Marchenko self-focusing theory interlayer multiple prediction method for focus point fitting wrinkled stratum

Abstract

According to the Marchenko self-focusing theory interlayer multiple prediction method with the focus point attached to the wrinkled stratum, the focus point is selected in the stratum with strong wave impedance underground, and a focus point layer penetrating through a strong wave impedance interface to the maximum degree is obtained; a direct wave between the underground focus point and the earth surface detector is obtained through simulation; taking the processed seismic record and the direct wave field as input, and adopting a Marchenko method to solve uplink and downlink Green functions between the underground focus point and the earth surface detector; a direct wave of the downlink Green function is cut off, convolution is carried out on the direct wave and the complete uplink Green function according to a specific spatial position, and primary coverage of interlayer multiples passing through a certain focus point is obtained; and the focus points are continuously replaced, multiple coverage of inter-layer multiple wave prediction results of different focus points is solved, integration is carried out, and a final inter-layer multiple wave prediction result is obtained. According to the method, on the premise that the calculation amount is not changed, more complete interlayer multiples can be predicted.

Description

Technical Field

[0001] The invention provides a method for predicting interlayer multiple waves based on the Marchenko self-focusing theory with a focal point fitted to a folded stratum, and belongs to the technical field of seismic exploration. Background Art

[0002] In recent years, the interlayer multiple wave suppression method based on Marchenko theory has gradually become a hot research topic. The Marchenko method is a method for reconstructing the Green's function at any position underground. It is based on the inverse scattering theory and only requires the input of unilaterally received seismic data and a macroscopic inaccurate model to reconstruct the upgoing and downgoing Green's functions relative to the surface detector at any position in the underground medium. Through the convolution and correlation of the upgoing and downgoing Green's functions, the effective prediction of interlayer multiple waves can be achieved, and the interlayer multiple waves can be effectively suppressed through the adaptive subtraction method.

[0003] The interlayer multiple wave prediction method based on Marchenko self-focusing theory uses the reflection wave records collected on the ground and the direct wave records between the underground virtual source point and the surface obtained under the macro model, and the method of selecting the focus point is often to simply select any depth underground to place the focus point horizontally. The upgoing and downgoing Green's function wave fields between the virtual source point and the surface are constructed through the Marchenko equation, and the Green's function downgoing wave field after removing the direct wave is convolved with the complete Green's function upgoing wave field to construct the interlayer multiple waves.

[0004] Conventional interlayer multiple wave prediction methods based on Marchenko self-focusing theory select the focus point on a horizontal line, without considering the folds and undulations of the underground strata. Although Marchenko self-focusing theory can obtain accurate measurements of the travel time and phase of interlayer multiple waves, the predicted interlayer multiple waves are highly correlated with the depth of the focus point selection. The deeper the focus point is, the larger the cut-off time window of the Green's function is, and the less effective information of the Green's function is.

[0005] When the strong wave impedance layers (high velocity and altitude layers) that generate interlayer multiple waves underground are irregular, the selected focusing points cannot all pass through the strong wave impedance strata. Some focusing points are below the strong wave impedance interface. The focusing points that are not in the strong wave impedance strata cannot construct the interlayer multiple waves inside the layer, which will cause the direct wave travel time from the focusing point to the surface detector to be too long, resulting in a large cut-off amount of the Green's function, which often results in the predicted interlayer multiple wave data volume being too small.

[0006] As the depth of the focus layer increases, the amount of wave field data of the predicted interlayer multiple waves gradually decreases. This is because as the depth of the focus layer increases, the travel time of the direct wave between the focus point and the surface detector gradually increases, and the interlayer multiple reflections inside the high wave impedance formation cannot be predicted, resulting in the gradual increase of the cut-off time window for solving the uplink and downlink Green's functions by the Marchenko method, which in turn leads to a gradual decrease in the effective information of the Green's function, and the amount of constructed interlayer multiple wave data is correspondingly reduced. Summary of the invention

[0007] The technical problem to be solved by the present invention is that as the depth of the focus point layer increases, the amount of predicted interlayer multiple wave data decreases. At the same time, when the underground strong wave impedance interface is not horizontal, the horizontally selected focus point layer cannot completely pass through the fold strong wave impedance interface, resulting in a large depth of some focus points, which is greater than the depth of the strong wave impedance stratum, and the interlayer multiple waves that are scattered multiple times in the strong wave impedance stratum cannot be effectively constructed. Therefore, it will also lead to a small number of predicted interlayer multiple waves.

[0008] The present invention proposes a method for predicting interlayer multiple waves based on the Marchenko self-focusing theory with a focus point close to a wrinkled stratum, comprising the following steps:

[0009] Step 1: On the premise of having sufficient geological knowledge of the underground velocity field, select several points inside the underground strong wave impedance stratum as focusing points, and use the Bezier curve smoothing method to obtain the focusing point layer that penetrates the strong wave impedance interface to the greatest extent.

[0010] Step 2: Through wave equation forward modeling or ray forward modeling, the direct wave between the underground focusing point and the surface detector is obtained.

[0011] Step 3: Take the de-noised, free surface multiple wave and ghost wave removed seismic records and the direct wave fields obtained in the previous step as input, and use the Marchenko method to solve the uplink and downlink Green's functions from the underground focusing point to the surface detector.

[0012] Step 4: Cut off the direct wave of the downward Green's function and convolve it with the complete upward Green's function at a specific spatial position to obtain a single coverage of the interlayer multiple waves passing through a certain focusing point.

[0013] Step 5: Continuously change the focus point, find the multiple coverage of the interlayer multiple wave prediction results at different focus points, and integrate them to obtain the final interlayer multiple wave prediction results.

[0014] The present invention, under the same theoretical framework of predicting interlayer multiple waves by Marchenko self-focusing theory, can predict more complete interlayer multiple waves without changing the amount of calculation by placing the focusing point in the strong wave impedance formation causing the interlayer multiple waves. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the focusing function;

[0016] Figure 2 is a schematic diagram of Green's function;

[0017] Figure 3 To predict the inter-layer multiple wave ray principle;

[0018] Figure 4 A schematic diagram of rays for constructing interlayer multiple waves in an embodiment;

[0019] Figure 5 The strong wave impedance interface of the embodiment is a velocity model of a wrinkled undulating layer;

[0020] Figure 6 The original earthquake record with the source position of (3000, 0) in the embodiment;

[0021] Figure 7 Schematic diagram of the layer where the focus point of the embodiment is located; the upper focus point is close to the folded stratum, and the depth of the lower focus point is 1600 meters;

[0022] Figure 8 The predicted interlayer multiple waves when the focus layer depth is 1600 meters in the embodiment;

[0023] Fig. 9 This is the predicted interlayer multiple waves when the focus layer of the embodiment is attached to the wrinkled formation. DETAILED DESCRIPTION

[0024] Marchenko self-focusing theory can reconstruct the precise Green's function from the underground virtual source point to the surface receiver, such as Figure 1 and Figure 2 , which is mathematically derived from the following system of equations:

[0025]

[0026] Modification point: G - (x i ,x0,t) represents the upward Green function, G + (x i ,x0,t) represents the downlink Green's function, R(x′0,x0,t) represents the reflection response after wavelet deconvolution, represents the upward focusing function, Represents the downlink focusing function.

[0027] The construction of a single reflection wave field passing through an underground virtual source point must contain its downstream direct wave field and a single upward reflection wave field. In other words, if the direct wave of the downstream Green's function is removed, or the single reflection wave of the upward Green's function is removed, it is impossible to construct a single wave. Only the interlayer multiple waves of the corresponding strata can be constructed. By subtracting the interlayer multiple waves from the original seismic data through adaptive matching filtering, the single reflection waves of each underground reflection layer can be obtained.

[0028] The Marchenko self-focusing theory predicts the interlayer multiple waves by convolving the downlink Green's function without the direct wave with the complete uplink Green's function to achieve the prediction of the interlayer multiple waves. Figure 3 , the interlayer multiple wave construction method based on Marchenko self-focusing theory is:

[0029]

[0030] G IM (x2,x1,t) means x1 is the earthquake source and x2 is the interlayer multiple waves received by the geophone. represents the downward Green's function after removing the direct wave, G - represents the complete upward Green's function, ρ(x) represents the density function, c(x) represents the wave velocity, S i It is the range interval for constructing the virtual source points of interlayer multiple waves.

[0031] The method for predicting interlayer multiple waves based on the Marchenko self-focusing theory with the focus point fitting the folded strata includes the following steps:

[0032] Step 1: On the premise of having sufficient geological knowledge of the underground velocity field, select several points inside the underground strong wave impedance stratum as focusing points, and use the Bezier curve smoothing method to obtain the focusing point layer that penetrates the strong wave impedance interface to the greatest extent.

[0033] Step 2: Through wave equation forward modeling or ray forward modeling, the direct wave between the underground focusing point and the surface detector is obtained.

[0034] Step 3: Take the de-noised, free surface multiple wave and ghost wave removed seismic records and the direct wave fields obtained in the previous step as input, and use the Marchenko method to solve the uplink and downlink Green's functions from the underground focusing point to the surface detector.

[0035] Step 4: Cut off the direct wave of the downward Green's function and convolve it with the complete upward Green's function at a specific spatial position to obtain a single coverage of the interlayer multiple waves passing through a certain focusing point.

[0036] Step 5: Continuously change the focus point, find the multiple coverage of the interlayer multiple wave prediction results at different focus points, and integrate them to obtain the final interlayer multiple wave prediction results.

[0037] In the interlayer multiple wave prediction method based on the Marchenko self-focusing theory, the ray path of the interlayer multiple wave must pass through the focusing point. Therefore, in this method, the deepest scattering point of the interlayer multiple wave ray path must be greater than the depth of the focusing point. When the depth of the focusing point is greater than the depth of a certain strong wave impedance interface, the interlayer multiple waves inside the strong wave impedance formation cannot be predicted. Figure 4 As shown, the focus point is on the third layer, which cannot construct the interlayer multiple waves inside the second layer.

[0038] The conventional Marchenko self-focusing theory simply selects a horizontal layer to place the focus point when selecting the focus point, without considering the contact relationship between the selected focus point and the underground stratum. When the selected focus point happens to be in a strong wave impedance stratum, the interlayer multiple reflections inside the stratum can be predicted. However, when the selected focus point is not inside the strong wave impedance stratum, the interlayer multiple waves formed by multiple scattering inside it cannot be constructed.

[0039] At the same time, when the focusing point is selected below the strong wave impedance interface, the direct wave from the focusing point to the surface detector is not the minimum arrival time of the multiple wave scattering wave related to the focusing point. Some interlayer multiple waves formed between shallow strata will be cut off by the larger direct wave time window, thus cutting off useful information.

[0040] The present invention proposes to place the focus point in a high wave impedance stratum, which can minimize the travel time of the direct wave from the focus point to the surface detector, so as to ensure that the cut-off time window is minimized when calculating the Green function, and correspondingly retain the most effective information. When performing convolution and integration to predict interlayer multiple waves, more complete interlayer multiple waves will be obtained. Under the premise of not increasing the amount of calculation for solving the Marchenko equation, the interlayer multiple reflections caused by the underground strong wave impedance interface can be predicted to the greatest extent, which is more conducive to the later adaptive matching filtering.

[0041] as follows Figure 5 The velocity model shown in the figure is 6,000 meters long and 3,000 meters deep. It is divided into 7 layers, of which the fourth layer is a high-velocity layer in the form of folds, which forms a strong wave impedance difference with the surrounding rock, thereby forming abundant interlayer multiple waves. Figure 6 As shown, this is the surface earthquake record when the spatial position of the earthquake source is (3000, 0).

[0042] Now, set the focus point layer at 1600 depth in the model and in the high-speed formation, such as Figure 7As shown in the figure, the shallowest red dotted line is the location of the focus point layer in the high-speed layer, and the focus point of this layer completely passes through the strong wave impedance layer. The depth of the deeper horizontal focus point layer is 1,600 meters, and the focus point of this layer partially passes through the strong wave impedance layer.

[0043] The interlayer multiple waves predicted by the two focus layers are calculated respectively. The interlayer multiple waves predicted when the focus layer depth is 1600 meters are as follows: Figure 8 As shown in Figure 2, the predicted interlayer multiple waves when the focus layer fits the folded stratum are as follows: Fig. 9 shown.

[0044] In such Figure 8 The interlayer multiple waves predicted by the horizontal focus point shown in the figure cannot predict the interlayer multiple waves inside the folded strata. Fig. 9 The data can be accurately predicted, which greatly improves the integrity of the multiple waves between layers in the whole data. Figure 8 , Fig. 9 It can be found that compared with the focus point placed in the horizontal layer, by placing the focus point in the strong wave impedance stratum that causes interlayer multiple waves, more complete interlayer multiple waves can be predicted without changing the amount of calculation.

[0045] The focus point of the present invention is modified from a simple horizontal selection to change with the structural morphology of the strong wave impedance stratum (using the Bezier curve interpolation method), so that the focus point fits the fold morphology, thereby making the predicted interlayer multiple wave data volume more complete.

[0046] Compared with other interlayer multiple wave suppression methods, the present invention has a very low dependence on the velocity model. The travel time errors of the uplink and downlink Green's functions caused by the erroneous velocity model will offset each other. Even when the model error is large, the interlayer multiple waves with the correct travel time can be predicted.

Claims

1. A method for predicting interlayer multiple waves based on the Marchenko self-focusing theory with the focus point fitting the folded strata, characterized in that: The following steps are involved: Step 1: On the premise of having sufficient geological knowledge of the underground velocity field, select several points inside the underground strong wave impedance stratum as focusing points, and use the Bezier curve smoothing method to obtain the focusing point layer that passes through the strong wave impedance interface to the greatest extent; Step 2: Obtain the direct wave from the underground focusing point to the surface detector through wave equation forward modeling or ray forward modeling; Step 3: The deconvolution process of the de-noised, free surface multiple wave and ghost wave removed seismic records and the direct wave fields obtained in the previous step are used as input, and the Marchenko method is used to solve the uplink and downlink Green's functions from the underground focus point to the surface detector. Step 4: Cut off the direct wave of the downlink Green's function and convolve it with the complete uplink Green's function at a specific spatial position to obtain a single coverage of the interlayer multiple waves passing through a certain focal point; Step 5: Continuously change the focusing point, find the multiple coverage of the inter-layer multiple wave prediction results at different focusing points, and integrate them to obtain the final inter-layer multiple wave prediction results.

Images