A method for optimizing the configuration of controlled source electromagnetic exploration data acquisition parameters

Abstract

The present invention discloses a method for optimizing the configuration of controlled source electromagnetic exploration data acquisition parameters, comprising the following steps: step 1, using the staggered grid finite difference method to complete the numerical calculation of the three-dimensional electromagnetic response of the controlled source electromagnetic exploration and analyze the electromagnetic response characteristics; step 2, using the Euclidean distance and the first-order difference of the singular value spectrum of the sensitivity matrix to construct the quality evaluation function; step 3, establishing the objective function, using the multi-objective non-dominated sorting genetic algorithm to search the Pareto optimal solution set; step 4, according to the actual problem, extracting the optimal transmission frequency and the data acquisition parameters such as the receiving and transmitting distance from the Pareto optimal solution. The present invention is helpful to improve the resolution of the observed data to the target body, improve the reliability of the model parameter estimation; can control the exploration cost, reduce the exploration risk; and deepen the understanding of the correlation between the data acquisition parameters and the electromagnetic properties of the geological target body.

Description

Technical Field

[0001] The invention relates to the technical field of electromagnetic exploration, and in particular to a method for optimizing configuration of controllable source electromagnetic exploration data acquisition parameters. Background Art

[0002] Controlled source electromagnetic exploration uses grounded wires or ungrounded return lines as artificial sources, avoiding the influence of uncertainty factors of natural sources. It can collect high-quality signals in higher noise areas and plays an important role in mineral resource exploration, oil and gas exploration and monitoring.

[0003] Theoretically, the wider the frequency band range of controlled source electromagnetic exploration observation and the more frequencies there are, the richer the observation data obtained and the better the exploration effect. However, some studies have found that under the same observation data and the same initial model, the inversion results are controlled by several key frequencies. Randomly adding more frequency points and wider frequency band data to participate in the inversion does not significantly improve the inversion results. Instead, it increases the calculation scale and time, limiting the practical application of three-dimensional controlled source electromagnetic exploration. When the frequency range is determined, the transmission and reception distance has a major impact on the detection effect of controlled source electromagnetic exploration. For conventional ground controlled source electromagnetic exploration, the transmitting source needs to be far away from the survey area so that the observation meets the plane wave field conditions, but because of the power limit of the transmitting source, in order to ensure the signal strength of the receiving point, the transmitting source cannot be too far from the survey area. It can be seen that the configuration of parameters such as the transmitting frequency and the transmitting and receiving distance is directly related to the quality of field data acquisition, which in turn affects the detection effect of controlled source electromagnetic exploration on geological targets.

[0004] Simple data acquisition parameter optimization configuration methods are generally designed according to the geological tasks and workload requirements and in accordance with industry technical regulations. In order to achieve the purpose of detection, it is also necessary to preset geophysical models based on known geological data in the survey area, repeat numerical calculations or physical experiments, plan the survey line direction, adjust the current intensity, and select appropriate receiving and transmitting distances and transmission frequencies and other field data acquisition parameters.

[0005] A slightly more complicated method is to express the relationship between model parameters and observation data as d = Gm, where m is the model parameter, d is the observation data, and G is the forward mapping matrix. Through model perturbation analysis, the sensitivity of the observation data to the change of model parameters is checked. In addition, some auxiliary information related to the observation data and model parameters can be obtained from the linear inversion solution process, such as the model parameter resolution matrix and the data importance function.

[0006] In recent years, optimization theory based on objective function has been gradually used to solve the problem of parameter configuration of geophysical exploration data acquisition, mainly applied to resistivity tomography, seismic tomography and seismic full waveform inversion. Its theory and methods generally follow the experimental design framework proposed by Maurer et al. (2000). In terms of numerical optimization technology, global optimization algorithms such as genetic algorithm, greedy algorithm, Bayesian algorithm, etc. are usually used. In terms of objective function construction, it is mainly based on model parameter resolution matrix, point spread function, sensitivity matrix, etc. The objective function is the core of the series connection model space and data space in the optimization design of data acquisition parameters, and it characterizes the ability of observation data to resolve underground anomalies. At present, related research has developed from a single objective function to multiple objective functions.

[0007] In the above-mentioned studies on controlled source electromagnetic exploration data acquisition parameters, most of them are independent parameter analysis. However, the actual data acquisition parameters are not independent of each other, but interact with each other and jointly affect the detection effect of the geological target body. The analysis results of a single parameter change cannot well meet the actual needs of effective detection of geological bodies. Simple data acquisition parameter optimization configuration methods all use physical experiments or numerical calculations for qualitative analysis, which cannot cover the complete data acquisition parameter space. The data acquisition parameter configuration using the global optimization algorithm is generally based on one-dimensional or two-dimensional model assumptions, and cannot fully obtain detection data that reflects the spatial distribution of underground three-dimensional geological bodies. However, there are few research cases on the coordination mechanism and optimization configuration between three-dimensional controlled source electromagnetic exploration data acquisition parameters in domestic and foreign research. Summary of the invention

[0008] The purpose of the present invention is to provide a method for optimizing the configuration of controllable source electromagnetic exploration data acquisition parameters to solve the problems raised in the above background technology.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A method for optimizing and configuring controlled source electromagnetic exploration data acquisition parameters comprises the following steps:

[0011] Step 1: construct a three-dimensional geoelectric model;

[0012] Step 2: Using the staggered grid finite difference method, the Helmholtz equation based on the secondary electric field is solved to complete the numerical calculation of the three-dimensional electromagnetic response of the controlled source electromagnetic exploration;

[0013] Step 3: define the relative anomaly formula of electromagnetic response, compare the field value variation law when there is an abnormal body and when there is no abnormal body, and analyze the electromagnetic response characteristics obtained by different data acquisition parameters;

[0014] Step 4: Gaussian random noise is added to the theoretical data, and the nonlinear conjugate gradient method is used to perform three-dimensional inversion, summarize the influence of different data acquisition parameter configurations on the detection effect of the target geological body, and analyze the main control parameters;

[0015] Step 5: Display the calculated sensitivity matrix. The sensitivity matrix reflects the relationship between the physical properties of a certain point underground and the geophysical observation data;

[0016] Step 6: Analyze the structural characteristics of the singular value spectrum of the sensitivity matrix of different data acquisition parameters, and combine the electromagnetic response characteristics and the three-dimensional inversion results to reveal that larger singular values ​​and flat singular value spectra have better resolution capabilities for model parameters;

[0017] Step 7: Construct a quality evaluation function for data acquisition parameter configuration;

[0018] Step 8: construct multiple objective functions according to the quality evaluation function, wherein objective function 1 is the Euclidean distance of the singular value spectrum of the sensitivity matrix, and objective function 2 is the first-order difference of the singular value spectrum of the sensitivity matrix;

[0019] Step 9: Use a multi-objective non-dominated sorting genetic algorithm to search for a set of Pareto optimal solutions that can simultaneously minimize objective function 1 and objective function 2 under constraints;

[0020] Step 10: According to the needs of the actual problem, extract the optimal data acquisition parameters such as the transmission frequency, the receiving and transmitting distance, and the measurement point location from the Pareto optimal solution;

[0021] Step 11: Design a three-dimensional geoelectric model of the target geological body with different parameters, compare the three-dimensional inversion results of the data collected by the standard and optimized data acquisition parameters, and prove the advantages of optimizing the data acquisition parameters.

[0022] As a further technical solution of the present invention: in step 5, the sensitivity matrix is ​​the derivative of the observed data to the model parameters, and its expression is derived in accordance with the framework of Egbert & Kelbert. In order to improve the resolution of the observed data to the target geological body, the sensitivity matrix is ​​calculated using the derivative of the observed data to the target body model parameters.

[0023] As a further technical solution of the present invention: the step 6 is specifically: using the singular value decomposition technology (SVD), the sensitivity matrix J (N×M) can be expressed as: J=U r Λ r V r T , where M represents the number of model parameters, N represents the number of observations, and U r and V r are N×r and M×r eigenvector matrices respectively, Λ ris a string containing r singular values ​​(λ1,λ2,...,λ r ), the singular values ​​are all greater than zero and arranged from large to small, r is the rank of the sensitivity matrix J, and T represents the transpose. Larger singular values ​​indicate disturbances in the model space, which will cause larger changes in the data space.

[0024] As a further technical solution of the present invention: the step 7 is specifically:

[0025] Define the sensitivity matrix J for all model parameters a , which is the derivative of the observed data with respect to all model parameters. The ability of the observed data to resolve model parameters is related to the size of the sensitivity singular value and the flatness of the singular value spectrum.

[0026] In order to quantify the size of the singular values ​​of the sensitivity matrix, the quality function 1 is defined as:

[0027]

[0028] Where D a =[ln(λ a 1),ln(λ a 2),...,ln(λ a ra )] T is the sensitivity matrix J a The singular value vector of , ra is the sensitivity matrix J a rank, is the sensitivity matrix J a The maximum singular value vector of , T represents transpose. QF1 represents the Euclidean distance of the singular values ​​of the sensitivity matrix. The smaller the QF1 value, the smaller the sensitivity matrix J a The larger the singular value of

[0029] In order to quantify the flatness of the singular value spectrum of the sensitivity matrix, the quality function 2 is defined as:

[0030]

[0031] Where W is the first-order difference operator, that is, the smoothing factor, and the expression is:

[0032]

[0033] QF2 represents the sensitivity matrix J a The first-order difference of the singular value, the smaller the QF2 value, the smaller the difference between adjacent singular values, and the flatter the singular value spectrum.

[0034] According to QF1 and QF2, the quality function QF of data acquisition parameter design is defined as

[0035] QF=QF1+QF2 (4)

[0036] It characterizes the quality of the optimized design of data acquisition parameters. The smaller the QF value, the better the data acquisition parameters are, and the greater the ability of the collected observation data to distinguish the model parameters.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) It helps to improve the ability of observation data to resolve the target object and improve the reliability of model parameter estimation;

[0039] (2) It can control exploration costs and reduce exploration risks;

[0040] (3) Deepening the understanding of the correlation between data acquisition parameters and the electromagnetic properties of geological targets can provide theoretical and methodological support for the optimization design of data acquisition parameters for the effective detection of target geological bodies. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of the present invention.

[0042] Figure 2 It is a three-dimensional geoelectric model and grid division diagram;

[0043] Figure 2 In the figure, (a) is the theoretical three-dimensional model diagram, (b) is the Z=1km grid division diagram, (c) is the X=0km grid division diagram, and (d) is the Y=0km grid division diagram.

[0044] Figure 3 It is a schematic diagram of the three-dimensional inversion results of different frequency combinations at X = 0km;

[0045] Figure 3 (a) shows different frequency combination diagrams, and (b)-(f) show the inversion result diagrams of AE group frequencies respectively. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, a method for optimizing the configuration of controlled source electromagnetic exploration data acquisition parameters comprises the following steps:

[0048] (1) Construct a three-dimensional geoelectric model;

[0049] (2) Using the staggered grid finite difference method, the Helmholtz equation based on the secondary electric field is solved to complete the numerical calculation of the three-dimensional electromagnetic response of controlled source electromagnetic exploration;

[0050] (3) Define the relative anomaly formula of electromagnetic response, compare the field value variation law when there is an abnormal body and when there is no abnormal body, and analyze the electromagnetic response characteristics obtained with different data acquisition parameters;

[0051] (4) Gaussian random noise is added to the theoretical data, and the nonlinear conjugate gradient method is used to perform three-dimensional inversion. The influence of different data acquisition parameter configurations on the detection effect of the target geological body is summarized, and the main control parameters are analyzed;

[0052] (5) Display the calculated sensitivity matrix. The sensitivity matrix reflects the relationship between the physical properties of a certain point underground and the geophysical observation data;

[0053] (6) Analyze the structural characteristics of the singular value spectrum of the sensitivity matrix with different data acquisition parameters, and combine the electromagnetic response characteristics and three-dimensional inversion results to reveal that larger singular values ​​and flat singular value spectra have better resolution capabilities for model parameters;

[0054] (7) Construct a quality evaluation function for data acquisition parameter configuration;

[0055] (8) constructing multiple objective functions according to the quality evaluation function, wherein objective function 1 is the Euclidean distance of the singular value spectrum of the sensitivity matrix, and objective function 2 is the first-order difference of the singular value spectrum of the sensitivity matrix;

[0056] (9) Using a multi-objective non-dominated sorting genetic algorithm, under constraints, search for a set of Pareto optimal solutions that can simultaneously minimize objective function 1 and objective function 2;

[0057] (10) According to the needs of the actual problem, extract the optimal data acquisition parameters such as the transmission frequency, receiving and transmitting distance, and measurement point location from the Pareto optimal solution;

[0058] (11) Design a three-dimensional geoelectric model of the target geological body with different parameters, compare the three-dimensional inversion results of the data collected by standard and optimized data acquisition parameters, and prove the advantage of optimizing data acquisition parameters.

[0059] Example 2. Based on Example 1, the applicant of the present invention designed a simple three-dimensional geoelectric model and discussed the influence of the transmission frequency on the three-dimensional inversion results of conventional ground controlled source electromagnetic exploration. The theoretical model is shown in 2(a). A low-resistance anomaly with a resistivity of 30Ω·m is embedded in a uniform half-space with a resistivity of 800Ω·m. The size of the low-resistance anomaly is 1000m×1000m×500m and the burial depth is 900m. The length of the transmitting source is 1km, the coordinates of the center point are (0,-10000,0), and the transmitting current is 1A. A 21×21 equally spaced measuring network is arranged within the range of 4000m×4000m on the ground, with a total of 441 measuring points.

[0060] The three-dimensional model is divided into 42×42×29 grids in the x, y and z directions. The grid size in the x and y directions is 100m×100m. The thickness of the first layer of the grid in the z direction is 20m, and the thickness of each layer below it increases by a factor of 1.1. In order to meet the Dirichlet boundary conditions, 5 expanded grids are set in the x, y and z directions respectively, and the expansion factor is 2. The model grid division is as follows Figure 2 As shown in (b, c, d).

[0061] The number and frequency values ​​of different combinations are shown in Table 1, where groups AD are randomly selected standard frequencies, and group E is the optimized frequency selected by the method proposed in this invention. The inversion uses a nonlinear conjugate gradient algorithm, and the inversion parameters of different frequency combinations are consistent. The initial model is 800Ω·m uniform half space, and iterates 120 times. The inversion results are shown in Figure 3 In the figure, the black solid rectangle is the location of the low-resistance anomaly. By comparing the inversion results of different frequency combinations, the following conclusions can be drawn:

[0062] ① The frequencies of group A, B and C are selected from low, medium and high frequency bands respectively, and the number of frequencies is equal. Figure 3 (b, c, d) It can be seen that the inversion results of group B (mid-frequency band) are the best, while the inversion results of group C (high frequency band) and group A (low frequency band) are poor. It can be seen that frequency selection has a great influence on the detection effect of controlled source electromagnetic exploration;

[0063] ② The number of frequencies in groups D and E are 18 and 6 respectively. Group D is the standard frequency and group E is the optimized frequency. Comparing the inversion results ( Figure 3 e, f) It can be seen that randomly adding more frequency points and wider frequency band data to the inversion does not significantly improve the inversion results.

[0064] For 3D controlled source electromagnetic exploration, the actual data acquisition parameters interact with each other and jointly affect the detection effect of the geological target body. Facing the target geological body, it is necessary to further analyze the main controlling factors affecting the detection effect, reveal the data acquisition parameter influence mechanism of the observation data on the model parameter resolution ability, construct data acquisition parameter quality evaluation indicators, and perform multi-parameter and multi-objective collaborative optimization.

[0065] Table 1: Different frequency combinations for three-dimensional controlled source electromagnetic exploration;

[0066]

[0067] The present invention is aimed at the development needs of exploration geophysics, and solves the problem of difficult detection of target geological bodies from the perspective of optimizing the configuration of three-dimensional controlled source electromagnetic exploration data acquisition. The key points are reflected in the following two aspects:

[0068] (1) Propose an optimization design for the data acquisition parameters of 3D controlled source electromagnetic exploration, explicitly calculate the sensitivity matrix of the target model parameters, and improve the resolution of the observed data for the target geological body;

[0069] (2) A mathematical optimization model with the goal of minimizing the singular value Euclidean distance and the first-order difference is constructed, and a multi-objective non-dominated sorting genetic algorithm is used to simultaneously solve the minimum distance and the most stable estimation.

[0070] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0071] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for optimizing the configuration of controlled source electromagnetic exploration data acquisition parameters, characterized in that: The following steps are involved: Step 1: construct a three-dimensional geoelectric model; Step 2: Using the staggered grid finite difference method, the Helmholtz equation based on the secondary electric field is solved to complete the numerical calculation of the three-dimensional electromagnetic response of the controlled source electromagnetic exploration; Step 3: define the relative anomaly formula of electromagnetic response, compare the field value variation law when there is an abnormal body and when there is no abnormal body, and analyze the electromagnetic response characteristics obtained by different data acquisition parameters; Step 4: Gaussian random noise is added to the theoretical data, and the nonlinear conjugate gradient method is used to perform three-dimensional inversion, summarize the influence of different data acquisition parameter configurations on the detection effect of the target geological body, and analyze the main control parameters; Step 5: Display the calculated sensitivity matrix; the sensitivity matrix reflects the relationship between the physical properties of a certain point underground and the geophysical observation data; Step 6: Analyze the structural characteristics of the singular value spectrum of the sensitivity matrix of different data acquisition parameters, and combine the electromagnetic response characteristics and the three-dimensional inversion results to reveal that larger singular values ​​and flat singular value spectra have better resolution capabilities for model parameters; Step 7: Construct a quality evaluation function for data acquisition parameter configuration; Step 8: construct multiple objective functions according to the quality evaluation function, wherein objective function 1 is the Euclidean distance of the singular value spectrum of the sensitivity matrix, and objective function 2 is the first-order difference of the singular value spectrum of the sensitivity matrix; Step 9: Use a multi-objective non-dominated sorting genetic algorithm to search for a set of Pareto optimal solutions that can simultaneously minimize objective function 1 and objective function 2 under constraints; Step 10: According to the needs of the actual problem, extract the optimal transmission frequency, receiving and transmitting distance, and measurement point location data acquisition parameters from the Pareto optimal solution; Step 11, designing a three-dimensional geoelectric model of the target geological body with different parameters, comparing the three-dimensional inversion results of the data collected by the standard data acquisition parameters and the optimized data acquisition parameters, and proving the advantages of optimizing the data acquisition parameters; The step 7 is specifically: Define the sensitivity matrix J for all model parameters a , that is, the derivative of the observed data with respect to all model parameters; based on the theory that the ability of the observed data to resolve model parameters is related to the size of the sensitivity singular value and the flatness of the singular value spectrum; In order to quantify the size of the singular values ​​of the sensitivity matrix, the quality function 1 is defined as: Where D a =[ln(λ a 1),ln(λ a 2),...,ln(λ a ra )] T is the sensitivity matrix J a The singular value vector of , ra is the sensitivity matrix J a rank, is the sensitivity matrix J a The maximum singular value vector of , T represents transpose; QF1 represents the Euclidean distance of the singular values ​​of the sensitivity matrix. The smaller the QF1 value, the smaller the sensitivity matrix J a The larger the singular value of In order to quantify the flatness of the singular value spectrum of the sensitivity matrix, the quality function 2 is defined as: Where W is the first-order difference operator, that is, the smoothing factor, and the expression is: QF2 represents the sensitivity matrix J a The first-order difference of the singular value, the smaller the QF2 value, the smaller the difference between adjacent singular values, and the flatter the singular value spectrum. According to QF1 and QF2, the quality function QF of data acquisition parameter design is defined as QF=QF1+QF2 (4) It characterizes the quality of the optimized design of data acquisition parameters. The smaller the QF value, the better the data acquisition parameters are, and the greater the ability of the collected observation data to distinguish the model parameters.

2. The method for optimizing the configuration of controlled source electromagnetic exploration data acquisition parameters according to claim 1, characterized in that: The step 5 is specifically: the sensitivity matrix is ​​the derivative of the observed data with respect to the model parameters, and its expression is derived in accordance with the framework of Egbert & Kelbert. In order to improve the resolution of the observed data to the target geological body, the sensitivity matrix is ​​calculated using the derivative of the observed data with respect to the target body model parameters.

3. The method for optimizing and configuring controlled source electromagnetic exploration data acquisition parameters according to claim 1, characterized in that: The step 6 is specifically: using the singular value decomposition technique, the sensitivity matrix J, J = N × M, where M represents the number of model parameters and N represents the number of observation data, so J is expressed as: J = U r Λ r V r T , U r and V r are N×r and M×r eigenvector matrices respectively, Λ r is a string containing r singular values ​​λ1,λ2,...,λ r The diagonal matrix of , the singular values ​​are all greater than zero, and are arranged from large to small, r is the rank of the sensitivity matrix J, T represents the transpose; larger singular values ​​indicate disturbances in the model space, which will cause larger changes in the data space.

Images