CN115016008A - Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network - Google Patents
Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network Download PDFInfo
- Publication number
- CN115016008A CN115016008A CN202210485742.2A CN202210485742A CN115016008A CN 115016008 A CN115016008 A CN 115016008A CN 202210485742 A CN202210485742 A CN 202210485742A CN 115016008 A CN115016008 A CN 115016008A
- Authority
- CN
- China
- Prior art keywords
- polarization
- neural network
- source induction
- conductivity
- electric source
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/02—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation operating with propagation of electric current
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/08—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation operating with magnetic or electric fields produced or modified by objects or geological structures or by detecting devices
- G01V3/083—Controlled source electromagnetic [CSEM] surveying
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/36—Recording data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/38—Processing data, e.g. for analysis, for interpretation, for correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
Abstract
The invention relates to an electrical source induction-polarization symbiotic effect multi-parameter imaging method based on a neural network. Substituting the conductivity formula into a Maxwell equation according to the polarized medium fractional order model to deduce an electric source induction-polarization symbiosis effect formula; according to geological data of an experimental area, constructing polarized medium models with different parameters such as conductivity, polarizability, frequency dispersion coefficient, time constant and the like, carrying out numerical simulation, and constructing a sample set; optimizing and selecting the structure and the activation function of the neural network, training the neural network and optimizing the performance; and (3) carrying out multi-parameter extraction on the polarized medium on the measured data of the electric source induction-polarization symbiosis effect by using a neural network, thereby realizing multi-parameter-depth imaging. The invention aims to extract the multi-parameter information of the polarization medium, and compared with the traditional resistivity imaging method, the imaging result precision of the conductivity and the polarization rate is higher.
Description
Technical Field
The invention relates to the field of research of geophysical signal processing technology, in particular to an electric source induction-polarization symbiotic effect inversion method.
Background
In the field of geophysical exploration, the electrical information of the underground medium is obtained by inverting the actually measured geophysical data. The medium containing polymetallic ore belongs to polarized medium, the resistivity is a complex number related to frequency, so under the excitation of alternating field, the electromagnetic induction and polarization effect are generated simultaneously. Electromagnetic induction can be used for distinguishing formation lithology information, and polarization effects can cause electromagnetic response to generate an 'opposite sign' phenomenon, wherein the electromagnetic response contains information of polarized media such as metal ores. Aiming at an underground complex geological structure and a polarized medium model, parameters such as conductivity, polarizability, frequency dispersion coefficient, time constant and the like need to be extracted for the induction-polarization symbiotic effect.
CN111413738B discloses a time domain induced polarization spectrum analysis method and system for porous media, which perform joint inversion through visual polarization rate data calculated by different charging times for many times, estimate relaxation time distribution and obtain the pore size distribution.
CN110133733A discloses a conductance-polarizability multi-parameter imaging method based on a particle swarm optimization algorithm, which comprises the steps of obtaining zero-frequency conductivity and a time constant through early-late data, and then adopting the particle swarm optimization algorithm to realize parameter extraction of a polarized medium Kerr-Kerr complex conductivity model.
CN110673218A discloses a method for extracting polarized medium parameter information in transient electromagnetic response of a grounded conductor source, which is to invert a vertical magnetic field that is less affected by polarization effect to obtain underground resistivity information, obtain electric field response through forward calculation, obtain pure polarization response in measured data, and invert the pure polarization response to obtain polarization parameters. However, the method solves parameters such as resistivity, polarizability and the like respectively, so that the method has important significance for multi-parameter imaging method research of the polarized medium fractional order model.
Disclosure of Invention
The invention aims to provide an electric source induction-polarization symbiosis effect multi-parameter imaging method based on a neural network, which can be used for quickly and accurately predicting an underground electric structure of an actual underground medium.
The present invention is achieved in such a way that,
the electrical source induction-polarization symbiotic effect multi-parameter imaging method based on the neural network comprises the following steps:
1) substituting a fractional order complex conductivity formula into a Maxwell equation according to the polarized medium fractional order model to derive an electric source induction-polarization co-occurrence effect formula;
2) acquiring parameter information of an underground medium model according to geological data of an experimental area, designing polarized medium models with different conductivities, polarizabilities, dispersion coefficients and time constants, performing numerical simulation by applying the electric source induction-polarization symbiosis effect formula in the step 1, and constructing a sample set;
3) optimizing and selecting a neural network structure and an activation function, training by applying the sample set in the step (2), optimizing the performance of the neural network and establishing the neural network;
4) preprocessing the measured data of the electric source induction-polarization symbiotic effect, applying the neural network in the step 3, and extracting polarization medium model parameters of the measured data, wherein the parameters comprise zero-frequency conductivity, polarization rate, frequency dispersion coefficient, time constant and depth;
5) and (4) performing conductivity-depth and polarizability-depth imaging by applying the result of the step (4) to obtain underground medium information.
Further, the step 2 includes calculating an electrical source response according to an electrical source induction-polarization co-occurrence effect formula, and analyzing the influence of the conductivity, the polarization rate, the dispersion coefficient and the time constant on the electrical source induction-polarization co-occurrence effect; and designing polarized medium models with different conductivities, polarizabilities, dispersion coefficients and time constants according to the analysis result and geological data of the experimental area, and calculating the electric source induction-polarization symbiosis effect of each model.
Further, the step 3 comprises the following steps:
constructing different polarized medium models and sample sets of electric source induction-polarization symbiotic effects thereof;
optimally designing a neural network structure and optimally selecting an activation function according to the sample set;
inputting a sample set to carry out neural network training;
optimizing the performance of the neural network, and calculating a loss function of the neural network;
judging whether the loss function reaches a threshold value, if so, optimally designing the structure of the neural network and optimally selecting an activation function;
and storing the trained neural network.
Compared with the prior art, the invention has the beneficial effects that:
aiming at the electric source induction-polarization symbiotic effect, the invention can accurately obtain the parameter results such as conductivity, polarization rate, frequency dispersion coefficient, time constant and the like by adopting neural network inversion, is favorable for the practicability of the electric source detection technology, and provides a new technical guarantee for developing electromagnetic detection and searching mineral resources.
Drawings
FIG. 1 is a flow chart of a neural network-based electrical source induction-polarization symbiotic effect multi-parameter imaging method;
fig. 2 is a resistivity-depth effect map and a polarizability-depth effect map of one embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Examples
Take the Kerr-Kerr model as an example
Referring to fig. 1, a multi-parameter imaging method based on neural network for electrical source induction-polarization symbiotic effect includes:
1) substituting a fractional order complex conductivity formula into a Maxwell equation according to a polarized medium fractional order model to deduce an electric source induction-polarization co-occurrence effect formula;
the kirl-kirl model complex resistivity expression is:
wherein σ 0 Is the zero frequency resistivity, η is the polarizability, τ is the time constant, c is the dispersion coefficient, and ω is the angular frequency. Substituting the fractional order complex resistivity expression into a Maxwell equation, assuming that the center of the electrical source is coincident with the origin of coordinates, and the center of the electrical source and the origin of coordinates extend to-L and L along the two sides of the x-axis respectively, wherein the magnetic field vertical component expression of the electrical source induction-polarization symbiotic effect is as follows:
where I is the emission current, μ 0 Is magnetic permeability, J 1 The method is a Bessel function first-order expression, r is a receiving and transmitting distance, y is a y coordinate of a receiving point, z is a z coordinate of the receiving point, and lambda and x' are integrated variables. Gamma ray TE In order to be the reflection coefficient of the light,
2) acquiring parameter information of an underground medium model according to geological data of an experimental area, designing polarized medium models with different conductivities, polarizabilities, frequency dispersion coefficients and time constants, performing numerical simulation by applying the electric source induction-polarization symbiosis effect formula in the step 1, and constructing a sample set;
calculating the electric source response according to an electric source induction-polarization co-occurrence effect formula, and analyzing the influence of the conductivity, the polarization rate, the frequency dispersion coefficient and the time constant on the electric source induction-polarization co-occurrence effect; and designing polarized medium models with different conductivities, polarizabilities, dispersion coefficients and time constants according to the analysis result and geological data of the experimental area, and calculating the electric source induction-polarization symbiosis effect of each model.
3) Optimizing and selecting a neural network structure and an activation function, training by applying the sample set in the step (2), optimizing the performance of the neural network and establishing the neural network;
comprises the following steps:
I. constructing different polarized medium models and sample sets of electric source induction-polarization symbiotic effects thereof;
II, optimally designing a neural network structure and optimally selecting an activation function according to the sample set;
III, inputting a sample set to carry out neural network training;
performing performance optimization on the neural network, and calculating a loss function of the neural network;
v, judging whether the loss function reaches a threshold value, if so, performing step II to optimally design the structure of the neural network and optimally select an activation function;
and VI, storing the trained neural network.
4) Preprocessing the actually measured data of the electric source induction-polarization symbiosis effect, applying the neural network in the step 3, and extracting polarization medium model parameters of the actually measured data, wherein the parameters comprise zero-frequency conductivity, polarization rate, frequency dispersion coefficient, time constant and depth;
5) and (4) performing conductivity-depth and polarizability-depth imaging by applying the result of the step (4) to obtain underground medium information.
The steps 4) and 5) comprise the following steps:
A. according to the detection requirement, the superconducting quantum sensor is applied to detect the actual electrical source, and the measured data of the induction-polarization symbiosis effect is collected;
B. preprocessing the measured data, including superposition, denoising and data sampling;
C. b, extracting parameters of the pre-processed actual measurement data in the step B by using a neural network, and extracting zero-frequency conductivity, polarization rate, frequency dispersion coefficient, time constant and depth;
D. and drawing conductivity-depth imaging and polarizability-depth imaging, analyzing imaging results and acquiring underground medium information.
Fig. 2 is a resistivity-depth and polarizability-depth effect graph according to the embodiment of the present invention shown in fig. 1, and the result conforms to the theoretical model of the embodiment, so that a new idea and a new method are provided for high-precision and high-efficiency inversion of electrical source induction-polarization effect data.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (3)
1. A multi-parameter imaging method of electric source induction-polarization symbiosis effect based on a neural network is characterized by comprising the following steps:
1) substituting a fractional order complex conductivity formula into a Maxwell equation according to a polarized medium fractional order model to deduce an electric source induction-polarization co-occurrence effect formula;
2) acquiring parameter information of an underground medium model according to geological data of an experimental area, designing polarized medium models with different conductivities, polarizabilities, dispersion coefficients and time constants, performing numerical simulation by applying the electric source induction-polarization symbiosis effect formula in the step 1, and constructing a sample set;
3) optimizing and selecting the neural network structure and the activation function, training by applying the sample set in the step (2), optimizing the performance of the neural network and establishing the neural network;
4) preprocessing the measured data of the electric source induction-polarization symbiotic effect, applying the neural network in the step 3, and extracting polarization medium model parameters of the measured data, wherein the parameters comprise zero-frequency conductivity, polarization rate, frequency dispersion coefficient, time constant and depth;
5) and (5) applying the result of the step (4) to carry out conductivity-depth and polarizability-depth imaging to obtain underground medium information.
2. The method of claim 1, wherein step 2 comprises calculating an electrical source response according to an electrical source induction-polarization co-occurrence formula, and analyzing the effects of conductivity, polarization, dispersion coefficient, and time constant on the electrical source induction-polarization co-occurrence; and designing polarized medium models with different conductivities, polarizabilities, dispersion coefficients and time constants according to the analysis result and geological data of the experimental area, and calculating the electric source induction-polarization symbiosis effect of each model.
3. The method of claim 1, wherein said step 3 comprises the steps of:
constructing different polarized medium models and sample sets of electric source induction-polarization symbiotic effects thereof;
optimally designing a neural network structure and optimally selecting an activation function according to the sample set;
inputting a sample set to carry out neural network training;
optimizing the performance of the neural network, and calculating a loss function of the neural network;
judging whether the loss function reaches a threshold value, if so, optimally designing the structure of the neural network and optimally selecting an activation function;
and storing the trained neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210485742.2A CN115016008A (en) | 2022-05-06 | 2022-05-06 | Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210485742.2A CN115016008A (en) | 2022-05-06 | 2022-05-06 | Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115016008A true CN115016008A (en) | 2022-09-06 |
Family
ID=83068210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210485742.2A Pending CN115016008A (en) | 2022-05-06 | 2022-05-06 | Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115016008A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829001A (en) * | 2022-11-08 | 2023-03-21 | 中国科学院地质与地球物理研究所 | Transient electromagnetic-excitation field separation and multi-parameter information extraction method and system |
-
2022
- 2022-05-06 CN CN202210485742.2A patent/CN115016008A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829001A (en) * | 2022-11-08 | 2023-03-21 | 中国科学院地质与地球物理研究所 | Transient electromagnetic-excitation field separation and multi-parameter information extraction method and system |
| US11892588B1 (en) | 2022-11-08 | 2024-02-06 | Institute Of Geology And Geophysics, Chinese Academy Of Sciences | Method and system for transient electromagnetic-induced polarization field separation and multi-parameter information extraction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Common-azimuth seismic data fault analysis using residual UNet | |
| Dannemiller et al. | A new method for determination of magnetization direction | |
| Berdichevsky et al. | Magnetotellurics in the context of the theory of ill-posed problems | |
| CN110133733B (en) | Conductance-polarizability multi-parameter imaging method based on particle swarm optimization algorithm | |
| Fedi et al. | A fast interpretation of self-potential data using the depth from extreme points method | |
| Chen et al. | Identifying gravity anomalies caused by granitic intrusions in Nanling mineral district, China: a multifractal perspective | |
| CN110187393A (en) | A kind of boat magnetic compensation method based on generalized regression nerve networks | |
| CN115016008A (en) | Electrical source induction-polarization symbiotic effect multi-parameter imaging method based on neural network | |
| CN115292890A (en) | Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development | |
| CN110119586B (en) | Axial conductivity anisotropy transient electromagnetic three-component three-dimensional FDTD forward modeling method | |
| Dalgaard et al. | Enhancing SNMR model resolution by selecting an optimum combination of pulse moments, stacking, and gating | |
| CN111626377B (en) | Lithology recognition method, device, equipment and storage medium | |
| Zhang et al. | Extraction of Multiple Electrical Parameters from IP-Affected Transient Electromagnetic Data Based on LSTM-ResNet | |
| Zheng et al. | A method of using geomagnetic anomaly to recognize objects based on HOG and 2D-AVMD | |
| CN105137495A (en) | Oil gas detection method and oil gas detection system | |
| CN113553773B (en) | Ground-air electromagnetic data inversion method based on Bayesian framework combined with neural network | |
| US11624853B2 (en) | Methods for performing formation evaluation and related systems | |
| CN108376204A (en) | A kind of electromagnetism broad sense skin depth computational methods based on underground harsh media model | |
| Zhang et al. | Determining bed boundaries from inversion of EM logging data using general measures of model structure and data misfit | |
| CN107193048B (en) | A kind of metal interference elimination method of tunnel geological forecast Transient electromagnetic response | |
| CN113406709B (en) | Ground-air electromagnetic data retrieval method based on high-generalization neural network | |
| Ma et al. | Three-dimensional gravitational and magnetic-data acquisition and analysis via a joint-gradient Euler-deconvolution method | |
| Liu et al. | Two-dimensional fast imaging of airborne EM data based on U-net | |
| CN113671582B (en) | Electrical source induction-polarization effect detection method based on three-component SQUID | |
| CN111158058B (en) | Method for detecting metal abandoned well based on geomagnetic memory |
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 |