting the apparent resistivity data into a pre-trained underground structure prediction model to generate an initial underground structure of the region to be detected;
s3, combining the apparent resistivity data with the initial underground structure, and inverting by utilizing a pre-constructed inversion rule to obtain the underground structure of the region to be detected;
the underground structure prediction model is obtained by inputting a pre-constructed training sample into a pre-defined width learning network for training;
the training sample comprises: and simulating an earth sample and observing data corresponding to the simulated earth sample.
Optionally, the method further comprises:
s4, judging the underground abnormal body information of the region to be detected based on the underground structure.
Optionally, the observation data corresponding to the simulated earth sample specifically includes:
and performing forward calculation on the simulated earth sample to obtain apparent resistivity data corresponding to the simulated earth sample.
Alternatively, for a simulated earth sample, the simulated earth sample includes only one anomaly.
Optionally, the resistivity value of the anomaly of the simulated earth sample is a random number between 1-500 Ω·m.
Alternatively, the size and location of the anomaly are randomly generated for different simulated earth samples.
Alternatively, the background value of the simulated earth sample is a random number conforming to the physical property variation range for different simulated earth samples.
Optionally, the background value of the simulated earth sample is a random number between 1-500 Ω·m.
In a second aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the method of inverting a subsurface structure based on width learning as set forth in any one of the first aspects above.
In a third aspect, an embodiment of the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method of inverting a subsurface structure based on width learning as described in any of the first aspects above.
(III) beneficial effects
According to the method for inverting the underground structure based on the width learning, which is provided by the application, the width learning network is trained through a large number of simulated earth samples and observation data corresponding to the simulated earth samples, the time consumption of the width learning training of the single-layer network structure is short, the mapping capability and the generalization capability are strong, the model training can be completed rapidly, an initial underground structure prediction model is generated, and the initial underground structure is obtained. And then carrying out inversion by combining the measured apparent resistivity data of the region to be measured and the initial underground structure to obtain the underground structure of the region to be measured. Based on the method, the underground structure is closer to an actual underground structure, and in the subsequent inversion calculation, the inversion convergence speed is high, the imaging precision is high, and the deep imaging deviation is small.
Drawings
FIG. 1 is a flow chart of a method for inverting a subsurface structure based on width learning according to an embodiment of the application;
FIG. 2 is a flow chart of a method for inverting a subsurface structure based on width learning according to another embodiment of the application;
FIG. 3 (a) is a schematic view of a practical underground structure according to another embodiment of the present application;
FIG. 3 (b) is a schematic diagram of an initial subsurface structure provided by an embodiment of the application.
FIG. 3 (c) is a schematic representation of the inversion results of a prior art inversion of a uniform initial subsurface structure, as provided in one embodiment of the application.
FIG. 3 (d) is a schematic diagram of inversion results for inverting a subsurface structure based on breadth-learning, as provided in one embodiment of the application.
FIG. 4 is a schematic diagram of the inverted RMS curve provided in the embodiment of FIG. 3 (c), FIG. 3 (d);
FIG. 5 (a) is a schematic diagram of three high-density measuring lines laid on the surface of the area to be measured according to the present embodiment;
FIG. 5 (b) is a low resistance anomaly iron case used in the present example;
FIG. 5 (c) is a high resistance anomaly foam box used in this example;
FIG. 6 (a) is a schematic view of a slice of the initial subsurface structure at a depth of 0.1 meters before embedding an anomaly provided by the embodiment of FIG. 5.
FIG. 6 (b) is a schematic representation of inversion slicing at a depth of 0.1 meters of inversion results for a uniform initial subsurface structure prior to embedding anomalies as provided by the embodiment of FIG. 5.
FIG. 6 (c) is a schematic view of an inversion slice of the inversion result of the initial subsurface structure at a depth of 0.1 meters before embedding an anomaly provided by the embodiment of FIG. 5.
Fig. 6 (d) is a schematic view of a slice of the initial subsurface structure at a depth of 0.1 meters after embedding an anomaly provided by the embodiment of fig. 5.
FIG. 6 (e) is a schematic representation of inversion slicing at a depth of 0.1 meters of inversion results for a uniform initial subsurface structure after embedding anomalies as provided by the embodiment of FIG. 5.
FIG. 6 (f) is a schematic view of an inversion slice of the initial subsurface structure inversion result at a depth of 0.1 meters after embedding an anomaly provided by the embodiment of FIG. 5.
FIG. 6 (g) is a schematic representation of a percent change slice of the initial subsurface structure at a depth of 0.1 meters before and after embedding an anomaly in the embodiment of FIG. 5.
FIG. 6 (h) is a graph of a percent change slice of the inversion result of the uniform initial subsurface structure at a depth of 0.1 meters before and after embedding anomalies in accordance with the embodiment of FIG. 5.
FIG. 6 (i) is a schematic representation of a percent change slice of the inversion result of the initial subsurface structure at a depth of 0.1 meters before and after embedding anomalies in accordance with the embodiment of FIG. 5.
FIG. 7 (a) is an inverted RMS curve for two subsurface structures prior to embedding anomalies, as provided in one embodiment of the application.
FIG. 7 (b) is an inverted RMS curve for two subsurface structures following the embedding of anomalies provided in one embodiment of the application.
Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The application will be better explained by the following detailed description of the embodiments with reference to the drawings. It is to be understood that the specific embodiments described below are merely illustrative of the relevant application and are not limiting of the application. In addition, it should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other; for convenience of description, only parts related to the application are shown in the drawings.
Geophysics study the earth through physical methods, and can be used for exploring underground resources such as water source metal veins, however, in actual exploration, because the underground structure information obtained directly is very little, the existing uniform half-space underground structure has larger deviation from the actual underground structure, so that inversion speed is slow, result precision is low, and imaging deviation is large along with the increase of exploration depth. Therefore, the application provides a method for inverting an underground structure based on width learning, which is used for determining an initial underground structure which is closer to an actual underground structure, and inverting based on the initial underground structure to obtain the underground structure of a region to be detected.
As shown in fig. 1 and fig. 2, fig. 1 and fig. 2 are diagrams showing a method for inverting a subsurface structure based on width learning according to various embodiments of the present application, where the steps mainly include:
s1, acquiring apparent resistivity data of a region to be detected.
S2, inputting the apparent resistivity data into a pre-trained underground structure prediction model, and generating an initial underground structure of the region to be detected.
S3, combining the apparent resistivity data with the initial underground structure, and inverting by utilizing a pre-constructed inversion rule to obtain the underground structure of the region to be detected.
The underground structure prediction model is obtained by inputting a pre-constructed training sample into a pre-defined width learning network for training.
The training sample comprises: the simulated earth sample and the observation data corresponding to the simulated earth sample are, in some embodiments, a set of simulated earth samples and the observation data corresponding to the simulated earth sample are also referred to as an observation data pair.
Specifically, in this embodiment, the observation data corresponding to the simulated earth sample is specifically:
and performing forward calculation on the simulated earth sample to obtain apparent resistivity data corresponding to the simulated earth sample.
In other embodiments, the method further includes step S4 of determining the underground abnormal body information of the region to be measured based on the underground structure of the region to be measured.
The abnormal body refers to a geological body which causes abnormal geophysical prospecting. The background value is a magnitude reflecting the geophysical background of the earth. In one embodiment, in the training samples, for each simulated earth sample, the simulated earth sample includes only one anomaly.
Further, in one embodiment, the resistivity value of the anomaly of the simulated earth sample is a random number between 1-500 Ω -m.
The size and location of the anomaly are randomly generated for different simulated earth samples.
In yet another embodiment, the background value of the simulated earth sample is a random number that matches the range of physical property variation for different simulated earth samples.
The background value of the simulated earth sample may be a random number between 1-500 Ω -m.
For example, in one embodiment, the training samples include at least ten thousand pairs of observation data, each pair of observation data simulating a background value of the earth sample, a size, a position, a resistivity value, etc. of the anomaly, are random numbers generated randomly in conformity with a range of physical properties. Further, in this embodiment, an observation data set corresponding to a simulated earth sample set is obtained by forward modeling calculation according to the observation system of the work area, as the training sample of this embodiment.
According to the technical scheme, before the actual observed apparent resistivity information is inverted, an initial underground structure is designed, a training sample set is used for training a width learning network, field actual measurement data is input into the trained width learning network, and a predicted underground structure is obtained, wherein the underground structure is closer to the actual underground structure, and in subsequent inversion iteration, compared with the prior art, the inversion iteration speed is higher, the precision is higher, and the deep exploration effect is better.
In order to better explain the above technical solutions, more specific embodiments will be introduced for more detailed explanation and description.
Example 1
In this embodiment, the background resistivity of the region to be measured is 100deg.OMEGA.m, six abnormal bodies are arranged in the region to be measured, three of which are high-resistance abnormal bodies with resistivity value of 400Ω.m, and the other three are low-resistance abnormal bodies with resistivity value of 10Ω.m. The actual underground structure of the region to be measured is shown in fig. 3 (a), and six abnormal bodies are separately and hierarchically arranged.
12000 simulated earth samples were set, each comprising only one anomaly of random size, random location, and random resistivity. And performing forward calculation on the simulated earth samples to obtain observation data paired with each simulated earth sample. And the simulated earth sample and the observed data are paired to be used as training samples, and are input into a predefined width learning network for training to obtain an underground structure prediction model.
And collecting apparent resistivity data of the region to be detected by using equipment such as an observation system of the region to be detected. The apparent resistivity data is input to the underground structure prediction model, and the initial underground structure of the region to be measured is output, as shown in fig. 3 (b), fig. 3 (b) is a schematic diagram of the initial underground structure provided in this embodiment.
And (3) carrying out inversion by combining the apparent resistivity data of the region to be detected and the initial underground structure, wherein the inversion result is shown in fig. 3 (d).
In this embodiment, in order to demonstrate the beneficial effects of the method, a control group is also provided, and a uniform initial subsurface structure in the prior art, that is, a subsurface structure based on uniform half-space inversion, is used. The inversion result is shown in fig. 3 (c).
In comparison of fig. 3 (c) with fig. 3 (d), the boundaries of the anomalies in fig. 3 (d) are more clear, the values and shapes are closer to the actual subsurface structure, and the redundant structures between anomalies are also significantly suppressed.
FIG. 4 is a schematic diagram of the inverted RMS curve provided in the embodiments of FIGS. 3 (c) and 3 (d).
As can be seen from fig. 4, the inversion is performed using the initial subsurface structure, and the RMS (root mean square) value is smaller, in this embodiment, the inversion iteration is performed 56 times using the initial subsurface structure, so that the effect of 100 times of inversion iteration with the uniform initial subsurface structure can be achieved. Therefore, the technical scheme provided by the application can be used for more efficiently and accurately inverting and determining the underground structure, and has a good inversion effect.
Example 2
Fig. 5 (a) to 5 (c) are schematic diagrams of a field experiment provided in this embodiment, and fig. 5 (a) is a schematic diagram of three high-density measuring line positions laid on the ground surface of the area to be measured in this embodiment. Fig. 5 (c) and 5 (b) are a high-resistance abnormal body foam box and a low-resistance abnormal body iron skin box, respectively, used in the present example.
The model for predicting the underground structure used in this embodiment is a model trained in advance. According to an observation system of a field experiment, modeling is performed to generate a large number of earth models (simulated earth samples) consisting of single abnormal bodies, the sizes and positions of the abnormal bodies are randomly generated, the resistivity values of the abnormal bodies and the background of the model are random numbers between 1 and 500 omega-m, and observation data corresponding to the resistivity model are obtained through forward modeling calculation. A pair of training samples consisted of a single resistivity model (modeling the earth sample) and corresponding observations, together resulting in 12000 pairs of samples as a training set. And inputting the training set into a width learning network for training to obtain the underground structure prediction model. And acquiring an initial underground structure based on the underground structure prediction model.
In this embodiment, three high-density measuring lines are L1, L2, and L3, respectively, wherein a foam box is buried under L1, a sheet iron box is buried under L2, and L3 is used as a control group. In order to better show the change of the underground structure after the abnormal body is buried, the change of the underground resistivity is reflected by adopting the percentage change rate, and the calculation formula of the percentage change is as follows:
m 1 representing inversion results before embedding the anomaly; m is m 2 The inversion result after the abnormal body is buried is represented by c > 0, the inversion result represents that the resistivity value of the underground is increased, and the resistivity value of the underground is reduced by c < 0.
FIG. 6 (a) is a schematic view of a slice of the initial subsurface structure at a depth of 0.1 meters before embedding an anomaly provided by the embodiment of FIG. 5.
FIG. 6 (b) is a schematic representation of inversion slicing at a depth of 0.1 meters of inversion results for a uniform initial subsurface structure prior to embedding anomalies as provided by the embodiment of FIG. 5.
FIG. 6 (c) is a schematic view of an inversion slice of the inversion result of the initial subsurface structure at a depth of 0.1 meters before embedding an anomaly provided by the embodiment of FIG. 5.
Fig. 6 (d) is a schematic view of a slice of the initial subsurface structure at a depth of 0.1 meters after embedding an anomaly provided by the embodiment of fig. 5.
FIG. 6 (e) is a schematic representation of inversion slicing at a depth of 0.1 meters of inversion results for a uniform initial subsurface structure after embedding anomalies as provided by the embodiment of FIG. 5.
FIG. 6 (f) is a schematic view of an inversion slice of the initial subsurface structure inversion result at a depth of 0.1 meters after embedding an anomaly provided by the embodiment of FIG. 5.
FIG. 6 (g) is a schematic representation of a percent change slice of the initial subsurface structure at a depth of 0.1 meters before and after embedding an anomaly in the embodiment of FIG. 5.
FIG. 6 (h) is a graph of a percent change slice of the inversion result of the uniform initial subsurface structure at a depth of 0.1 meters before and after embedding anomalies in accordance with the embodiment of FIG. 5.
FIG. 6 (i) is a schematic representation of a percent change slice of the inversion result of the initial subsurface structure at a depth of 0.1 meters before and after embedding anomalies in accordance with the embodiment of FIG. 5.
Comparing fig. 6 (a) (b) (c), fig. 6 (a) shows the initial subsurface structure as being close to the actual subsurface structure, and the inversion results of the subsurface structure in fig. 6 (c) are also closer to the actual subsurface structure than the inversion results of the uniform initial subsurface structure in fig. 6 (b).
Further, in connection with the percent change slices of fig. 6 (g) (h) (i), it can be verified that the initial subsurface structure is closer to the actual subsurface structure than the uniform initial subsurface structure, and the inversion is performed using the initial subsurface structure, and the obtained subsurface structure is also closer to the actual structure. Inversion is performed by using a uniform initial underground structure, and the accuracy of the inversion underground structure is limited in an electrical change area where errors occur at the left ends of L1 and L3 measuring lines.
With reference to fig. 6 (a), 6 (d) and 6 (g), after the foam box and the iron sheet box are buried, a region with an increased resistivity value appears at the position of the electrode No. 5 on the left side of the L1 line in fig. 6 (d), and a region with a significantly reduced resistivity appears at the middle of the L2 line; in fig. 6 (g), there is a region of increased resistivity under the L1 line, and there is a region of reduced resistivity in the middle of L2, these two abnormal regions correspond to the positions of the foam tank and the leather tank, respectively. It can be stated that the technical scheme provided by the application can acquire the initial underground structure which is closer to the actual underground structure, and can acquire the accurate underground structure by inverting based on the initial underground structure.
Comparing fig. 6 (e), 6 (f), 6 (h) and 6 (i), it can be seen that the inversion is performed by using the initial underground structure, the obtained underground structure is closer to reality, the abnormal region of the percentage change slice before and after the abnormal body is buried is closer to the real underground structure change region, and the electrical change regions with errors at the left ends of the L1 and L3 measuring lines are also suppressed.
FIG. 7 (a) is an inverted RMS curve of the first two subsurface structures with buried anomalies provided in this example. FIG. 7 (b) is an inverted RMS curve of two subsurface structures after embedding anomalies as provided in this example. Comparing fig. 7 (a) with fig. 7 (b), it can be seen that inversion is performed using the initial subsurface structure, the inversion RMS value decreases faster, the fitting to the observed data is better, and the inversion result is more reliable.
Based on the embodiment, the method provided by the application can be verified to be effective, and the underground structure can be more accurately and rapidly determined when actual exploration is carried out, so that the application value is high.
According to the method for inverting the underground structure based on the width learning, the width learning network is trained through a large number of simulated earth samples and observation data corresponding to the simulated earth samples, the width learning training time of the single-layer network structure is short, the mapping capability and the generalization capability are strong, the model training can be completed rapidly, and an initial underground structure prediction model is generated. And inputting the measured apparent resistivity data of the region to be measured into the initial underground structure prediction model to obtain an initial underground structure, and further, inverting the initial underground structure based on the measured apparent resistivity data of the region to be measured to obtain a final underground structure. Based on the method, the subsurface structure is closer to an actual subsurface structure, the inversion convergence speed is high, the imaging precision is high in subsequent inversion calculation, and the imaging deviation is small when exploration is deepened.
In addition, the application also provides electronic equipment, which comprises: a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the method of inverting a subsurface structure based on width learning of any of the above embodiments.
The electronic device as shown in fig. 8 may include: at least one processor 101, at least one memory 102, at least one network interface 104, and other user interfaces 103. The various components in the electronic device are coupled together by a bus system 105. It is understood that the bus system 105 is used to enable connected communications between these components. The bus system 105 includes a power bus, a control bus, and a status signal bus in addition to a data bus. But for clarity of illustration the various buses are labeled as bus system 105 in fig. 8.
The user interface 103 may include, among other things, a display, a keyboard, or a pointing device (e.g., a mouse, a trackball (trackball), or a touch pad, etc.).
It will be appreciated that the memory 102 in this embodiment may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable EPROM (EEPROM), or a flash Memory. The volatile memory may be random access memory (Random Access Memory, RAM) which acts as an external cache. By way of example, and not limitation, many forms of RAM are available, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (Double Data Rate SDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), and Direct memory bus RAM (DRRAM). The memory 102 described herein is intended to comprise, without being limited to, these and any other suitable types of memory.
In some implementations, the memory 102 stores the following elements, executable units or data structures, or a subset thereof, or an extended set thereof: an operating system 1021, and application programs 1022.
The operating system 1021 includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, for implementing various basic services and processing hardware-based tasks. Applications 1022 include various applications for implementing various application services. A program for implementing the method of the embodiment of the present application may be included in the application program 1022.
In an embodiment of the present application, the processor 101 is configured to execute the method steps provided in the first aspect by calling a program or an instruction stored in the memory 102, specifically, a program or an instruction stored in the application 1022.
The method disclosed in the above embodiment of the present application may be applied to the processor 101 or implemented by the processor 101. The processor 101 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in the processor 101 or instructions in the form of software. The processor 101 described above may be a general purpose processor, a digital signal processor, an application specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software elements in a decoding processor. The software elements may be located in a random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in the memory 102, and the processor 101 reads information in the memory 102, and in combination with its hardware, performs the steps of the method described above.
Furthermore, an embodiment of the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method of inverting a subsurface structure based on width learning as described in any of the above embodiments.
In the description of the present application, it should be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the description of the present specification, the terms "one embodiment," "some embodiments," "examples," "particular examples," or "some examples," etc., refer to particular features, structures, materials, or characteristics described in connection with the embodiment or example as being included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that alterations, modifications, substitutions and variations may be made in the above embodiments by those skilled in the art within the scope of the application.
Claims (10)
1. A method for inverting a subsurface structure based on width learning, comprising:
s1, acquiring apparent resistivity data of a region to be detected;
s2, inputting the apparent resistivity data into a pre-trained underground structure prediction model to generate an initial underground structure of the region to be detected;
s3, combining the apparent resistivity data with the initial underground structure, and inverting by utilizing a pre-constructed inversion rule to obtain the underground structure of the region to be detected;
the underground structure prediction model is obtained by inputting a pre-constructed training sample into a pre-defined width learning network for training;
the training sample comprises: and simulating an earth sample and observing data corresponding to the simulated earth sample.
2. The method of inverting a subsurface structure based on width learning of claim 1, further comprising:
s4, judging the underground abnormal body information of the region to be detected based on the underground structure.
3. The method for inversion of subsurface structures based on breadth-learning according to claim 1, wherein the observed data corresponding to the simulated earth sample is specifically:
and performing forward calculation on the simulated earth sample to obtain apparent resistivity data corresponding to the simulated earth sample.
4. The method of inverting a subsurface structure based on width learning of claim 1,
for a simulated earth sample, the simulated earth sample includes only one anomaly.
5. The method of inverting a subsurface structure based on width learning of claim 4,
the resistivity value of the abnormal body of the simulated earth sample is a random number between 1 and 500 omega-m.
6. The method of inverting a subsurface structure based on width learning of claim 5,
the size and location of the anomaly are randomly generated for different simulated earth samples.
7. The method of inverting a subsurface structure based on width learning of claim 6,
for different simulated earth samples, the background value of the simulated earth sample is a random number conforming to the physical property change range.
8. The method of inverting a subsurface structure based on width learning of claim 7,
the background value of the simulated earth sample is a random number between 1 and 500 omega m.
9. An electronic device, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the method of inverting a subsurface structure based on width learning as claimed in any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of the method of inverting subsurface structures based on width learning as claimed in any of the preceding claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310178893.8A CN116595706A (en) | 2023-02-28 | 2023-02-28 | Method, electronic equipment and storage medium for inverting underground structure based on width learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310178893.8A CN116595706A (en) | 2023-02-28 | 2023-02-28 | Method, electronic equipment and storage medium for inverting underground structure based on width learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116595706A true CN116595706A (en) | 2023-08-15 |
Family
ID=87603259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310178893.8A Pending CN116595706A (en) | 2023-02-28 | 2023-02-28 | Method, electronic equipment and storage medium for inverting underground structure based on width learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116595706A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927806A (en) * | 2019-10-29 | 2020-03-27 | 清华大学 | Magnetotelluric inversion method and magnetotelluric inversion system based on supervised descent method |
| CN111126591A (en) * | 2019-10-11 | 2020-05-08 | 重庆大学 | Magnetotelluric deep neural network inversion method based on space constraint technology |
| CN113807020A (en) * | 2021-09-28 | 2021-12-17 | 成都理工大学 | Magnetotelluric inversion method based on deep learning constraint |
| CN115238565A (en) * | 2022-06-06 | 2022-10-25 | 清华大学 | Resistivity model reconstruction network training method, electromagnetic inversion method and device |
-
2023
- 2023-02-28 CN CN202310178893.8A patent/CN116595706A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126591A (en) * | 2019-10-11 | 2020-05-08 | 重庆大学 | Magnetotelluric deep neural network inversion method based on space constraint technology |
| CN110927806A (en) * | 2019-10-29 | 2020-03-27 | 清华大学 | Magnetotelluric inversion method and magnetotelluric inversion system based on supervised descent method |
| CN113807020A (en) * | 2021-09-28 | 2021-12-17 | 成都理工大学 | Magnetotelluric inversion method based on deep learning constraint |
| CN115238565A (en) * | 2022-06-06 | 2022-10-25 | 清华大学 | Resistivity model reconstruction network training method, electromagnetic inversion method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mahmud et al. | Simulation of Earth textures by conditional image quilting | |
| Foglia et al. | Evaluating model structure adequacy: The case of the Maggia Valley groundwater system, southern Switzerland | |
| Ramirez et al. | Stochastic inversion of electrical resistivity changes using a Markov Chain Monte Carlo approach | |
| Loke et al. | Computation of optimized arrays for 3-D electrical imaging surveys | |
| Bailey et al. | Ensemble smoother assimilation of hydraulic head and return flow data to estimate hydraulic conductivity distribution | |
| Zhang et al. | An improved and stable downward continuation of potential field data: The truncated Taylor series iterative downward continuation method | |
| CN110346835B (en) | Magnetotelluric forward modeling method, forward modeling system, storage medium, and electronic device | |
| Orlandini et al. | Analytical basis for determining slope lines in grid digital elevation models | |
| CN112364911A (en) | Resistivity imaging inversion method and device and storage medium | |
| CN116842813B (en) | Three-dimensional geoelectromagnetic forward modeling method | |
| Lucieer et al. | Analyzing uncertainty in multibeam bathymetric data and the impact on derived seafloor attributes | |
| Udosen et al. | A finite integration forward solver and a domain search reconstruction solver for electrical resistivity tomography (ERT) | |
| Alvarez-Aramberri et al. | Dimensionally adaptive hp-finite element simulation and inversion of 2D magnetotelluric measurements | |
| Loke et al. | The inversion of data from very large three‐dimensional electrical resistivity tomography mobile surveys | |
| Husain et al. | Fractal dimension of India using multicore parallel processing | |
| Akça et al. | Extraction of structure-based geoelectric models by hybrid genetic algorithms | |
| Andersen et al. | Bayesian inversion of geoelectrical resistivity data | |
| Dai et al. | The forward modeling of 3D gravity and magnetic potential fields in space-wavenumber domains based on an integral method | |
| Song et al. | Finite element method for modeling 3D resistivity sounding on anisotropic geoelectric media | |
| Yi et al. | Three-dimensional anisotropic inversion of resistivity tomography data in an abandoned mine area | |
| Marescot et al. | Nonlinear inversion of geoelectric data acquired across 3D objects using a finite-element approach | |
| Sungkono | Robust interpretation of single and multiple self-potential anomalies via flower pollination algorithm | |
| Bickel | Discretization, simulation, and the value of information | |
| Kang et al. | mCSEM inversion for CO2 sequestration monitoring at a deep brine aquifer in a shallow sea | |
| Yu et al. | Imaging discrete fracture networks using the location and moment tensors of microseismic events |
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 |