Seismic inversion low-frequency model establishment method and device
Through the transformation of seismic data and neural network modeling, the problem of poor seismic inversion modeling accuracy under drilling conditions is solved, and more accurate reservoir prediction and formation change reflection are achieved, which is suitable for deep and offshore oil and gas exploration.
Patent Information
- Application Number
- CN202310575174.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-19
- Publication Date
- 2024-11-22
AI Technical Summary
In the absence of drilling, seismic inversion modeling is difficult to accurately predict reservoir distribution and hydrocarbon detection. The existing technology often uses constant modeling to lead to poor accuracy and cannot effectively reflect the longitudinal and lateral changes of underground media.
By obtaining seismic data and velocity spectrum, converting it into layer velocity sequence, extracting seismic attributes based on seismic phases, establishing a dummy well and determining the logging curve, using neural network algorithm to establish a low-frequency model, fusing the velocity spectrum and seismic attribute information, and building a geology-compliant geology A regular model.
It improves the seismic inversion modeling accuracy under well-free conditions, can reflect formation changes more accurately, enhances the accuracy of reservoir prediction, and is suitable for deep and offshore oil and gas exploration.
Smart Images

Figure CN119001838A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method and a device for establishing a seismic inversion low-frequency model. Background Art
[0002] Seismic inversion is the core technology for predicting reservoir distribution and hydrocarbon detection in oil and gas exploration. In the inversion process, how to establish an accurate low-frequency model is the key step to achieve ideal prediction results. At present, seismic inversion modeling usually adopts the method of drilling data interpolation. First, the velocity and other information at the well point are obtained according to the drilling data, and then interpolation is performed according to certain mathematical operations. However, in many cases, there are no wells in the study area where seismic inversion is required, such as deep oil and gas exploration, offshore oil and gas exploration, etc., which brings difficulties to seismic inversion modeling. Summary of the invention
[0003] In order to enrich the types of seismic inversion low-frequency model construction methods in the prior art and improve the accuracy of modeling and reservoir prediction, the present invention provides a seismic inversion low-frequency model establishment method and device. The technical solution proposed by the present invention is as follows:
[0004] As a first aspect of the present invention, a method for establishing a seismic inversion low-frequency model is provided, comprising:
[0005] Obtain the seismic data volume and velocity spectrum of the study area and the structural horizons of each target layer in the study area;
[0006] Converting the velocity spectrum to obtain a layer velocity sequence of each target layer;
[0007] Extracting various seismic attributes of the target layer based on the structural horizon of the target layer, and predicting the seismic phase corresponding to the target layer according to the various seismic attributes;
[0008] In each type of seismic phase of the target layer, at least one position within the range of the seismic phase is selected to establish a pseudo well, the interval velocity sequence of the pseudo well is obtained from the interval velocity sequence of each target layer, and a well logging curve is determined according to the interval velocity sequence of the pseudo well;
[0009] In various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve.
[0010] In some embodiments, the velocity spectrum includes a stacked velocity sequence; and converting the velocity spectrum to obtain a layer velocity sequence of each target layer includes:
[0011] Input velocity spectrum;
[0012] The stacking velocity sequence in the velocity spectrum is converted into the layer velocity sequence by the following formula:
[0013]
[0014] In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v a,n is the stacking velocity of the nth layer, v a,n-1 is the stacking velocity of the n-1th layer.
[0015] In some embodiments, the well logging curve includes a density curve; in each type of seismic phase of the target layer, selecting at least one position within the range of the seismic phase to establish a pseudo well, obtaining the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determining the well logging curve according to the interval velocity sequence of the pseudo well, comprising:
[0016] For each type of seismic phase of the target layer, selecting at least one position within the range of the seismic phase to establish a pseudo well;
[0017] Acquire the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, determine the density sequence according to the interval velocity sequence of the pseudo well by the following formula, and obtain a density curve;
[0018]
[0019] Where D x,y is the density sequence at (x, y), C1 is a constant, V x,y is the layer velocity sequence at (x,y).
[0020] In some embodiments, the logging curve includes a wave impedance curve; and determining the logging curve according to the interval velocity sequence of the pseudo well further includes:
[0021] According to the interval velocity sequence of the pseudo well and the density sequence, the wave impedance sequence is determined by the following formula to obtain the wave impedance curve:
[0022] P x,y =V x,y *D x,y ;
[0023] Where P x,y is the wave impedance sequence at (x, y), D x,y is the density sequence at (x,y), V x,y is the layer velocity sequence at (x,y).
[0024] In some embodiments, the velocity spectrum includes a stacked velocity sequence; and converting the velocity spectrum to obtain a layer velocity sequence of each target layer includes:
[0025] The stacking velocity sequence is converted into a root mean square velocity sequence by the following formula:
[0026]
[0027] In the formula, v R,n is the root mean square velocity of the nth layer, v a,n is the stacking velocity of the nth layer, is the inclination of the nth stratum interface;
[0028] The root mean square velocity series is converted into a layer velocity series by the following formula:
[0029]
[0030] In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v R,n is the root mean square velocity of the nth layer, v R,n-1 is the RMS velocity of the n-1th layer.
[0031] In some embodiments, extracting various types of seismic attributes of the target layer based on the structural horizon of the target layer, and predicting the seismic phase corresponding to the target layer according to the various types of seismic attributes, includes:
[0032] A time window is set according to the structural horizon of the target layer, and various seismic attributes of the target layer are extracted based on the set time window; the seismic attributes include amplitude, frequency and phase;
[0033] According to the various earthquake attributes, a preset clustering method is used to predict the corresponding earthquake phases.
[0034] In some embodiments, predicting the corresponding seismic phases using a preset clustering method according to the various types of seismic attributes includes:
[0035] The inter-class distances between various types of seismic attributes are calculated by the cross-correlation algorithm, and the two types of seismic attributes with the closest inter-class distances are aggregated and merged into a new attribute.
[0036] Calculating the inter-class distance between the new attribute and the remaining types of seismic attributes, repeating the aggregation process until all types of seismic attributes are divided, and obtaining multiple seismic attribute sets;
[0037] The multiple seismic attribute sets are clustered using a K-means clustering algorithm to obtain the seismic phase of the target layer.
[0038] In some embodiments, in various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve, including:
[0039] In various seismic phases of the target layer, the seismic data and the well logging curve corresponding to the pseudo well are used as training samples, and a nonlinear mapping relationship between the seismic data and the well logging curve is established through a neural network algorithm;
[0040] The nonlinear mapping relationship corresponding to each earthquake is applied to the entire seismic data volume to obtain a low-frequency model volume.
[0041] In some embodiments, the well logging curve includes a density curve or a wave impedance curve: in various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve, including:
[0042] In various seismic phases of the target layer, the seismic data and density curve corresponding to the pseudo well are used as training samples, and a first nonlinear mapping relationship between the seismic data and the density curve is established through a neural network algorithm;
[0043] Applying the first nonlinear mapping relationship corresponding to each earthquake to the entire seismic data volume to obtain a velocity model volume;
[0044] or,
[0045] In various seismic phases of the target layer, the seismic data and wave impedance curve corresponding to the pseudo well are used as training samples, and a second nonlinear mapping relationship between the seismic data and the wave impedance curve is established through a neural network algorithm;
[0046] The second nonlinear mapping relationship corresponding to each earthquake is applied to the entire seismic data volume to obtain a wave impedance model volume.
[0047] As a second aspect of the present invention, there is provided a seismic inversion low-frequency model establishment device, comprising:
[0048] An acquisition module is used to acquire the seismic data volume and velocity spectrum of the study area and the structural horizons of each target layer in the study area;
[0049] A layer velocity sequence determination module, used for converting the velocity spectrum to obtain a layer velocity sequence of each target layer;
[0050] A seismic phase determination module, used to extract various seismic attributes of the target layer based on the structural horizon of the target layer, and predict the seismic phase corresponding to the target layer according to the various seismic attributes;
[0051] A well logging curve determination module; used for selecting at least one position within the range of the seismic phase in each type of seismic phase of the target layer to establish a pseudo well, obtaining the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determining the well logging curve according to the interval velocity sequence of the pseudo well;
[0052] The model building module is used to obtain seismic data corresponding to the pseudo well from the seismic data volume in various seismic phases of the target layer, and to establish a low-frequency model according to the seismic data and the well logging curve.
[0053] As a third aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method for establishing a seismic inversion low-frequency model as described in the first aspect is implemented.
[0054] As a fourth aspect of the present invention, there is provided an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0055] Memory, used to store computer programs;
[0056] The processor is used to implement the method for establishing a seismic inversion low-frequency model as described in the first aspect when executing the program stored in the memory.
[0057] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:
[0058] The method for establishing a seismic inversion low-frequency model provided by the present invention obtains a seismic data body, a velocity spectrum and a target layer structural layer; converts the velocity spectrum into a layer velocity sequence; extracts various seismic attributes of the target layer based on the target layer structural layer, and predicts corresponding seismic phases according to the various seismic attributes; the target layer structural layer is a layer with significantly different depths and relative activity in a specific crust deformation process in an orogenic belt, and the seismic phase information predicted by the multi-attribute extraction based on the target layer structural layer can reflect the vertical and horizontal changes of the stratum and conform to the actual situation of the underground medium; by fusing the velocity spectrum information and the seismic phase information corresponding to the prediction of various seismic attributes into the modeling process, the established model is more in line with geological laws, and the technical difficulty of poor precision of the previous constant modeling is overcome, thereby improving the accuracy of modeling and reservoir prediction, which has obvious practical significance for oil and gas exploration. Compared with the prior art method of seismic inversion modeling without wells that uses drilling data interpolation, the present invention extracts seismic attributes and predicts seismic phases based on the acquired seismic data, and then converts the velocity spectrum information into pseudo-well data, i.e., the logging curve of the pseudo-well, to establish an inversion low-frequency model, so that the established model is more in line with geological laws, thereby improving the accuracy of modeling and reservoir prediction, and is suitable for seismic inversion modeling without wells.
[0059] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.
[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0062] Figure 1 A schematic diagram of a flow chart of a method for establishing a low-frequency model for seismic inversion provided by an embodiment of the present invention;
[0063] Figure 2 It is a schematic diagram of a process of converting a velocity spectrum into a layer velocity sequence according to an embodiment of the present invention;
[0064] Figure 3 A schematic diagram of a process for predicting earthquake phases according to an embodiment of the present invention;
[0065] Figure 4 A schematic diagram of a process of establishing pseudo wells in various seismic phases and determining well logging curves according to an embodiment of the present invention;
[0066] Figure 5 A schematic diagram of a process of establishing an inversion low-frequency model according to seismic phase prediction results and well logging curves of a modeling well according to an embodiment of the present invention;
[0067] Figure 6 A schematic diagram of the structure of a seismic inversion low-frequency model establishment device provided in an embodiment of the present invention;
[0068] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;
[0069] Figure 8a Schematic diagram of the low-frequency model established using conventional methods;
[0070] Figure 8b A schematic diagram of a low-frequency model established by using the low-frequency model establishment method provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0071] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0072] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.
[0073] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.
[0074] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0075] The inventor found in his work that, at present, seismic inversion modeling usually adopts the method of drilling data interpolation. First, the velocity and other information at the well point are obtained according to the drilling data, and then interpolation is performed according to certain mathematical operations. However, in many cases, there is no drilling in the study area where seismic inversion is required, such as deep oil and gas exploration, offshore oil and gas exploration, etc., which brings difficulties to seismic inversion modeling. In order to solve this problem, the inversion model is usually set to a constant. The model established by this method does not change in the vertical and horizontal directions, which is quite different from the actual situation of the underground medium, resulting in the deterioration of the accuracy of seismic inversion and cannot meet the expectations of the inventor. After further research and development, the inventor made the present invention. The present invention relates to the field of petroleum geophysical exploration technology, and in particular to the technology of using seismic data to predict oil and gas reservoirs. Specifically, it is a seismic inversion modeling method under well-free conditions.
[0076] Embodiment 1
[0077] The embodiment of the present invention provides a method for establishing a low-frequency model for seismic inversion, referring to Figure 1 As shown, including:
[0078] S101, obtaining the seismic data volume and velocity spectrum of the study area and each target layer structural horizon of the study area;
[0079] The above-mentioned seismic data body is a three-dimensional post-stack seismic data body that has been processed by surface acquisition and subsequent pre-stack time migration. For the method of pre-stack time migration processing, those skilled in the art can refer to the specific description in the prior art, which will not be repeated here. The above-mentioned target layer structural horizon is generally the time horizon of the top and bottom of the target layer interpreted on the seismic data, and can also be the time horizon of the top or bottom of the target layer.
[0080] S102, converting the velocity spectrum to obtain a layer velocity sequence of each target layer;
[0081] The above velocity spectrum is obtained by flattening the common imaging point gather data through velocity analysis during seismic data processing, and is usually the stacking velocity corresponding to different time depths in the study area.
[0082] S103, extracting various types of seismic attributes of the target layer based on the structural horizon of the target layer, and predicting the seismic phase corresponding to the target layer according to the various types of seismic attributes;
[0083] The seismic phase of the target layer is predicted by integrating various seismic attributes to obtain the seismic phase prediction result, that is, the seismic phase type included in the target layer, and then modeling is carried out based on the seismic phase prediction result.
[0084] S104, in each type of seismic phase of the target layer, selecting at least one position within the range of the seismic phase to establish a pseudo well, obtaining the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determining a well logging curve according to the interval velocity sequence of the pseudo well;
[0085] For each type of seismic phase, a pseudo well is established at any position within the range of the seismic phase, or multiple pseudo wells are established at any multiple positions. The established pseudo wells are used as modeling wells, and a low-frequency model is established based on the seismic data and logging curves of the modeling wells.
[0086] S105 . In various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve.
[0087] After obtaining the seismic phase prediction result of the target layer, for each type of predicted seismic phase, the seismic data corresponding to each pseudo-well in this type of seismic phase is obtained from the seismic data body obtained in step S101; the layer velocity sequence of each pseudo-well in this type of seismic phase is obtained from the layer velocity sequence converted in the above step S102; and a low-frequency model is established using the seismic data and logging curves of the pseudo-wells.
[0088] Reference Figure 8a and Figure 8b As shown in the figure, they are respectively the conventional method under the condition of no well ( Figure 8a ) and the method of the present invention ( Figure 8b ) The low-frequency model established by the conventional method has no changes in the vertical and horizontal directions, which is quite different from the actual situation of the underground medium. The low-frequency model established by the method of the present invention can reflect the changes in the stratum in the vertical and horizontal directions, which is in line with the actual situation of the underground medium. The established model is more in line with geological laws.
[0089] The method for establishing a seismic inversion low-frequency model provided by an embodiment of the present invention is a seismic inversion modeling method under well-free conditions, which obtains a seismic data body, a velocity spectrum and a target layer structural horizon; converts the velocity spectrum into a layer velocity sequence; extracts various seismic attributes of the target layer based on the target layer structural horizon, and predicts the corresponding seismic phase according to the various seismic attributes; the target layer structural horizon is a horizon with significantly different depths and relative activity in a specific crust deformation process in an orogenic belt, and the seismic phase information predicted by the multi-attribute extraction based on the target layer structural horizon can reflect the changes in the stratum in the vertical and horizontal directions, which conforms to the actual situation of the underground medium; by integrating the velocity spectrum information and the seismic phase information corresponding to the prediction of various seismic attributes into the modeling process, the established model is more in line with geological laws, and the technical difficulty of poor accuracy of modeling using constants in the past is overcome, thereby improving the accuracy of modeling and reservoir prediction, which has obvious practical significance for oil and gas exploration.
[0090] Compared with the method of using drilling data interpolation in seismic inversion modeling in the prior art, the present invention converts the velocity spectrum into the layer velocity sequence of each target layer, and establishes a pseudo-well in each type of seismic phase of the target layer, obtains the layer velocity sequence of the pseudo-well from the layer velocity sequence of the target layer, and determines the corresponding logging curve according to the layer velocity sequence of the pseudo-well, so as to realize the conversion of velocity spectrum information into pseudo-well data, and establish an inversion low-frequency model based on the determined pseudo-well data, thereby improving the accuracy of the low-frequency model under the condition of no well. Moreover, by selecting at least one position within the range of each type of seismic phase to establish a pseudo-well, and determining the logging curve according to the layer velocity sequence of the pseudo-well; establishing a low-frequency model according to the seismic data and logging curve corresponding to the pseudo-well, and establishing different low-frequency models for different seismic phases, the accuracy of modeling and reservoir prediction is further improved.
[0091] The velocity spectrum in step S102 is usually in the format of (L, T, t, V a ), where L is the seismic line number, T is the channel number of each seismic line, t is the time depth, V a is the underground medium stacking velocity at point (L, T, t). In a specific embodiment, the velocity spectrum includes a stacking velocity sequence; the velocity spectrum is converted to obtain the layer velocity sequence of each target layer, referring to Figure 2 As shown, including:
[0092] S1021, input velocity spectrum;
[0093] S1022. Convert the stacking velocity sequence in the velocity spectrum into a layer velocity sequence using the following formula:
[0094]
[0095] In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v a,n is the stacking velocity of the nth layer, v a,n-1 is the stacking velocity of the n-1th layer.
[0096] In the above step S102, the stacking velocity sequence can be directly converted into a layer velocity sequence, or the stacking velocity sequence can be converted into a root mean square velocity sequence, and then converted into a layer velocity sequence. In a specific embodiment, the velocity spectrum includes a stacking velocity sequence; the above conversion of the velocity spectrum to obtain the layer velocity sequence of each target layer includes:
[0097] Convert the stacking velocity series into a root mean square velocity series:
[0098]
[0099] In the formula, v R,n is the root mean square velocity of the nth layer, v a,n is the stacking velocity of the nth layer, is the inclination of the nth stratum interface;
[0100] Convert the RMS velocity series into layer velocity series:
[0101]
[0102] In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v R,n is the root mean square velocity of the nth layer, v R,n-1 is the RMS velocity of the n-1th layer.
[0103] In a specific embodiment, the various types of seismic attributes in the above step S103 include attributes such as amplitude, frequency and phase. The above-mentioned extraction of various types of seismic attributes of the target layer based on the structural horizon of the target layer, and prediction of the seismic phase corresponding to the target layer according to the various types of seismic attributes, refer to Figure 3 As shown, including:
[0104] S1031, setting a time window according to the structural horizon of the target layer, and extracting various seismic attributes of the target layer based on the set time window; the seismic attributes include amplitude, frequency and phase;
[0105] A time window is set according to the target layer structure layer, and the time window is [H1, H2], where H1 is the top structure layer of the target layer and H2 is the bottom structure layer of the target layer. Based on the set time window, the amplitude, frequency, phase and other attributes of the target layer are obtained.
[0106] S1032. According to the various earthquake attributes, a preset clustering method is used to predict corresponding earthquake phases.
[0107] The various seismic attributes in the target layer are clustered using a preset clustering method, and the layers with similar seismic attributes are classified into one category, thereby dividing the target layer into multiple categories, each with different seismic phases, to obtain seismic phase prediction results. Those skilled in the art can also use the waveform clustering method in the prior art to cluster the extracted seismic attributes to achieve seismic phase prediction.
[0108] The basic principle of using the preset clustering method to predict earthquake phases in the above step S1032 is: there are N attribute samples, each of which is a class, and the inter-class distance is calculated by the cross-correlation algorithm. The classes with the closest distance are merged into a new class, which is called a clustering. At this time, there are N-1 classes in the sample set. The inter-class distances of these N-1 classes are recalculated, and another clustering is performed. Repeat the above steps until all attribute samples are classified into several classes. K attribute samples are selected from several classes as the initial clustering centers, and the distance between each attribute sample and each clustering center is calculated. Each attribute sample is assigned to the clustering center closest to it. The clustering center and the attribute sample assigned to it represent an earthquake phase. In a specific embodiment, the above-mentioned prediction of the corresponding earthquake phase based on the various types of earthquake attributes using the preset clustering method includes:
[0109] S10321, calculating the inter-class distances between various types of earthquake attributes by using a cross-correlation algorithm, and aggregating and merging the two types of earthquake attributes with the closest inter-class distances into a new attribute;
[0110] S10322, calculating the inter-class distance between the new attribute and the remaining types of earthquake attributes, repeating the aggregation process to complete the division, and obtaining multiple earthquake attribute sets;
[0111] By calculating the inter-class distances between various types of seismic attributes, the attributes are divided into several classes, and multiple seismic attribute sets are obtained, so that the data similarity within the same class is maximized as much as possible, and the data difference in different classes is maximized as much as possible.
[0112] S10323. Use a K-means clustering algorithm to cluster the multiple seismic attribute sets to obtain the seismic phase of the target layer.
[0113] First, K types of seismic phases are set in the study area, and then K seismic attributes are randomly selected from the multiple attribute sets obtained by division as the initial clustering centers. The distance between each seismic attribute and each clustering center is calculated, and each seismic attribute is assigned to the clustering center closest to it. The clustering center and the seismic attributes assigned to the clustering center represent a seismic phase.
[0114] In a specific embodiment, the logging curve in the above step S104 may be a density curve; in each type of seismic phase of the target layer, at least one position within the range of the seismic phase is selected to establish a pseudo well, the interval velocity sequence of the pseudo well is obtained from the interval velocity sequence of each target layer, and the logging curve is determined according to the interval velocity sequence of the pseudo well, referring to Figure 4 As shown, including:
[0115] S1041. For each type of seismic phase of the target layer, select at least one position within the range of the seismic phase to establish a pseudo well;
[0116] In any seismic phase A, select a position (x, y) within its range, and obtain the layer velocity sequence V at that position in step S102 x,y (t), t is the two-way travel time of the seismic wave.
[0117] S1042, acquiring the interval velocity sequence of the pseudo well from the interval velocity sequences of the respective target layers, determining the density sequence according to the interval velocity sequence of the pseudo well by the following formula, and obtaining a density curve;
[0118]
[0119] Where D x,y is the density sequence at (x, y), C1 is a constant, V x,y is the layer velocity sequence at (x,y). Where C1 is 0.31.
[0120] In a specific embodiment, the logging curve in the above step S104 may be a wave impedance curve; and determining the logging curve according to the interval velocity sequence of the pseudo well further includes:
[0121] S1043, determining the wave impedance sequence according to the interval velocity sequence of the pseudo well and the density sequence by the following formula to obtain a wave impedance curve:
[0122] P x,y =V x,y *D x,y ;
[0123] Where P x,y is the wave impedance sequence at (x, y), D x,y is the density sequence at (x,y), V x,y is the layer velocity sequence at (x,y).
[0124] In a specific embodiment, in the above step S105, a neural network algorithm can be used to establish a nonlinear mapping relationship between seismic data and well logging curves; in the above-mentioned various seismic phases of the target layer, the seismic data corresponding to the pseudo well is obtained from the seismic data body, and a low-frequency model is established according to the seismic data and the well logging curve, and a reference is made to the above-mentioned Figure 5 As shown, including:
[0125] S1051, in various seismic phases of the target layer, using the seismic data and logging curves corresponding to the pseudo well as training samples, and establishing a nonlinear mapping relationship between the seismic data and the logging curves through a neural network algorithm;
[0126] In this embodiment, an existing neural network model is used for training. The input of the model is seismic data, and the output is a well logging curve. The neural network model can be a BP neural network model. The specific training process is:
[0127] Initialize the model parameters of the neural network model;
[0128] Inputting the seismic data corresponding to the pseudo well into the neural network model to obtain the model output result, comparing the model output result with the well logging curve corresponding to the seismic data, and determining the corresponding loss function; optimizing the model parameters of the neural network model through the back propagation algorithm based on the loss function to obtain an updated network model;
[0129] Then, the seismic data corresponding to the pseudo well is input into the updated network model to obtain the model output result, and the model output result is compared with the well logging curve corresponding to the seismic data to determine the corresponding loss function; based on the loss function, the model parameters of the updated neural network model are optimized by the back propagation algorithm to obtain the updated network model again;
[0130] Repeat the above process of updating model parameters until the loss function reaches the preset requirement, that is, the seismic data and the well logging curve reach the maximum correlation, and obtain a trained neural network model. Based on the trained neural network model, a nonlinear mapping relationship between seismic data and well logging curves can be obtained.
[0131] S1052: Apply the nonlinear mapping relationship corresponding to each earthquake to the entire earthquake data volume to obtain a low-frequency model volume.
[0132] The nonlinear mapping relationship corresponding to each earthquake is applied to the seismic data corresponding to the earthquake of this type to obtain the corresponding low-frequency model, thereby obtaining a low-frequency model body.
[0133] The inversion low-frequency model established in the embodiment of the present invention may be a velocity model, a density model, a wave impedance model, etc. The type of the model is determined according to the well logging curve of the pseudo well, i.e., the modeling well. In a specific embodiment, the well logging curve includes a density curve or a wave impedance curve: in the various seismic phases of the target layer as described in the above step S105, the seismic data corresponding to the pseudo well is obtained from the seismic data body, and a low-frequency model is established according to the seismic data and the well logging curve, including:
[0134] S201, in various seismic phases of the target layer, using the seismic data and density curve corresponding to the pseudo well as training samples, and establishing a first nonlinear mapping relationship between the seismic data and the density curve through a neural network algorithm;
[0135] S202, applying the first nonlinear mapping relationship corresponding to each earthquake to the entire seismic data volume to obtain a velocity model volume;
[0136] or,
[0137] S301, in various seismic phases of the target layer, using the seismic data and wave impedance curve corresponding to the pseudo well as training samples, and establishing a second nonlinear mapping relationship between the seismic data and the wave impedance curve through a neural network algorithm;
[0138] S302: Apply the second nonlinear mapping relationship corresponding to each earthquake to the entire seismic data volume to obtain a wave impedance model volume.
[0139] The process of establishing the first nonlinear mapping relationship between seismic data and density curve, and the process of establishing the second nonlinear mapping relationship between seismic data and wave impedance curve are similar to the process of step 1051 above. Please refer to the specific description of S1051 and will not be repeated here.
[0140] When establishing the velocity model, the interval velocity sequence of the established pseudo-well is directly obtained from the interval velocity sequence of the target layer obtained in the above step S102, and the velocity model is established using the seismic data and interval velocity sequence of the pseudo-well.
[0141] Embodiment 2
[0142] The embodiment of the present invention provides a seismic inversion low-frequency model establishment device, referring to Figure 6 As shown, including:
[0143] Acquisition module 401, used to acquire seismic data volume and velocity spectrum of the study area and each target layer structural horizon of the study area;
[0144] A layer velocity sequence determination module 402 is used to convert the velocity spectrum to obtain a layer velocity sequence of each target layer;
[0145] A seismic phase determination module 403 is used to extract various seismic attributes of the target layer based on the structural horizon of the target layer, and predict the seismic phase corresponding to the target layer according to the various seismic attributes;
[0146] The well logging curve determination module 404 is used to select at least one position within the range of the seismic phase in each type of seismic phase of the target layer to establish a pseudo well, obtain the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determine the well logging curve according to the interval velocity sequence of the pseudo well;
[0147] The model building module 405 is used to obtain the seismic data corresponding to the pseudo well from the seismic data volume in various seismic phases of the target layer, and to establish a low-frequency model according to the seismic data and the well logging curve.
[0148] The implementation principle and technical effect of the seismic inversion low-frequency model establishment device provided in the embodiment of the present invention are similar to those of any of the aforementioned method embodiments, and will not be described in detail here.
[0149] Embodiment 3
[0150] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for establishing a seismic inversion low-frequency model as described in any of the aforementioned method embodiments.
[0151] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.
[0152] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus, or device.
[0153] Embodiment 4
[0154] An embodiment of the present invention provides an electronic device, referring to Figure 7 As shown, it includes a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112 and the memory 113 communicate with each other through the communication bus 114.
[0155] Memory 113, used for storing computer programs;
[0156] The processor 111 is used to implement the method for establishing a seismic inversion low-frequency model as described in any of the above method embodiments when executing the program stored in the memory 113.
[0157] The implementation principle and technical effect of the electronic device provided by the embodiment of the present invention are similar to those of any of the aforementioned method embodiments, and will not be repeated here.
[0158] The above-mentioned memory 113 can be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk or a ROM. The memory 113 has a storage space for program codes for executing any method steps in the above-mentioned method. For example, the storage space for program codes may include various program codes for implementing various steps in the above method respectively. These program codes can be read from or written into one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks. Such computer program products are usually portable or fixed storage units. The storage unit may have a storage segment or storage space arranged similarly to the memory 113 in the above-mentioned electronic device. The program code can be compressed, for example, in an appropriate form. Generally, the storage unit includes a program for executing the method steps according to an embodiment of the present invention, that is, a code that can be read by, for example, a processor 111, and when these codes are run by an electronic device, the electronic device performs the various steps in the method described above.
[0159] In this document, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. The orientation or position relationship indicated by the terms "upper", "lower", etc. is based on the orientation or position relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as a limitation on the present invention.
[0160] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict. The present invention is not limited to any single aspect, any single embodiment, or any combination and / or permutation of these aspects and / or embodiments. Each aspect and / or embodiment of the present invention can be used alone, or in combination with one or more other aspects and / or other embodiments.
[0161] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims
1. A method for establishing a low-frequency model for seismic inversion, characterized in that: include: Obtain the seismic data volume and velocity spectrum of the study area and the structural horizons of each target layer in the study area; Converting the velocity spectrum to obtain a layer velocity sequence of each target layer; Extracting various seismic attributes of the target layer based on the structural horizon of the target layer, and predicting the seismic phase corresponding to the target layer according to the various seismic attributes; In each type of seismic phase of the target layer, at least one position within the range of the seismic phase is selected to establish a pseudo well, the interval velocity sequence of the pseudo well is obtained from the interval velocity sequence of each target layer, and a well logging curve is determined according to the interval velocity sequence of the pseudo well; In various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve.
2. The method for establishing a low-frequency model for seismic inversion according to claim 1, characterized in that: The velocity spectrum includes a stacked velocity sequence; and converting the velocity spectrum to obtain a layer velocity sequence of each target layer includes: Input velocity spectrum; The stacking velocity sequence in the velocity spectrum is converted into the layer velocity sequence by the following formula: In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v a,n is the stacking velocity of the nth layer, v a,n-1 is the stacking velocity of the n-1th layer.
3. The method for establishing a low-frequency model for seismic inversion according to claim 1, characterized in that: The well logging curve includes a density curve; in each type of seismic phase of the target layer, selecting at least one position within the range of the seismic phase to establish a pseudo well, obtaining the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determining the well logging curve according to the interval velocity sequence of the pseudo well, including: For each type of seismic phase of the target layer, selecting at least one position within the range of the seismic phase to establish a pseudo well; Acquire the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, determine the density sequence according to the interval velocity sequence of the pseudo well by the following formula, and obtain a density curve; Where D x,y is the density sequence at (x, y), C1 is a constant, V x,y is the layer velocity sequence at (x,y).
4. The method for establishing a low-frequency model for seismic inversion according to claim 3, characterized in that: The logging curve includes a wave impedance curve; the logging curve is determined according to the interval velocity sequence of the pseudo well, and further includes: According to the interval velocity sequence of the pseudo well and the density sequence, the wave impedance sequence is determined by the following formula to obtain the wave impedance curve: P x,y =V x,y ·D x,y ; Where P x,y is the wave impedance sequence at (x, y), D x,y is the density sequence at (x,y), V x,y is the layer velocity sequence at (x,y).
5. The method for establishing a low-frequency model for seismic inversion according to claim 1, characterized in that: The velocity spectrum includes a stacked velocity sequence; and converting the velocity spectrum to obtain a layer velocity sequence of each target layer includes: The stacking velocity sequence is converted into a root mean square velocity sequence by the following formula: In the formula, v R,n is the root mean square velocity of the nth layer, v a,n is the stacking velocity of the nth layer, is the inclination of the nth stratum interface; The root mean square velocity series is converted into a layer velocity series by the following formula: In the formula, v n is the layer velocity of the nth layer, t 0.n is the vertical propagation time of the seismic wave in the nth layer, t 0.n-1 is the vertical propagation time of the seismic wave in the n-1th layer, v R,n is the root mean square velocity of the nth layer, v R,n-1 is the RMS velocity of the n-1th layer.
6. The method for establishing a low-frequency model for seismic inversion according to claim 1, characterized in that: The extracting various seismic attributes of the target layer based on the structural horizon of the target layer, and predicting the seismic phase corresponding to the target layer according to the various seismic attributes, includes: A time window is set according to the structural horizon of the target layer, and various seismic attributes of the target layer are extracted based on the set time window; the seismic attributes include amplitude, frequency and phase; According to the various earthquake attributes, a preset clustering method is used to predict the corresponding earthquake phases.
7. The method for establishing a low-frequency model for seismic inversion according to claim 6, characterized in that: The method of predicting corresponding seismic phases according to the various seismic attributes by using a preset clustering method includes: The inter-class distances between various types of seismic attributes are calculated by the cross-correlation algorithm, and the two types of seismic attributes with the closest inter-class distances are aggregated and merged into a new attribute. Calculating the inter-class distance between the new attribute and the remaining seismic attributes, repeating the aggregation process until all seismic attributes are divided, and obtaining multiple seismic attribute sets; The K-means clustering algorithm is used to cluster the multiple seismic attribute sets to obtain the seismic phase of the target layer.
8. The method for establishing a low-frequency model for seismic inversion according to claim 1, characterized in that: In various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data volume, and a low-frequency model is established according to the seismic data and the well logging curve, including: In various seismic phases of the target layer, the seismic data and the well logging curve corresponding to the pseudo well are used as training samples, and a nonlinear mapping relationship between the seismic data and the well logging curve is established through a neural network algorithm; The nonlinear mapping relationship corresponding to each earthquake is applied to the entire seismic data volume to obtain a low-frequency model volume.
9. The method for establishing a low-frequency model for seismic inversion according to claim 8, characterized in that: The logging curve includes a density curve or a wave impedance curve: in various seismic phases of the target layer, seismic data corresponding to the pseudo well is obtained from the seismic data body, and a low-frequency model is established according to the seismic data and the logging curve, including: In various seismic phases of the target layer, the seismic data and density curve corresponding to the pseudo well are used as training samples, and a first nonlinear mapping relationship between the seismic data and the density curve is established through a neural network algorithm; Applying the first nonlinear mapping relationship corresponding to each earthquake to the entire seismic data volume to obtain a velocity model volume; or, In various seismic phases of the target layer, the seismic data and wave impedance curve corresponding to the pseudo well are used as training samples, and a second nonlinear mapping relationship between the seismic data and the wave impedance curve is established through a neural network algorithm; The second nonlinear mapping relationship corresponding to each earthquake is applied to the entire seismic data volume to obtain a wave impedance model volume.
10. A seismic inversion low-frequency model building device, characterized in that: include: An acquisition module is used to acquire the seismic data volume and velocity spectrum of the study area and the structural horizons of each target layer in the study area; A layer velocity sequence determination module, used for converting the velocity spectrum to obtain a layer velocity sequence of each target layer; A seismic phase determination module, used to extract various seismic attributes of the target layer based on the structural horizon of the target layer, and predict the seismic phase corresponding to the target layer according to the various seismic attributes; Well logging curve determination module; For selecting at least one position within the range of each type of seismic phase of the target layer to establish a pseudo well, obtaining the interval velocity sequence of the pseudo well from the interval velocity sequence of each target layer, and determining a well logging curve according to the interval velocity sequence of the pseudo well; The model building module is used to obtain seismic data corresponding to the pseudo well from the seismic data volume in various seismic phases of the target layer, and to establish a low-frequency model according to the seismic data and the well logging curve.
11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for establishing a seismic inversion low-frequency model as described in any one of claims 1 to 9 is implemented.
12. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the method for establishing a seismic inversion low-frequency model as described in any one of claims 1 to 9 when executing the program stored in the memory.
