• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    method and system for the separation of seismic sources in maritime acquisition of simultaneous firing. the present invention relates to the seismic data that are obtained by simultaneous recording on seismic streamers, acquired by the almost simultaneous activation of two or more seismic sources, towed in two positions in the vicinity of the seismic streamers. a waste is iteratively updated to an inversion solution for the activation of two or more seismic sources. the iterative residue update uses a sequence of overlapping temporary windows containing reflection events and uses normal change corrections based on larger reflection events in each temporary window. an updated final waste is added to a final updated result of the model.
    公开号:BR102012028800B1
    申请号:R102012028800-1
    申请日:2012-11-09
    公开日:2021-04-27
    发明作者:Rolf H. Baardman；Roald G. Van Borselen
    申请人:Pgs Geophysical As；
    IPC主号:
    专利说明:

    BACKGROUND
    [001] In the oil and gas industry, geophysical prospecting is commonly used to assist in the search and evaluation of subsurface land formations. Geophysical prospecting techniques provide insight into the structure of the subsurface land, which is useful for finding and extracting valuable natural resources, particularly hydrocarbon deposits such as oil and natural gas. A well-known geophysical prospecting technique is seismic research. In a earth-based seismic survey, a seismic signal is generated on or near the surface of the earth and then travels down and into the earth's subsurface. In a marine seismic survey, the seismic signal can also travel downward through a body of water that covers the earth's subsurface. Seismic energy sources are used to generate the seismic signal which, after propagation into the earth, is reflected at least partially by subsurface seismic reflectors. Such seismic reflectors are usually interfaces between underground formations having different elastic properties, specifically the speed of the sound wave and the density of the rock, which lead to differences in acoustic impedance at the interfaces. The reflected seismic energy is detected by seismic sensors (also called seismic receptors) on or near the surface of the earth, in an overlying body of water, or in wells at known depths. Seismic sensors generate signals, usually electrical or optical, from the detected seismic energy, which are recorded for further processing.
    [002] The appropriate seismic sources for generating the seismic signal in terrestrial seismic surveys may include explosives or vibrators. Marine seismic surveys normally use a submerged seismic source towed by a ship and activated periodically to generate an acoustic wave field. The seismic source generating the wave field can be of various types, including a small explosive charge, a spark or electric arc, a marine vibrator, and generally, a cannon. The seismic source can be a water cannon, a steam cannon and, more typically, an air cannon. Typically, a marine seismic source consists not of a single source element, but of a spatially distributed set of source elements. This is particularly true for air cannons, currently the most common form of marine seismic source.
    [003] Appropriate types of seismic sensors typically include particle velocity sensors, particularly in land surveys, and water pressure sensors, particularly in marine surveys. Sometimes, particle displacement sensors, particle acceleration sensors, or pressure gradient sensors are used in place of or in addition to particle speed sensors. Particle velocity sensors and water pressure sensors are generally known in the art as geophones and hydrophones, respectively. Seismic sensors can be deployed on their own, but are most commonly used in sensor assemblies. In addition, pressure sensors and particle motion sensors can be deployed together in a marine survey, placed in pairs or pairs of matrices.
    [004] In a marine seismic survey, a seismic survey vessel travels on the water surface, usually at approximately 5 knots, and contains seismic acquisition equipment, such as navigation control, seismic source control, sensor control seismic and recording equipment. The seismic source control equipment causes a seismic source towed in the water body, by the seismic vessel to act at selected times (the activation commonly known as "firing"). Seismic streamers, also called seismic cables, are elongated cable-like structures, towed in the body of water by the seismic research vessel tows the seismic source that or by another seismic research vessel. Typically, a plurality of seismic streamers are towed behind a seismic vessel. Seismic streamers contain sensors to detect reflected wave fields starting from the seismic source and returning from reflective interfaces. BRIEF DESCRIPTION OF THE DRAWINGS
    [005] The invention and its advantages can be more easily understood by reference to the following detailed description and the accompanying drawings, in which:
    [006] figure 1 is a diagram of one of an exemplary system for acquiring seismic data suitable for use with incorporations of the invention;
    [007] figure 2 is a diagram illustrating, by way of example, one of the many different types of computer systems that can be used in accordance with embodiments of the invention;
    [008] figure 3 is a flowchart that illustrates an exemplary modality of a method for the separation of seismic sources in marine acquisition by simultaneous firing;
    [009] figure 4 is a flow chart illustrating an exemplary embodiment of a method for solving iterative inversion of models for the activation of two or more seismic sources; and
    [0010] figure 5 is a flow chart illustrating an exemplary embodiment of a method for iteratively updating a waste. DETAILED DESCRIPTION
    [0011] Figure 1 is a diagram of one of an exemplary system for acquiring seismic data suitable for use with incorporations of the invention. In several embodiments, a simple seismic sensor cable (also called a seismic streamer) or a simple ocean floor cable is shown for simplicity of illustration. This cable illustration is only to demonstrate the principles of the invention more clearly and is not intended to be a limitation of the invention. Multiple streamers or cables can also be used.
    [0012] In figure 1, a seismic acquisition system is generally designated as 100. A seismic vessel 101 is arranged in a body of water 102 and carries equipment 103 for navigation, control of the seismic source, and recording of the seismic sensor. Seismic vessel 101 or another service vessel (not shown) tows a seismic source 104 through the water body 102 below the water surface 105. Seismic source 104 comprises any type of suitable source, typically in sequences. In one embodiment, one or more seismic sources 104 are towed at the front of seismic streamer 106 by seismic vessel 101, while one or more seismic sources 117 are towed at the back of seismic streamer 106, for example, by another service ship 118. The configuration of seismic sources 104 illustrated in seismic acquisition system 100 is not intended to be a limitation of the invention. Other configurations of seismic sources 104 can also be used.
    [0013] In one embodiment, seismic vessel 101 or another service vessel (not shown) tows a seismic streamer 106 through the body of water 102. Seismic streamer 106 comprises seismic sensors 107 in positions spaced from each other along the streamer seismic 106, in such a way that the seismic streamer 106 containing the seismic sensors 107 is arranged in the water body 102. Seismic sensors 107 are normally pressure sensors, such as hydrophones. In another embodiment, the seismic streamer 106 comprises a double sensor streamer, in which the seismic sensors 107 comprise pairs of pressure and particle motion sensors ordered together. Particle motion sensors are usually particle speed sensors, such as geophones, or accelerometers. Seismic sensors 107 typically comprise arrays of sensors in each spaced position. An alternative to having the pressure and particle motion sensors ordered together is to have sufficient spatial density of sensors in such a way that the respective wave fields recorded by the pressure and particle motion sensors can be interpolated or extrapolated to produce the two signals. wave field in the same location.
    [0014] In another embodiment, seismic vessel 101 or other service vessel (not shown) has an ocean floor cable 108 on the water bottom 109. Ocean floor cable 108 also includes seismic sensors 107 in spaced positions along the cable, also normally in sensor arrangements at each spaced position. Seismic sensors 107 on the ocean floor sensor 108 can also comprise pairs of pressure and particle motion sensors. In yet another embodiment, both seismic streamers 106 and ocean floor cable 108 are employed. The type of sensors illustrated in the seismic acquisition system 100 is not intended to be a limitation of the invention. For example, in other embodiments, discrete seismic sensors 107 located at seabed nodes (not shown) or seismic sensors 107 located in vertical arrangements or in drillholes (not shown) can be included in the seismic acquisition system 100. Others types of seismic sensors 107 and their arrangements can also be used.
    [0015] When the seismic sources 104 are activated, the acoustic energy moves downwards, in 110, through the water body 102 and the bottom of the water 109 to the limits of the layers, such as 111 and 112, around a underground formation layer, such as 113. Some of the acoustic energy is reflected from the boundary / boundary of the layer at 111 and travels upward at 114. The upward displacement acoustic energy 114 is detected in seismic sensors 107 on the bottom of the ocean 108 or the seismic streamer 106. The upward displacement acoustic energy continues upward at 115 until it reflects off the water surface 105 and then moves downward again at 116. The acoustic energy moving at Downward direction 116 can be detected by seismic sensors 107 again on seismic streamer 106 or on the ocean floor cable 108, resulting in a phantom signal. The acoustic energy detected in seismic sensors 107 can be recorded on any type of storage media, in any appropriate location, such as, but not restricted to, seismic streamer 106 or ocean floor cable 108, on seismic vessel 101 or other service ship, or ashore.
    [0016] Figure 2 is a diagram illustrating, by way of example, one of the many different types of computer systems that can be used in accordance with embodiments of the invention. A central processor 20 is coupled to user input devices, such as a 61 (wired or wireless) keyboard and a mouse 22 (wired or wireless). The processor 20 is, in addition, coupled to a screen, such as a monitor 23. A computer program according to embodiments of the invention can reside in any of a series of computer-readable media, such as a disk 24 insertable into a disk drive 25 or an internal or external hard drive (not shown).
    [0017] As shown above in Figure 1, offshore seismic exploration normally employs a submerged seismic source 104 towed by a ship and periodically activated to generate an acoustic wave field (the trigger). The wave field can be generated by a small explosive charge, a spark or electric arc, a marine vibrator, or normally, a cannon. The cannon can be a water cannon, a steam cannon or, more typically, an air cannon. Each air gun contains a volume of compressed air normally to approximately 2000 psi (pounds per square inch) or more. An air cannon abruptly releases its compressed air to create an air bubble, leading to an expanding sound wave in the water. The resulting wavefront propagates down and into the earth under water, is reflected in the subterranean layers of earth, and returns upwards towards the water surface.
    [0018] In seismic exploration, there is a continuous drive for more dense data sampling to produce a better image of complex geological structures. Recent advances in acquisition such as Amplo-Azimuth, Multi-Azimuth or Rico-Azimuth can deliver a more diverse range of source, azimuth and compensation sampling. To collect this data, several seismic service vessels to supply on one side and receiving vessels to receive on the other are used, thereby significantly increasing research costs.
    [0019] In conventional acquisition, there is an overlap between the zero time firing registers, and seismic data are recorded discontinuously. As a result, the seismic source domain is often poorly sampled, leading to scaling / knurling. When acquiring simultaneous shots, data can be recorded continuously, and temporary overlap between shots is allowed. As a result, more sources are triggered during the same acquisition period, which improves flexibility in search geometries. As a result, a set of more densely sampled data established in terms of font spacing, but also azimuth and displacement distributions, can be obtained. The resulting recorded shot records are also known as "mixed" shot records.
    [0020] In terms of efficiency, simultaneous acquisition can contribute to the reduction of research times, which is of special value in critical situations, where small time windows for acquisition dominate due to strict security, environmental, or economic restrictions.
    [0021] The simultaneous triggering of marine seismic sources can provide significant advantages in the efficiency of seismic acquisition. However, each source is affected by seismic interference from one or more of the other sources. This interference associated with each seismic source needs to be computed so that effective processing techniques can be applied to acquired seismic data to separate seismic data sources.
    [0022] The randomization of the seismic source synchronism allows the attenuation of the energy of the sources of interference after the classification of the acquired data for a domain that makes it possible to align the contributions of a specific source to zero time, while the other source (s) will appear be inconsistent. As a result, methods that can discriminate between coherent and incoherent energy allow the sources to be separated.
    [0023] Using techniques in accordance with embodiments of the invention, two or more sources can be separated, or "unbundled", by being built by minimizing a cost function. The cost function describes the size of the residual, or mismatch between input data and modeled data. How to use the cost function to calculate an iterative solution for investing in a sequence of overlapping temporary windows will be described below in exemplary embodiments of the invention.
    [0024] One embodiment includes a method for separating seismic fonts into seismic data that were acquired using simultaneous triggering sea acquisition. Using matrix notation, seismic data can be represented by a P data matrix, with its columns representing shot records and its rows representing receiving concentrations. Therefore, the element Pij represents a single trace related to the position of the source 'j' and position 'i' of the detector. The position of each element in the matrix P corresponds to the spatial coordinates of the sources and receivers; with different rows corresponding to different detector numbers and different columns corresponding to different experiments, that is, trigger records. In general, a seismic source mixture can be formulated as follows:

    [0025] Where P 'is the matrix of mixed data, and zd, and zs are the detector and the depth levels of the source, respectively. The mixing matrix r contains the mixing parameters. In the case of a marine survey with random firing times, but equal source forces, only phase encoding is used. In this case, the mixing elements rkl consist only of phase terms that express the time delay Tkl given to source k in the mixed source matrix I.
    [0026] In order to retrieve individual “decompensated” trigger records from the mixed data matrix P ', a matrix inversion has to be performed. In general, the mixing problem is indeterminate (that is, P 'has fewer columns than P), meaning that unique solutions to the inverse problem do not exist. Therefore, the mixing matrix r is not invertible.
    [0027] An objective, also known as an objective function, can be defined for the inverse problem. The objective is a functional one (that is, a map of a vector space to the field underlying the vector space) that measures how close the decomposed data, represented by the vector m, of the recovered model fits to the mixed input data d. The standard objective function is usually of the form:

    [0028] Which represents the L2 standard of the residual, which is defined as the mismatch between the recorded mixed data d and the predicted mixed data mr, which can be constructed from the predicted combined data m. Note that an L2 standard is used here as a generic measurement of the distance between the predicted data and the observed data, but other standards are also possible to use. In addition, this embodiment is illustrated with two sources, but this is not a restriction of the invention. Two or more sources can be used.
    [0029] The goal of the objective function is to minimize the difference between predicted and observed data. To minimize the objective function (that is, solve the inverse problem), the gradient of the objective function is calculated, using the same reasoning to minimize a function of just one variable. The gradient of the objective function in Equation (2), defined as equal to zero, is:

    [0030] What is simplified after the rearrangement in the least square solution:

    [0031] In the marine case, the mixing matrix r contains only phase terms (phase coding). As a result, its least square inverse corresponds to the complex transposed conjugate (Hermitian). The application of the hermitian transposed complex, designated by an "H" envelope, to the mixed input data then leads to "pseudo-decomposed" firing concentrations:

    [0032] From a physical point of view, the "pseudo-decombination" process performs an expansion corresponding to the number of seismic sources that are mixed together. For example, if the number of sources is b, each mixed trigger register is copied b times. Then, each copy of these trigger records is corrected for the corresponding time delays introduced in the field or decoded, in the case of encoded sources (correlation). This corresponds to simply correcting each shot to its own zero time. Since responses from multiple sources are included in a single mixed trigger register and the source codes are not orthogonal, the “pseudo-decombination” process generates correlation noise. This correlation noise is known as "mixing noise" or "cross line". However, these pseudo-combined fire records serve a purpose in further processing, either as a final product, after separation, or as an input to a subsequent source separation method to be applied.
    [0033] An alternative to the least squares solution is to impose restrictions on possible solutions, emphasizing certain possible characteristics in our models. This type of restriction technique is known as regularization.
    [0034] One approach to regularization is to restrict inversion by imposing low density to some convenient domain (For example, transformed CMP Radon concentrations). However, these schemes are known to have, at a minimum, the following drawbacks. Imposing low density not only suppresses noise from the mix, but, unfortunately, also suppresses weaker reflection events. Even in the Radon domain (stacking energy for coherent events), the weakest reflection events cannot be distinguished from mixing noise. In addition, reflection events are usually too complex to be described by just a few Radon parameters, as the result is not sparse enough.
    [0035] An alternative to using low density constraints is to use coherence constraints. In such methods, incoherent energy re-known as a representation of coherent events belonging (one of) to the interfering shots. The inversion approach aims to distribute all the energy in the mixed trigger records by reconstructing the individual unmixed trigger records in their respective locations.
    [0036] In exemplary embodiments, a method can be used that restricts inversion by imposing that the recorded wave field is regular, that is, that nearby sources produce records that are similar to each other in the sense that the desired records m they contain data that is, at least locally, predictable and contain coherent events. There are several methods to separate the coherent signal from incoherent interference from overlapping shots, for example, F-X multidimensional de-convolution, however, alternative methods can also be used.
    [0037] To assist in the separation of coherent from incoherent energy, an NMO correction can be used to flatten events so that they become more parabolic. However, an extensive speed profile that aims to flatten all events will result in stretching effects that can be serious. At the same time, a simple speed model (for example, very smooth and with low speeds) will produce less serious stretch effects, but will not flatten all events as desired.
    [0038] In order to overcome this problem, it is proposed to build the decomposed triggering records using temporary overlapping windows. In the first update step, the reflection of the bottom of the water is considered, where the NMO correction with the water velocity is applied to all pseudo-decomposed concentrations. Typically, the NMO correction is applied after the classification of the seismic data for a Common Depth Point (CDP) domain in order to flatten the reflection events as much as possible. Note that only minimal stretch effects are introduced when doing this, because a constant speed is used.
    [0039] To extract coherent signals from NMO-corrected data concentrations, a median multidimensional filter can be used. The type of filter used is not limited to medium filters. The type of filter used may include, but is not limited to, other filters such as frequency filters - wave number, Radon filters and curvelet or wavelet domain filters. Subsequently, an inverse NMO is applied to the signals obtained, after which time delays are applied, the results are added and subtracted from the residue. The residue will then serve as an entry for the next update, progressively expanding the temporary window of interest, thus using an appropriate NMO correction for the extended temporary window under consideration.
    [0040] An advantage of the temporary window approach is that it is able to align the target events with less stretching effects. In addition, it allows the removal of strong incoherent peaks of shallow data first, so that coherent energy extraction is easier when the process is repeated for longer arrival times, that is, for deeper events. The invention is not limited to the use of restricted NMO correction time windows. Other methods, such as static changes, can be used to avoid the effects of stretching.
    [0041] The proposed multidimensional median filter improves the separation of any overlapping events, that is, a reflection event and a peak of noise superimposed on the mixture. Through the use of a medium filter the noise of the mixture is significantly reduced. The destination event cannot be completely reconstructed in a single iteration, but the remaining reflection information still resides in the residual data, that is, it has not leaked to the wrong trigger. As such, the remaining reflection energy can still be added in subsequent iterations. A similar conservative principle is used for weaker or less aligned events. The application of aggressive extraction filters should be avoided to ensure that, if an event is not added to the correct decompiled concentration, it is also not added to the wrong concentration. After the last iteration, the residue is added to all decomposed concentrations to ensure that no relevant information has been lost.
    [0042] The inversion approach starts with the calculation of concentrations of pseudo-decomposed data, obtained as a scaled version of the least squares solution to minimize the objective, or cost, function Φ (m1, ..., mn) :

    [0043] Here, d are the mixed input data (as originally measured) and m = [m1, ... mN] T constitutes the decomposed data concentrations in the time domain and | | 2 represents the L2 or Euclidean norm (distance function).
    [0044] In the first update step, an NMO correction using water velocity is applied to all pseudo-decomposed concentrations. It is again observed that only minimal stretching effects are introduced when doing this, because a constant speed is used. Then, a temporary window is considered to contain the deep reflection event of the flattened water. Then, a median multidimensional filter is used to extract the coherent signals. Subsequently, an inverse NMO is applied to the obtained signals, after which time delays are applied; the results are added and subtracted from the residue. The residue will then serve as an entry for the next update, progressively expanding the temporary window of interest.
    [0045] Initially, the models are empty and thus, starting from Equation (6), the initial residue r (0) is equal to the mixed input data d. In iteration i °, the residue i ° r (i) is given by:

    [0046] Models m1 to mN for the separate sources are built iteratively, where in each subsequent iteration, a larger temporary window is considered, focusing on the next event of strong reflection (deeper). This procedure is repeated until all events in the seismic data have been reconstructed. After the iterative scheme has been completed, the residual data is added to the decomposed data sets m1, ..., mN.
    [0047] The NMO correction is normally applied in the Common Depth Point (CDP) domain, and as such, this domain is convenient for calculating the residual and reconstructing the m1, ..., mN models. However, the invention is not limited to this choice of concentrations. The process can be applied to other concentrations, such as common source or concentrations of common receptor.
    [0048] The proposed median multidimensional filter will have a leveling effect by averaging / smoothing events, which is good for stopping overlapping events. Otherwise, a reflection event and the noise peaks superimposed on the mixture (belonging to coherent events in the other trigger) will both be added to the update (leakage, seen as loopholes in the other trigger). The target event may not be completely removed in this way (the rest is still in the residual data, and is not leaked there). However, the corresponding noise mix is significantly reduced in such a way, in a later iteration, a deeper and weaker event can be discarded as well and added to the correct decomposed concentration. While the events are in the residue, the events have not leaked to the wrong trigger and can be added later. The same conservative principle can be used for weaker or less aligned events. Overly aggressive filters should not be used to ensure that if an event is not added to the correct decompiled concentration, it is also not added to the wrong concentration. As a result, after the update, the residue must be added to all decomposed concentrations, giving a slightly noisier output.
    [0049] Figure 3 is a flowchart that illustrates an exemplary modality of a method for separating the effects of seismic sources in marine acquisition by simultaneous firing.
    [0050] In block 30, the seismic data is obtained by simultaneous recording in seismic streamers, acquired by the almost simultaneous activation of two or more seismic sources, towed in two or more positions close to the seismic streamers.
    [0051] In block 31, a residue is iteratively updated to a model inversion solution for the activation of two or more seismic sources in block 30. The iterative update of the residue uses a sequence of overlapping temporary windows containing reflection events and uses normal change corrections based on major reflection events in each temporary window.
    [0052] In block 32, an updated final residue from block 31 is added to a result of the final updated model in block 31.
    [0053] Figure 4 is a flowchart that illustrates an exemplary embodiment of a method for solving iterative inversion of models for the activation of two or more seismic sources.
    [0054] In block 40, one or more seismic sources are towed in a first position in the vicinity of seismic streamers and one or more seismic sources are towed in one or more positions in the vicinity of seismic streamers. In one embodiment, the first position can be in front of the seismic streamers and another position can be in the back of the seismic streamers. However, the invention is not intended to be restricted to these positions.
    [0055] In block 41, the two or more seismic sources separate from block 40 are activated at approximately the same time. Random time delays from -500 ms to +500 ms are applied to the activation time of the two or more seismic sources.
    [0056] In block 42, seismic data is recorded in seismic streamers in response to activations of the seismic source in block 41. Responses to two or more seismic sources are recorded simultaneously.
    [0057] In block 43, a set of temporary windows, each containing reflection events, is selected from the seismic data re-recorded in block 42, starting with the reflection from the bottom of the water and proceeding downwards. Let the number of temporary windows be designated as N. These temporary windows do not have to follow the recorded reflection events or any subsurface reflectors. Temporary windows can reflect isochrones or pseudolimites representing windows in time, specified only by a window length.
    [0058] In block 44, in the initial iterative step 0, the results of the uncombined model m (0) are placed equal to zero. Thus, the initial residue r (0) is equal to the mixed input data d of block 42.
    [0059] In block 45, in each iterative step i, from i = 1, 2, ..., N, the residue r (i) is updated to a solution for inversion of the results of the model m (i) in the event i ° in the selected set of temporary windows in block 43.
    [0060] In block 46, after the final iterative step, the updated final residue r (N) from block 45 is added to the result of the m (N) model.
    [0061] Figure 5 is a flowchart that illustrates an exemplary embodiment of a method for iteratively updating the waste. This flowchart describes in more detail the iterative step in the process cited in block 45 of figure 4.
    [0062] In block 50, the iterative residue r (i-1) ° anterior r (i-1) is taken from block 45 in figure 4.
    [0063] In block 51, residue r (i-1) from block 50 is pseudo-decomposed for the two or more seismic sources in block 40 of figure 4 and classified in CDP concentrations.
    [0064] Because two or more sources are triggered with only small random delays between them, the responses of the sources are recorded at the same time. Decomposition of seismic data means that data from one source is aligned by reapplying random time delays and then that part of the data (mixing noise) that was triggered by other sources is removed. Pseudo-decombination means that the data is aligned only to the desired source, while the mixing noise is uncontaminated.
    [0065] In block 52, a correction - NMO is applied to the pseudo-decomposed CDP concentrations of block 51. Although this NMO correction can work better in CMP concentrations, the invention is not restricted to this choice of collecting. The process can be applied to other concentrations, such as common source or concentrations of common receptor.
    [0066] In block 53, a median multidimensional filter is applied to recover the coherent signals in the corrected CDP concentrations NMO of block 52. A temporary window can be used to limit the area where the filter is active.
    [0067] In block 54, an inverse NMO correction is applied to the signals recovered from block 53.
    [0068] In block 55, the random time delays of block 41 of figure 4 are applied to the corrected reverse signals NMO of block 54.
    [0069] In block 56, mixing is applied to signals with delay time in block 55.
    [0070] In block 57, the mixed signals from block 56 are subtracted from the previous residue r (i-1) to generate the new residue r (i).
    [0071] The seismic data obtained in the conduct of a seismic survey, representative of the terrestrial subsurface are processed to obtain the information related to the geological structure and the properties of the land formations in the subsurface in the area being surveyed. The processed seismic data is processed to present and analyze the potential hydrocarbon content of these underground formations. The purpose of seismic data processing is to extract as much information from seismic data as possible about underground formations, in order to adequately produce the geological image of the subsurface. In order to identify locations on the Earth's subsurface, where there is a high probability of encountering oil accumulations, large sums of money are spent on collecting, processing and interpreting seismic data. The process of construction of the reflective surfaces that define the layers of underground earth of interest from the seismic data provides an image of the earth in depth or time. A prerequisite for discovering any oil or gas reservoir is a well-resolved seismic image of the earth's subsurface.
    [0072] The image of the structure of the earth's subsurface is produced in order to allow an interpreter to select the places most likely to have accumulations of oil. To check for the presence of oil, a well must be drilled. Drilling wells to determine whether oil deposits are present or not is an extremely expensive and time-consuming undertaking. For this reason, there is a permanent need to improve the processing and presentation of seismic data, in order to produce an image of the structure of the Earth's subsurface that will improve the capacity of an interpreter, whether the interpretation is done by a computer or a human being , to assess the likelihood that an accumulation of oil exists at a particular location on the earth's subsurface. The processing and visualization of acquired seismic data facilitates more accurate decisions about drilling and where to drill, thereby reducing the risk of drilling dry wells.
    [0073] The embodiments according to the invention have been discussed above as methods for illustrative purposes only. Achievements can also be implemented as systems. Systems according to embodiments of the invention can, for example, be implemented by means of computers, particularly in digital computers, together with conventional data processing equipment. Such data processing equipment, well known in the art, may comprise any appropriate combination or network of computational processing equipment, including, but not limited to, hardware (processors, temporary and permanent storage devices, and any other processing equipment. computer-based), software (operating systems, application programs, mathematical program libraries and any other appropriate software), connections (electrical, optical, wireless or otherwise), and peripheral (input and output devices such as keyboards, pointing devices, and scanners, display devices such as monitors and printers, computer-readable storage media such as tapes, discs and hard drives, and any other appropriate equipment).
    [0074] A computer program can be stored on a computer-readable medium with the program having operable logic to make a programmable computer perform any of the methods described above.
    [0075] It should be understood that the foregoing is simply a detailed description of specific embodiments of the present invention and that numerous changes, modifications, and alternatives to the described corporations can be made in accordance with the disclosure here without departing from the scope of the invention . The foregoing description, therefore, does not mean limiting the scope of the invention. More precisely, the scope of the invention is to be determined only by the appended claims and their equivalents.
    权利要求:
    Claims (20)
    [0001]
    1. Method for the separation of effects from multiple seismic sources (104) in the acquisition of simultaneous triggering, characterized by comprising: obtaining (30) of the mixed seismic data recorded simultaneously in seismic streamers (106), acquired by the approximately simultaneous activation of two or more seismic sources (104), towed in two or more positions close to the seismic streamers (106); for each activation of the two or more seismic sources (104): - calculate predicted mixed seismic data as a product of a mixing matrix based on time delay and unmixed seismic data models; - calculate a residual difference between the mixed seismic data and the expected mixed seismic data; - update (31) the residue to a solution of inversion of the unmixed seismic data models, in which the update of the residue uses a sequence of overlapping temporal windows containing reflection events and uses normal change corrections based on events greater reflection rates in each time window; and addition (32) of an updated final waste to a result of the final updated unmixed seismic data model.
    [0002]
    2. Method according to claim 1, characterized by the fact that the activation of the two or more seismic sources (104) approximately simultaneously comprises: application of random time delays from -500 ms to +500 ms to the activation time of the two or more seismic sources (104).
    [0003]
    3. Method according to claim 2, characterized by the fact that the two or more positions comprise positions in front and behind, respectively, of the seismic streamers (106).
    [0004]
    4. Method according to claim 1, characterized by the fact that the update of the residue comprises: selection of a set of N time windows, each containing reflection events, in the recorded seismic data recorded, starting with background reflection of the water and proceeding downwards; placing, in the initial iterative step 0, the results of the unmixed data model m (0) equal to zero; and updating, at each iterative step i, of i = 1, 2, ..., N, the residue r (i) for a solution of inversion of the results of the unmixed seismic data model m (i) in the i-th event in the selected set of N time windows.
    [0005]
    5. Method according to claim 4, characterized by the fact that the addition of the updated final residue to the result of the final updated unmixed seismic data model comprises: Add, after the final iterative step, the updated final residue r (N) to the final result of the m (N) model.
    [0006]
    6. Method according to claim 4, characterized by the fact that updating, in each iterative step i, of i = 1, 2, ... N, the residual r (i) comprises: pseudo-non-mixing of the residue r (i-1) for the two or more seismic sources (104), generating pseudo-unmixed CDP concentrations as a product of the mixed seismic data multiplied by a time delay-based mix matrix transposition (51) ; applying an NMO correction to pseudo-unmixed CDP concentrations (52); applying a filter to recover the coherent signals at the corrected NMO CDP concentrations (53); applying an inverse NMO correction to the recovered signals (54); application of random time delays to corrected inverse NMO signals (55); mixing application to time-delayed signals (56); and subtracting the mixed signals from the previous residue r (i-1) to generate the new residue r (i) (57).
    [0007]
    Method according to claim 6, characterized in that the application of a filter comprises the application (53) of a median multidimensional filter.
    [0008]
    8. Method according to claim 7, characterized by the fact that the application of a multidimensional median filter additionally comprises: using a temporal window to limit the area where the multidimensional median filter is active (43).
    [0009]
    9. Method according to claim 6, characterized by the fact that the application of a filter comprises the application of a filter selected from the group consisting of frequency filters - wave number, Radon filters and curvelet or wavelet domain filters.
    [0010]
    10. Method according to claim 1, characterized by the fact that the iterative update of the waste uses a static displacement.
    [0011]
    11. System for the separation of effects of various seismic sources (104) in the acquisition of simultaneous triggering, characterized by comprising: two or more seismic sources (104) configured to be activated with random time delays from -500 ms to +500 ms the activation time of the two or more seismic sources (104); and towed seismic streamers (106) configured to simultaneously record mixed seismic data in response to the activations of two or more seismic sources (104), configured to be towed in front and behind, respectively of the towed seismic streamers (106); a programmable computer (20) used to perform at least the following: for each activation of the two or more seismic sources (104): - calculate predicted mixed seismic data as a product of a mixture matrix based on time delay and models unmixed seismic data; - calculate a residual difference between the mixed seismic data and the expected mixed seismic data; - update the residue for an inversion solution of unmixed seismic data models, in which the residue update uses a sequence of overlapping time windows containing reflection events and uses normal change corrections based on larger reflection events in each time window (31); and adding an updated final waste to a result of the final updated unmixed seismic data model (32).
    [0012]
    12. System according to claim 11, characterized by the fact that the update of the residue comprises: selection of a set of N time windows, each containing reflection events, in the recorded mixed seismic data, starting with reflection from the bottom of the water and proceeding downwards; placing, in the initial iterative step 0, the results of the unmixed seismic data model m (0) equal to zero; and updating, at each iterative step i, of i = 1, 2, ..., N, the residue r (i) for a solution of inversion of the results of the unmixed seismic data model m (i) in the i-th temporal window in the selected set of N temporal windows.
    [0013]
    13. System according to claim 12, characterized by the fact that the update, in each iterative step i, of i = 1,2, ... N, the residue r (i) comprises: pseudo-non-mixing of residue r (i-1) for the two or more seismic sources (104), generating pseudo-unmixed CDP concentrations as a product of the mixed seismic data multiplied by a conjugated transposition of the mixing matrix based on time delay (51); applying an NMO correction to pseudo-unmixed CDP concentrations (52); application of a multidimensional median filter to recover the coherent signals in the corrected CDP concentrations NMO (53); applying an inverse NMO correction to the recovered signals (54); application of random time delays to corrected inverse NMO signals (55); mixing application to time-delayed signals (56); and subtracting the mixed signals from the previous residue r (i-1) to generate the new residue r (i) (57).
    [0014]
    14. Computer readable medium with a computer program stored in it, the program having operable logic makes a programmable computer (20) execute at least one method characterized by the fact that it comprises: obtaining (30) of the mixed seismic data recorded simultaneously in seismic streamers (106), acquired by the approximately simultaneous activation of two or more seismic sources (104), towed in two or more positions close to the seismic streamers (106); for each activation of the two or more seismic sources (104): - calculate predicted mixed seismic data as a product of a mixing matrix based on time delay and unmixed seismic data models; - calculate a residual difference between the mixed seismic data and the expected mixed seismic data; update (31) the residue to a solution for inversion of the unmixed seismic data models, in which the update of the residue uses a sequence of overlapping temporal windows containing reflection events and uses normal change corrections based on events of greater reflection in each time window; and adding an updated final waste to a result of the final updated unmixed seismic data model (32).
    [0015]
    15. Medium according to claim 14, characterized by the fact that the activation of two or more seismic sources (104) approximately simultaneously comprises: application of random time delays from -500 ms to +500 ms to the activation time of the two or more seismic sources (104).
    [0016]
    16. Medium according to claim 15, characterized by the fact that the two or more positions comprise positions in front and behind, respectively, of the seized seismic streamers (106).
    [0017]
    17. Medium according to claim 14, characterized by the fact that the update of the residue comprises: selection of a set of N time windows, each containing reflection events, in the recorded seismic data, starting with reflection of the bottom of the water and proceeding downwards; placing, in the initial iterative step 0, the results of the unmixed seismic data model m (0) equal to zero; and updating, at each iterative step i, of i = 1, 2, ..., N, of the residue r (i) for a solution of inversion of the results of the unmixed seismic data model m (i) in the i-th temporal window in the selected set of temporal windows.
    [0018]
    18. Medium according to claim 17, characterized by the fact that the addition of the final updated residue to the result of the final updated model comprises: adding, after the final iterative step, the finalized update r (N) to the final result of the m (N) model.
    [0019]
    19. Medium according to claim 17, characterized by the fact that, in each iterative step i, of i = 1,2, ... N, the residual r (i) comprises: pseudo-non-mixing of the residue r (i-1) for the two or more seismic sources (104), generating pseudo-unmixed CDP concentrations, as a product of the mixed seismic data multiplied by a conjugated transposition of the mixture matrix based on time delay (51 ); applying an NMO correction to pseudo-unmixed CDP concentrations (52); application of a multidimensional median filter to recover the coherent signals in the corrected CDP concentrations NMO (53); applying an inverse NMO correction to the recovered signals (54); application of random time delays to the corrected reversed NMO signals (55); mixing application to time-delayed signals (56); and subtracting the mixed signals from the previous residue r (i-1) to generate the new residue r (i) (57).
    [0020]
    20. Medium according to claim 19, characterized by the fact that the application of a multidimensional median filter that additionally comprises: using a temporal window to limit the area where the multidimensional median filter is active (43).
    类似技术:
    公开号 | 公开日 | 专利标题
    BR102012028800B1|2021-04-27|METHOD AND SYSTEM FOR THE SEPARATION OF EFFECTS FROM VARIOUS SEISMIC SOURCES ON THE ACQUISITION OF SIMULTANEOUS SHOOTING AND MEDIA LEGIBLE BY COMPUTER
    BRPI1104046B1|2021-03-09|method for creating an image of the earth's subsurface
    EP2641109B1|2019-02-20|Separating interfering signals in seismic data
    BRPI0900890B1|2020-10-13|method to interpolate traces in seismic data
    US9229123B2|2016-01-05|Method for handling rough sea and irregular recording conditions in multi-sensor towed streamer data
    BRPI1004588A2|2012-08-28|method for full bandwidth pursuit of marine discharge seismic data
    BR102013002915A2|2015-09-01|Method and system for determining font signatures after font phantom removal
    BRPI0903062A2|2010-11-23|method for wave field separation in 3d double sensor towed recording cable data with distorted energy in the direction of the transverse recording cable
    BRPI0514989B1|2017-05-09|method for attenuation of water bottom multiples in marine seismic data
    BRPI0719140B1|2019-01-08|imaging methods of recorded seismic data, and, computer readable medium that stores computer executable instructions
    BR102012013014B1|2020-03-10|METHOD FOR SEISMIC EXPLORATION, CONFIGURED DEVICE TO GENERATE SEISMIC IMAGES, AND LEGIBLE STORAGE MEDIA BY COMPUTER
    BRPI1101394A2|2012-08-14|Method for separating pressure fields and vertical velocity from upward and downward propagation of pressure and triaxial motion sensors into towed strips
    BRPI1005340B1|2020-02-04|method for converting recorded seismic data from marine streamers to seismic data from non-ghosted sources
    EP2135113A2|2009-12-23|Processing of seismic data acquired using twin over/under streamers
    BRPI1004966A2|2015-09-01|Method for attenuating interference noise in dual sensor seismic data
    BR112019012803A2|2019-12-03|system and method for reconstructed wave field imaging
    BR102013009382A2|2015-06-09|Methods and apparatus for generating withdrawn ghost seismic data
    BR102013003162B1|2020-11-10|method for processing cable data from two sensors with anti-aliasing protection
    BR112020015737A2|2021-03-23|seismic imaging with a temporal decomposition imaging condition
    BR112019014491A2|2020-03-31|COHERENT 3D FILTERING TAU-P
    Cunnell et al.2015|High Quality and Efficient Seismic Acquisition for Frontier Deepwater Areas
    Howie et al.2010|SS: Thunder Horse and Atlantis Deepwater Frontier Developments in the Gulf of Mexico: Paper Title-Two Assets, Two Approaches to Tackling Subsalt Imaging-Thunder Horse and Atlantis
    同族专利:
    公开号 | 公开日
    EP2592439A3|2013-07-17|
    US9075162B2|2015-07-07|
    US20150293250A1|2015-10-15|
    EP2592439B1|2018-10-03|
    BR102012028800A2|2013-12-17|
    AU2012244128A1|2013-05-30|
    US9945972B2|2018-04-17|
    EP3413094A1|2018-12-12|
    US20130121109A1|2013-05-16|
    US20210141114A1|2021-05-13|
    AU2012244128B2|2016-04-14|
    US20180172862A1|2018-06-21|
    EP2592439A2|2013-05-15|
    EP3413094B1|2022-03-16|
    AU2016204363B2|2017-12-14|
    US10921473B2|2021-02-16|
    AU2018201492B2|2020-03-12|
    SG190512A1|2013-06-28|
    AU2016204363A1|2016-07-14|
    AU2018201492A1|2018-03-22|
    MX2012013088A|2013-05-10|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    GB1233027A|1968-09-13|1971-05-26|
    US4458339A|1980-10-06|1984-07-03|Texas Instruments Incorporated|Seismic prospecting using a continuous shooting and continuous recording system|
    US5671136A|1995-12-11|1997-09-23|Willhoit, Jr.; Louis E.|Process for seismic imaging measurement and evaluation of three-dimensional subterranean common-impedance objects|
    GB9927395D0|1999-05-19|2000-01-19|Schlumberger Holdings|Improved seismic data acquisition method|
    US6545944B2|2001-05-30|2003-04-08|Westerngeco L.L.C.|Method for acquiring and processing of data from two or more simultaneously fired sources|
    US6906981B2|2002-07-17|2005-06-14|Pgs Americas, Inc.|Method and system for acquiring marine seismic data using multiple seismic sources|
    US6751559B2|2002-09-10|2004-06-15|Pgs Exploration Limited|Method for suppressing noise from seismic signals by source position determination|
    US6889142B2|2003-05-30|2005-05-03|Pgs Exploration Limited|Method of correcting for time shifts in seismic data resulting from azimuthal variation|
    US7675812B2|2008-06-30|2010-03-09|Pgs Geophysical As|Method for attenuation of multiple reflections in seismic data|
    US8559270B2|2008-08-15|2013-10-15|Bp Corporation North America Inc.|Method for separating independent simultaneous sources|
    US9213119B2|2008-10-29|2015-12-15|Conocophillips Company|Marine seismic acquisition|
    US9052410B2|2009-02-12|2015-06-09|Conocophillips Company|Multiple seismic signal inversion|
    US8818730B2|2010-07-19|2014-08-26|Conocophillips Company|Unique composite relatively adjusted pulse|
    KR101908278B1|2010-09-27|2018-10-17|엑손모빌 업스트림 리서치 캄파니|Simultaneous source encoding and source separation as a practical solution for full wavefield inversion|
    MX2013006454A|2010-12-09|2013-12-06|Bp Corp North America Inc|Seismic acquisition method and system.|
    EP2681589B1|2011-04-18|2015-04-29|Siemens Aktiengesellschaft|Methods for deblending of seismic shot gathers|
    US9562983B2|2014-04-17|2017-02-07|Saudi Arabian Oil Company|Generating subterranean imaging data based on vertical seismic profile data|US9188693B2|2012-03-26|2015-11-17|Apache Corporation|Method for acquiring marine seismic data|
    US10359528B2|2013-03-15|2019-07-23|Pgs Geophysical As|Systems and methods for randomizing firing times of simultaneous sources in marine surveys|
    EP2787371A3|2013-04-03|2015-10-07|CGG Services SA|Acquisition system and method for blended seismic data|
    US9864084B2|2013-06-07|2018-01-09|Cgg Services Sas|Coherent noise attenuation method|
    WO2014195505A2|2013-06-07|2014-12-11|Cgg Services Sa|Virtual marine seismic spread acquisition|
    BR112015030104A2|2013-08-23|2017-07-25|Exxonmobil Upstream Res Co|simultaneous acquisition during both seismic acquisition and seismic inversion|
    US9651695B2|2013-09-19|2017-05-16|Pgs Geophysical As|Construction and application of angle gathers from three-dimensional imaging of multiples wavefields|
    EP3063567A1|2013-11-01|2016-09-07|CGG Services SA|Method and device for deblending seismic data using self-adapting and/or selective radon interpolation|
    EP2999978B1|2013-11-01|2019-12-04|CGG Services SAS|Hybrid deblending method and apparatus|
    US20150185348A1|2013-12-30|2015-07-02|Pgs Geophysical As|Methods and systems for attenuating residual acoustic energy in seismic data|
    US10345470B2|2014-01-13|2019-07-09|Cgg Services Sas|Device and method for deghosting seismic data using sparse tau-p inversion|
    US10598807B2|2014-02-18|2020-03-24|Pgs Geophysical As|Correction of sea surface state|
    GB2524656B|2014-02-26|2020-10-14|Pgs Geophysical As|Methods and systems for quantifying coherency and constraining coherency-based separation in simultaneous shooting acquisition|
    US10670757B2|2014-02-26|2020-06-02|Pgs Geophysical As|Methods and systems for quantifying coherency and constraining coherency-based separation in simultaneous shooting acquisition|
    EP3117246A2|2014-03-14|2017-01-18|CGG Services SA|Method and apparatus for estimating source signature in shallow water|
    US9903966B2|2014-04-14|2018-02-27|Pgs Geophysical As|Seismic data acquisition|
    JP6347480B2|2014-05-27|2018-06-27|独立行政法人石油天然ガス・金属鉱物資源機構|Vibration detection system, signal processing apparatus, and signal processing method|
    US9851463B2|2014-07-01|2017-12-26|Pgs Geophysical As|Interference attenuation of a residual portion of seismic data|
    NO338421B1|2014-07-03|2016-08-15|Kongsberg Seatex As|Method and system for dynamic positioning of instrumented tow cable in water|
    US10132946B2|2014-08-13|2018-11-20|Pgs Geophysical As|Methods and systems that combine wavefields associated with generalized source activation times and near-continuously recorded seismic data|
    US10317553B2|2014-08-13|2019-06-11|Pgs Geophysical As|Methods and systems of wavefield separation applied to near-continuously recorded wavefields|
    US10094944B2|2014-09-05|2018-10-09|Westerngeco L.L.C.|Separating survey data for a plurality of survey sources|
    US10073183B2|2014-10-20|2018-09-11|Pgs Geophysical As|Methods and systems that attenuate noise in seismic data|
    US10101480B2|2014-10-20|2018-10-16|Pgs Geophysical As|Methods and systems to separate seismic data associated with impulsive and non-impulsive sources|
    US10261207B2|2014-12-18|2019-04-16|Pgs Geophysical As|Seismic noise mitigation system and method|
    MA41215A|2014-12-18|2017-10-24|Brewer Joel D|SIMULTANEOUS SOURCE SEPARATION PROCESSES|
    US11112518B2|2015-02-24|2021-09-07|Cgg Services Sas|Method and apparatus for deblending seismic data using a non-blended dataset|
    US10768322B2|2015-08-27|2020-09-08|Pgs Geophysical As|Analogous processing of modeled and measured marine survey data|
    US10267940B2|2015-10-05|2019-04-23|Pgs Geophysical As|Noise template adaptation|
    US10436924B2|2015-10-14|2019-10-08|Pgs Geophysical As|Denoising seismic data|
    US11143774B2|2015-12-22|2021-10-12|Shell Oil Company|Method and system for separating blended seismic data|
    CN106226818B|2016-04-01|2018-05-04|中国石油天然气股份有限公司|Seismic data processing technique and device|
    US10267936B2|2016-04-19|2019-04-23|Pgs Geophysical As|Estimating an earth response|
    CN107315196B|2016-04-26|2020-08-25|中国石油化工股份有限公司|Improved VSP wave field separation method and device based on median filtering|
    US10571589B2|2016-08-17|2020-02-25|Pgs Geophysical As|Constraint of dithering of source actuations|
    US10520621B2|2016-11-23|2019-12-31|Cgg Services Sas|Method and apparatus for simultaneous geostatistical inversion of time-lapse seismic data|
    US10809402B2|2017-05-16|2020-10-20|Conocophillips Company|Non-uniform optimal survey design principles|
    US10649108B2|2017-10-19|2020-05-12|Cgg Services Sas|System and method for generating dithering sequences for seismic exploration|
    US11086038B2|2017-10-20|2021-08-10|Pgs Geophysical As|Seismic noise attenuation using adaptive subtraction with improved noise estimation|
    CN108957525B|2018-03-13|2021-04-16|中国海洋石油集团有限公司|Continuous recording method for seismic data acquisition|
    CN110824562A|2018-08-08|2020-02-21|中国石油化工股份有限公司|Microseism signal correction value calculation method and system|
    US20200073001A1|2018-08-29|2020-03-05|Pgs Geophysical As|Separation of blended marine seismic survey dataacquired with simultaneous multi-source actuation|
    US11009618B2|2018-09-04|2021-05-18|Sercel|System and method for generating dithering sequences with minimum value for seismic exploration|
    CA3115062A1|2018-10-12|2020-04-16|Bp Corporation North America Inc.|Separation of multiple seismic sources of different types by inversion|
    US11175424B2|2019-02-25|2021-11-16|Saudi Arabian Oil Company|Seismic data de-blending|
    EP3938810A1|2019-03-15|2022-01-19|BP Corporation North America Inc.|Signal recovery during simultaneous source deblending and separation|
    CN112147692A|2019-06-28|2020-12-29|中国石油化工股份有限公司|Secondary seismic source attenuation method for marine seismic data|
    WO2021162987A1|2020-02-12|2021-08-19|Schlumberger Technology Corporation|System and method for simultaneously acquiring wide azimuth and ocean bottom node surveys|
    法律状态:
    2013-12-17| B03A| Publication of a patent application or of a certificate of addition of invention [chapter 3.1 patent gazette]|
    2018-12-04| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|
    2020-06-09| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|
    2021-03-09| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|
    2021-04-27| B16A| Patent or certificate of addition of invention granted [chapter 16.1 patent gazette]|Free format text: PRAZO DE VALIDADE: 20 (VINTE) ANOS CONTADOS A PARTIR DE 09/11/2012, OBSERVADAS AS CONDICOES LEGAIS. |
    优先权:
    申请号 | 申请日 | 专利标题
    US13/293,832|2011-11-10|
    US13/293,832|US9075162B2|2011-11-10|2011-11-10|Method and system for separating seismic sources in marine simultaneous shooting acquisition|
    [返回顶部]