• 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:
    INVERSION OF A FULL WAVES FIELD USING VARIABLE FILTERS WITH TIME.An improved method to reduce the precision requirements of the starting model when performing a multi-scale inversion of seismic data (65) for local optimization of the objective function (64). The different inversion scales are caused by incorporating a low-pass filter for objective function (61), and then decreasing the amount of high frequency data, which is filtered from one scale to the next. In addition, the filter is designed to be in a variation of time, in which the cutoff frequency of the low-pass filter decreases with the increase in the transit time of the seismic data being filtered (62). The filter can be designed using Pratt's criterion to eliminate local minimums, and averaging (or other statistical measure) of the transit time period and having the error only with respect to the source and location in the receiver, but not in transit time (63).
    公开号:BR112012019562A2
    申请号:R112012019562-0
    申请日:2011-02-21
    公开日:2020-08-18
    发明作者:Jerome R. Krebs；John E. Anderson
    申请人:Exxonmobil Upstream Research Company；
    IPC主号:
    专利说明:

    “INVERSION OF A FULL WAVES FIELD USING VARIABLE FILTERS WITH TIME”
    CROSS REFERENCE FOR RELATED APPLICATION This application claims the benefit of Provisional Patent Application 61/318, 561, —USA registered on March 29, 2010, titled INVERSION OF A FULL WAVY FIELD USING TIME VARIABLE FILTERS, all of which are here is incorporated by reference.
    FIELD OF THE INVENTION The invention generally refers to the field of numerical inversion of seismic data to infer elastic parameters of the propagation medium. More particularly, the invention is a method to reduce the precision requirements in a starting model when it performs optimization of the objective function in inversions such as the inversion of seismic data in places.
    BACKGROUND OF THE INVENTION Inversion [see, for example, Tarantola, 1984] of attempts to find a model that perfectly explains the observed data. Local inversion methods that minimize the value of an objective function that measures the difference between simulated and observed data, and is often the only practical way to solve an inversion problem for a model with a large number of free parameters. These local methods require an initial estimate for the model to be inverted. They then interactively update the model to bring it closer to the real solution by looking for a change in the current model in a direction based on the gradient of the objective function. Unfortunately, the objective function often has many minimums, not just a minimum corresponding to the solution model. These other minimums are called local minimums, while the minimum corresponding to the desired solution is called the global minimum. If the starting model for the inversion is very close to the model corresponding to one of these local minimums, then the local inversion methods will stay closer to this local minimum and never repeat away from it towards the global minimum. Thus, the wrong solution is produced, no matter how much effort is spent on iterating.
    This local minimum problem can be solved first by an iteration inversion over an altered objective function that has fewer local minimums but has an overall minimum near the point of the desired solution. The result of iterating over this changed objective function should produce a model closer to the desired solution. This more accurate model is then used as the initial model for iterating over the original objective function. Since this new initial model is close to the global minimum of the original objective function, the iteration in the original objective function must now produce an accurate solution.
    This technique of iterating an altered objective function is often referred to as multi-resolution, or multi-grid, or multi-scale inversion, 5, which is discussed below.
    There are a large number of well-known inversion methods. These methods fall into one of two categories, inversion of non-iteration and inversion of iteration. The following are definitions of what it commonly means for each of the two categories: * Non-repetitive inversion - inversion that is performed by assuming some simple background model and updating the model based on the input data. This method does not use the update model as an input for another inversion step. For seismic data, these methods are commonly referred to as imaging, migration, diffraction tomography, linear inversion, or Born inversion.
    * Repetitive inversion - inversion involving repetitive improvements in the properties of the subsurface model, just as a model is found in a way that satisfactorily explains the observed data. If the inversion converges, then the final model will be better explained and the data observed, and they will be closer to the current surface model property. Repetitive inversion usually produces a more accurate model than non-repetitive inversion, but it is much more expensive to calculate.
    Wave inversion means an inversion based on a wave simulator, such as acoustic or seismic inversion. The repetitive method most commonly used in wave inversion is the optimization of the objective function. Optimization of the objective function involves minimizing the repetitive value of, with respect to the M model, the objective function (M) which is a measure of mismatch between the calculated and observed data (this is also sometimes referred to as the function of cost). The calculated data are simulated with a computer programmed to use the physical propagation that governs the source signal in a medium represented by the current model. Simulation calculations can be done by any of several numerical methods, including, but not limited to finite differences, finite elements or radius location. After Tarantola (Tarantola, 1984), the most commonly used objective function is the function, of at least square objectives: S (M) = (u (M) -d) ”C” (u (M) -d), O where T represents the transposition operator vector and: M = the model that is a vector of N parameters [Mm ;, m>, ma], d = measured vector data (shown with respect to the source, receiver and time), u ( M) = vector of simulated data for model M (shown with respect to source, receiver and time), C = covariance matrix.
    Methods for optimizing the objective function are local or global [Fallat, et al., 1999]. Global methods are involved, simply calculating the objective function (M) for a population of models (Mi, M2, M3, ...) and selecting a set of one or more models from the population of approximately a minimum of (M). If further improvement of this newly selected set of models is desired, then they can be used as a basis to generate a new population of models, which can be re-tested for objective function (M). Global methods are more likely to converge to the correct solution than local methods, but they are very expensive to apply to large-scale inversion problems having many model parameters. Well-known global inversion methods include Monte Carlo, simulated pairing, genetic and algorithmic evolution.
    Optimization of the local objective function involves:
    1. Define the current model for the starting model,
    2. Calculate the gradient, VM5 (M), of the objective function with respect to the parameters that describe the model,
    3. Search for an update model, which is a variation of the initial model from a direction based on the gradient, which better explains the observed data.
    4. If the updated model is not accurate enough, return to step 2 using the updated model as the current model, in order to finish otherwise.
    Algorithm 1 - Algorithm for updating a model using the optimization of the objective function.
    Local inversion methods are much more efficient than global methods, and therefore are the only practical methods for applying to a large scale inversion problem. Local inversion methods of function objectives are commonly used, including a steeper descent, with conjugated variations and Newton's method.
    It should be noted that in calculating VMS (M), algorithm 1 of the second step, requires the derivative of calculating (M) with respect to each of the mu parameters of the model N When N is very large (roughly over a thousand) , this calculation can be extremely time consuming if it had to be performed for each parameter model. Fortunately, the - adjunct method can be used to efficiently perform this calculation for all model parameters at once [Tarantola, 1984]. The adjunct method for the objective function with fewer squares and a latticed or meshed model parameterization (where M is a vector with each element representing the model value in a meshed cell) is summarized by the following algorithm:
    1. Calculate the frontal simulation of the data using the current model, with M * º with k: being the current iteration, to obtain u (M ”9),
    2. Subtract the observed data from the simulated data that arise SM
    3. Calculate the reverse simulation (ie, going back in time) using C * SM) as the source, producing an adjunct u (M),
    4. Finally, VuSMO) adjunct (MO) A u M, where A represents the adjunct operator (for example, the identity for slopes with respect to components of the module M representing a mode of elasticity, or a spatial variation for variations with respect to components of M representing density). Algorithm 2 - Algorithm for calculating the variations of the square cost functions of a meshed model using the adjunct method. Optimizing the local objective function is generally much less expensive than optimizing the global objective function, but requires a more accurate starting model. This more accurate starting model is required, because the objective function often has many minimal local optimization methods that will generally find the closest to these minimums. The minimum corresponding to the true model is called the global minimum and all other minimums are called local minimums. If the starting model is no longer close to the global minimum, then a local technical optimization is likely to produce an inaccurate inverted model that will correspond to the closest local minimum. This is illustrated in fig. 1, where the objective is to invert to an M model, which has two parameters mi and m.2. The dashed outlines 110 show the values of the objective function as a function of parameters mi and tri2. The global minimum 120 is marked by a solid black circle and the two local minimums 130 and 140 are shown by two filled gray circles. The inversion starts in the initial model Mº (150) and continues through local optimization the iteration of a model M and so on for model Mº * (160). It does not matter how many more local optimization iterations are attempted for the inverted model to get closer to the local minimum 130 near MO, rather than approximately the global minimum of 120. Several methods have been proposed in an attempt to overcome this local minimum problem. As mentioned above, many of these methods involve iterating over an altered objective of the function during the first iterations of the inversion. This altered objective of the function is chosen to have fewer local minimums, but to have an overall minimum close to the global minimum of the original objective function. Hereby, the first iterations will produce a model which, although inaccurate, is significantly closer to the global minimum of the original objective function. Figure 2 illustrates a local optimization corresponding to Fig.1, but using an altered objective function having fewer local minimums. The changed objective function has an overall minimum 210 (the black solid circle) that is close to, but —not the same location as the global minimum 220 for the original objective function (the circle with diagonal streaks). From the initial model Mº 230, which is the same used in the initial model in fig. 1, two iterations using the results of the function's changed objectives in a model] vr 240. This model] JVr it can then be used as the initial model for further iterations, but now using the original objective function. This is illustrated in Fig. 3, where the iteration of model 2 starts from fig. 2 (shown in gray), which becomes 310, is used as the starting model. The iteration now converges to an M * 320 model, close to the global minimum 220 instead of close to a local minimum, as in Fig. 1. Because the starting model is more accurate than the original starting model, the inversion now interacts for the right solution. Typically, when changing the original objective function, the number of local minimums in the changed objective function is inversely related to the distance between the original global minimum and the changed objective function. Thus, it may be advantageous to repeat in a sequence of the changed function objectives starting with one having the smallest number of local minimums and the global minimum precision, proceeding through the objectives of functions that have an increasing number of local minimums and increasing precision of the global minimum, then ending up by iterating over the original objective function. Methods that perform initial iterations over altered function objectives having few local minimums are often referred to as multi-scale or multi-grid methods, and a flow diagram for this technique as illustrated in Figure 4.
    The process starts at step 410, choosing a change to optimize the original purpose of the function. This changed objective function, which depends on the data being in 420 form, is repeated in step 430 until the changed objective function is considered sufficiently minimized in step 440. (The value is lower when the selected maximum or other stop condition is met. .) When this occurs, it is determined in step 450 whether the current model inverted sufficiently minimizes the original purpose of the function. If not, the process returns to step 410 and choose between a new objective function changed or to optimize the original objective of the function. Eventually, the process ends 460.
    Two altered objectives of the function have been proposed in the literature to solve the local minimum problem in the full field of seismic wave inversion ("FWI"): + High cut filters - Bunks (. Bunks et al., 1995) describes the objectives of least squares function, applying high-cut filters in invariant time (sometimes called a low-pass filter, that is, a filter that passes frequencies below its cut-off frequency and rejects frequencies above) for both the measured data and the signature source used to calculate simulated seismic data. The high cut, ie the cut, the frequency of these filters is then increased as the inversion is repeated, with no filter being applied for the final iterations (not a filter corresponding to the original purpose of the function).
    It is well known how to design these filters, see, for example, Press et al., Numerical Recipes in FORTRAN, The Art of Scientific Calculation, Cambridge University Press (1992). They can also be obtained from sources such as Se * Un * x (see http://www.cwp.mines.edu/cwpcodes/).
    * Removal layer - Maharramov (Maharramov et al., 2007) teaches that the initial inversion iterations should be located in shallower layers, and that this depth interval should be extended as the iteration proceeds. Correspondingly, when only small depths are inverted, then only shorter data times are inverted, because a shallow model can only predict the shorter time portions of the data. In general F WI will converge to the global minimum if the starting model is accurate enough to predict transit times for any internal propagation mode having half the period in this mode. This can be called the Pratt criterion. ("In practice, with the inversion of the seismic waveform this implies that much of the waveform energy must be predicted (by the initial model) within half-wavelength of the observed waveforms; if not, a minimum mismatch model will be obtained when the predicted waveforms coincide with the wrong cycle in the observed waveforms "- Pratt, 1999).
    The present invention is an improved method for reducing the precision requirements on the starting model when performing local optimization of the objective function.
    SUMMARY OF THE INVENTION The present inventive method is applicable to any inversion based on a wave simulator, such as acoustic or seismic inversion. In one of its aspects, the invention is a specific method to alter the objective function that will, due to an imprecision given in the starting model, reduce the number of iterations necessary to find the global minimum. By reducing the number of iterations, the cost and time calculation will be correspondingly reduced. The change includes incorporating a variation of the filtering time for the objective function.
    This filter is chosen so that some statistical measure of the difference between transit times of the measured data and the calculated one is less than some fractions (usually a quarter) of the dominant data period. This implies that the filter is a high cut filter. The filter is also chosen in such a way that the high cutoff frequency of this filter decreases with the transit time increasing, making a variation in the filtering time.
    With reference to the flowchart of Fig. 6, in one embodiment of the invention the —method is a method for the inversion of measured seismic data 65 to infer a model of physical property of a subsurface region, successively comprising the model updated through the realization iteration, multi-scale, inversion of the measured seismic data using a minimal local optimization 64 of an objective function, which calculates the mismatch between the simulated seismic data model and the measured seismic data 61, in which a change in the low-pass filter is used to successively change the objective function from scale to scale by filtering the measured and simulated seismic data in the calculation of mismatch 62, the referred filter being variable with the time where the low-pass filter varies with the transit time of the seismic data, being filtered on one or more scales of the inversion. The filter can be designed using the Pratt criterion to reduce the number of local minimums, modified to involve a statistical measure (such as the mean) of the data, with respect to the source and location of the receiver, but not with respect to the time of transit, leaving the filter with a variable time 63. At the final stage of the multi-scale inversion, preferably it uses the objective function unchanged, it is thus, more efficiently, provided with a more accurate model starting model to help converge to a global minimum, resulting in an optimized physical property model 66.
    As with any inversion data, the process in practical applications is highly automated, that is, it is performed with the aid of a computer programmed in accordance with the disclosures described here.
    BRIEF DESCRIPTION OF THE DRAWINGS The present invention and its advantages will be better understood with reference to the following detailed description and the accompanying drawings in which: Fig.1 is a schematic illustration of an inversion converging to a local minimum; Fig.2 is a schematic diagram of local optimization corresponding to fig. 1, but with an altered objective function having fewer local minimums; Fig.3 illustrates a local optimization with the original objective of the function of fig. 1, but using iteration 2 of the model from fig. 2 (shown in gray) as the starting model; Fig.4 is a flow chart showing the basic steps in multi-scale optimization; Fig.5 is a schematic diagram showing the error transit time, tenor, between - measured and simulated data, and the instantaneous period, T, of a seismic trace; and Fig.6 is a flow chart showing the basic steps in an embodiment of the present invention. The invention will be described in connection with the embodiments of the example. However, insofar as the following detailed description is specific to a particular embodiment or particular use of the invention, it is intended to be illustrative only, and is not to be construed as limiting the scope of the invention. On the contrary,
    it is intended to cover all alternatives, modifications and equivalents that may be included within the scope of the invention, as defined by the attached claims.
    DETAILED DESCRIPTION OF THE EXAMPLES OF EXAMPLES Mathematically, the Pratt criterion can be defined as: max | ... (M, 5,7,1)) where tenor IS The transit time error between the measured and simulated data, and T is the instantaneous period of the measured data as illustrated in fig. 5. In fig. 5, the transit time error between the measured and simulated data is indicated as teror, € The instantaneous period of a seismic trace is indicated by T. The transit time error is the amount of time displacement required to align the data measured and simulated. The instantaneous period is the time between similar phases (for example, peaks or troughs) of the data. (Transit time means the time that elapsed from the generation of a seismic wave by the seismic source until the wave is registered at a receiver.).
    Both the transit time error and the instantaneous period are functions of the source Ss, the receiver r and the transit time t. Furthermore, terror is a function of the accuracy of the current model. In practice, Equation 2 may be more restrictive than necessary to ensure the convergence of FWl. In particular, a less rigid statistic measured than the maximum (for example, the average 15) could be used for a purpose other than 1/2 and could be used on the right side of inequality. Thus, in practice, Equation 2 can be replaced by the following: statho, (M, s, r, 1) | <a E CNH seat T (s, r, 1) 2 where stat is some statistic such as the mean, the mode or an average square error, and a is a constant that is approximately equal to one. The results are not expected to be very sensitive to the choice of statistics.
    The Bunks Alteration in the objective function follows a line of reasoning similar to the Pratt criterion, using a high frequency filter to increase the T (s, r, t), thus allowing greater tenor values. It is understood that the main cause of local minimums is the hopping cycle, and longer periods can make this less likely. In theory, tero, it could be reduced instead of limiting the data to decrease the frequency; however the only way to do this would be to have a more accurate starting model, which is very difficult and perhaps impossible. In addition, the objective of FWI is to produce a more accurate model, - requiring a much more accurate model of starting by reducing the value of the FWI. On the other hand, the Maharramov method of separation layer avoids large errors in transit time by inverting only a superficial model that propagates only unique modes, which have small transit times. These transit time errors generally increase with transit time, limiting the inversion of shorter transit times while keeping within the thumb rule.
    5 In the present invention, a proposed alternative is Equation 3 of the "rule of thumb" that will lead to a new strategy to ensure convergence. The rule of thumb is an alternative as follows: sia lromor (M, 5,7,0) | from sat (s) | ST: (4) This rule of thumb differs from Equation 3 in which statistical analysis is no longer performed with respect to time. After applying the statistical calculations, the numerator and denominator on the left side of Equation 4 are not functions of the source and locations of the receiver. Equation 4 is equivalent to: 5, 2 T (1)> =, (M1), (5) where IS (1) 2 Star T (ssno) q Tora (M, 1) = sterlCoo (Mo 5,710) Na suit practice It is an increasing function of transit time t, because transit time errors tend to accumulate like waves propagated through an inaccurate model. Equation 5 suggests that the ideal strategy for multi-scale inversion would be to change the simulated measurements and data so that the average instantaneous period of the seismic data increases with the transit times in a similar way to the fear. This can be achieved by applying a time variation of the high frequency of the filter to the measured data. The high filter frequency of this varying time interval, should decrease with increasing transit times. The advantage of this method over the Bunks technique is that more information (that is, the higher frequency information in short transit times) would be used for the first inversion iterations, thus better restricting the inverted model leading to faster convergence. The advantage of our proposal over the Maharramov extraction layer is that more of the data (ie, seismic propagation modes through deeper portions of the model) would be used in the initial iterations, leading to faster convergence.
    The suit function, in both cases, depends on how the transit time errors are measured and also on what statistical measures these transit time error measurements are applied. However, it can be expected, as proposed in the multi-scale inversion strategy, that they will not be highly sensitive to fear. In fact, instead of putting a significant effort to measure terror from the data, an alternative strategy would be to assume a simple and functional form for tenor, such as the linear function teror =
    (Mo) t where M, is the initial model. This assumed functional form would then be used to design a variation in the time of the high frequency filter satisfactory to Equation 5, and inversion using this filter would be attempted. If the inversion does not converge, then the value of B would be increased and the inversion with this more conservative estimate of having, would be attempted. This test would continue until one is verified with the convergence yields for the initial model.
    After finding a B that converges to the initial model M ,, then the iteration will produce a current inverted model, which is more accurate than the initial model. This greater precision implies that it must be reduced as the iteration proceeds. This decrease - does not imply a filter varying in time corresponding to what passes at the highest frequencies. The inversion process with the variable time filter that passes more and more high frequencies will be until all the frequencies in which the data are used at the end of the inversion iterations.
    To make practical use of the variable time of the filters in the inversion, it is necessary to be able to calculate the adjunct gradient (Algorithm 2) in a way consistent with the variable time of the filter. The most mathematically recommended way to do this is to apply the variable filter time using the inverse covariance matrix C "'in Equation 1. To do this, the inverse covariance matrix C"' would be chosen to be non-diagonal (in the dimension time), with elements equal to a temporal representation varying with the variable time coefficients of the filter. Since the filtering would take place over a variable time, the filter coefficients would vary over time. Example 1 shows an example sub matrix of C "', corresponding to a particular source and a receiver, which would implement the variable time of the filter. The first line of this sub matrix is equal to zero, except for one diagonally.
    time> 1 0 0 0 O O O O O O
    0.95. .05 OOOOO 0 OO 0 .90.10.05 QOOOO 0 OO .85 15 10.05 OOO time 2/0 OOO .80.20 15 10.05 O 0 0 OOO .5 25.20 15 10 0 0 OOOO .70 .30 25 20 0 0 0 OOOO. 65 35 30 0 0 OOOOOO .60 40 00 0 0 0 0 0 0 055) Example 1
    This implies that this particular variation in the filtering time does not perform zero time filtering. The elements outside the diagonal increase for increasing lines in the sub-matrix, which implies that the high cutoff frequency of this time interval of the filtering variation decreases with increasing time. Note that the transit time error can be seen as a function of the source or the receiver. In this case, this method could be applied in a more general way than just for a simple time variation of filtration. For example:
    1. Transit time error is often not just a function of transit time, but is also often a function of the displacement between the source and the receiver.
    In this case, it would be advantageous to use different variations in the filtering time for credits from the receiver from different sources. As the filter iteration is relaxed as to the accuracy of the model increases.
    2. The transit time error is often a function of the immersion time of the data in the source and / or in the domain of the receiver. This can happen due to steep time dives corresponding to waves that travel in the predominantly horizontal direction, which are more sensitive to the accuracy of the initial model. In this case, the time variation in the immersion of the filters (for example, the frequency of the wave number filter, which removes seismic events with a high decrease time) could be used instead of temporal variables in the filtering time. As the iteration decreases in the filter it should be relaxed as to the accuracy of the model's increments.
    In any case, the filter (for example, a space and the variable time of the filter, decrease of the variable time variable etc.) can be implemented in the covariance matrix C "'as explained above, and the gradient calculation then also proceed as described above.
    A preferred approach may be that if teror (M, t) can be estimated for the current M model, then the variable time filter must be designed to be consistent with Equation 5. Otherwise, it is reasonable to estimate that teror (M , t) is an adjustment of the linear function, using an initial estimate for the EB value. Again, the variable time filter must be designed to be consistent with both an estimate of this É and Equation 5. simultaneously. If this inversion converges then the process is completed. If the inversion does not converge, then É would be increased and another inversion would be attempted. This process of increasing B would continue until reaching convergence.
    In practice, the matrix C "'representing the variable time filter would be very scarce and, therefore, its application to residual data in step 3 of algorithm 2 would be better performed by application of a variable time filter operator, instead of a multiplication of matrices. In fact, this inversion method is probably not strongly sensitive to the method used to implement the variable time filter. Therefore, the variable time filter could be more efficiently implemented as an invariable window time filter. in other words, the data would be separated into time windows and then different invariable time filters would be applied to different windows.
    The foregoing description is directed to particular embodiments of the present invention for the purpose of illustrating it. It will be apparent, however, to one skilled in the art, that many modifications and variations of the embodiments described herein are possible. For example, the inventive method is not limited to seismic data, and can be applied to any data where multiscale inversion is used to avoid local minimums. All such modifications and variations are intended to be within the scope of the present invention, as defined by the appended claims.
    REFERENCES Bunks, C, F.M. Saleck, S. Zaleski, G. Chaven, 1995, "Multiscale seismic inversion in waveform", Geophysics, 60, pp 1457-1473.
    Maharramov, M., U. Albertin, 2007, "Image difference tomography located in wave equation 5," SEG Summaries of the expanded abstract annual meeting ", San Antonio, 2007, PP, 3009-3013.
    Fallat, M.R., Dosso, SE, "Geo-acoustic inversion via local, global, and hybrid algorithms," Journal of the Acoustical Society of America 105, pp. 219-3230 (1999).
    Pratt, RG, "Inversion of seismic waveform in the frequency domain, Part 1: Theory and verification in a physical scale model", Geophysics 64, PP. 888-901 (1999).
    Tarantola, A., "Data inversion of seismic reflections in acoustic approaches, Geophysics 49, pp 1259-1266 (1984).
    权利要求:
    Claims (23)
    [1]
    1. Method for inversion of measured seismic data to infer a model of physical property of an underground region, successively comprising updating the model through the iterative multi-scale inversion of seismic data - measured without a computer, is characterized by the fact that it aims to objective is a function that computes the mismatch between a simulation model of seismic data and measured seismic data, in which a low filter change is then called a filter, and is used to successively change the objective scale function for filtering scale of seismic data measured and simulated in the computed mismatch calculation, the referred filter varying in time when the low frequency cut decreases and varies with the travel time of seismic data being filtered in one or more scales of the inversion.
    [2]
    2. Method according to claim 1, characterized by the fact that the cutoff frequency of the low-pass filters decreases with increasing duration.
    [3]
    3. Method according to claim 1, characterized by the fact that each objective function changed successively corresponds to the filter being changed to reject less data.
    [4]
    4. Method according to claim 3, characterized by the fact that the objective function is changed successively until a final scale is reached in which the objective function is unchanged and in which the iteration continues until a convergence criterion is met, or another stop condition is encountered.
    [5]
    5. Method according to claim 4, characterized by the fact that the filter corresponding to the final scale transmits all data and does not reject any.
    [6]
    6. Method according to claim 3, characterized by the fact that the filter is changed to reject less data by increasing the cutoff frequency of the low-pass filter.
    [7]
    7. Method according to claim 1, characterized by the fact that one or more iterations are performed on each scale of the multi-scale inversion, that is, for each change of high-pass filter and alteration of the objective function.
    [8]
    8. Method according to claim 1, characterized by the fact that the minimum local optimization of the objective function comprises the calculation of the gradient of the objective function, with respect to one or more parameters of a current physical property model, then , look for an updated physical property model, that is, a disturbance of the current physical property model in a gradient-based direction, which better explains the measured seismic data.
    [9]
    9. Method according to claim 8, characterized by the fact that the gradient is calculated by an adjunct method.
    [10]
    10. Method according to claim 1, characterized by the fact that the inversion is a total inversion in the wave field.
    [11]
    11. Method according to claim 2, characterized by the fact that to ensure that the iterations converge to a global minimum of the objective function, the following criterion is used to design a variable low-pass filter over time: the model when iteration initiation must be sufficiently accurate to predict the duration of any propagation mode within half a period of the mode, after applying the variable filter over time.
    [12]
    12. Method according to claim 11, characterized by the fact that the said convergence criterion can be mathematically expressed as: stato, (M, sr) 2 statT (s, r, 1) "2 where M represents the model of physical property, fear É The duration error between the measured and simulated seismic data, T is the instantaneous period of the measured data, a is a selected constant, t is the duration of the seismic wave, s represents the coordinates of the sources seismic, r represents the coordinates of the seismic receiver, and stat means the mean or other measure, to reduce the variability of being for constants.
    [13]
    13. Method according to claim 12, characterized by the fact that a is a number within the range 1/2 to 1.
    [14]
    14. Method according to claim 12, characterized by the fact that the rate of decrease in the low-pass cutoff frequency with increasing duration time is determined according to the convergence criterion.
    [15]
    15. Method according to claim 1, characterized by the fact that it also comprises using different time-varying filters for different displacements of source receivers.
    [16]
    16. Method according to claim 9, characterized by the fact that the filter is implemented using the inverse c ”variance matrix in Equation 1.
    [17]
    17. Method according to claim 16, characterized by the fact that the inverse c ”variance matrix C” * is chosen to be non-diagonal in the dimension of time, with elements equal to a temporal representation of the coefficients of the filter variable with time, and said coefficients varying with time.
    [18]
    18. Method according to claim 12, characterized by the fact that fear = state o (M, 5,7.) Is assumed to be a linear function of the duration time of the form Ferr = 6E (M,) :, in that M, is an initial choice for the model when the iteration starts, and B is a selected constant, in which the iterative inversion does not converge using this filter design, so the value of B would be increased and the inversion with this more conservative estimate de Íorror would be attempted, and so on, until a B is found that produces a convergence to the initial model.
    [19]
    19. Method according to claim 1, characterized by the fact that the inversion is a total inversion of the wave field
    [20]
    20. Method according to claim 1, characterized by the fact that a computer is programmed to perform at least some of the steps of the method.
    [21]
    21. Computer program product, characterized by the fact that it comprises a usable computer medium having a readable computer program code incorporated therein, said readable computer program code adapted to be executed to implement a method for inversion total wave field of measured data to infer a model of a propagation medium for the wave field, said method comprising: successively updating the model by iterative multiscale inversion performance of the measured data using a minimal local optimization of an objective function , which —calculates the adjustment between the simulated data model and the measured data; in which a change in the low-pass filter, hereinafter referred to as the filter, is used to successively change the objective function from scale to scale by filtering the measured and simulated data in the maladjusted calculation, said filter being variable with time; in which the cutoff frequency of the low-pass filter varies with the duration of the wave field through the medium corresponding to the data being filtered in one or more scales of the inversion.
    [22]
    22. Computer program product according to claim 21, characterized by the fact that the cut-off frequency of the low-pass filter decreases with increasing duration.
    [23]
    23. Computer program product according to claim 21, characterized by the fact that the data being inverted is seismic data, and the propagation medium is a subsurface region of the earth.
    类似技术:
    公开号 | 公开日 | 专利标题
    BR112012019562A2|2020-08-18|inversion of a full wave field using time-varying filters
    Mons et al.2016|Reconstruction of unsteady viscous flows using data assimilation schemes
    EP2389602B1|2019-08-14|Stochastic inversion of geophysical data for estimating earth model parameters
    RU2573174C2|2016-01-20|Artefact reduction in iterative inversion of geophysical data
    US10845493B2|2020-11-24|Model compression
    Stuart et al.2016|A two-stage Markov chain Monte Carlo method for velocity estimation and uncertainty quantification
    Liu et al.2017|Effects of conjugate gradient methods and step-length formulas on the multiscale full waveform inversion in time domain: Numerical experiments
    WO2015042386A1|2015-03-26|Eikonal solver for quasi p-waves in anisotropic media
    Rey et al.2019|Quantifying uncertainties in contact mechanics of rough surfaces using the multilevel Monte Carlo method
    Glowinski et al.2015|A penalization-regularization-operator splitting method for eikonal based traveltime tomography
    Cha et al.2010|Two-dimensional Laplace-domain waveform inversion using adaptive meshes: an experience of the 2004 BP velocity-analysis benchmark data set
    Griffiths2013|Kelvin wave propagation along straight boundaries in C-grid finite-difference models
    Dubrule2018|Kriging, splines, conditional simulation, Bayesian inversion and ensemble Kalman filtering
    Shaw et al.2020|Stochastic Galerkin finite volume shallow flow model: well-balanced treatment over uncertain topography
    BR102016024241B1|2022-01-25|Method for modeling a geophysical system and method for determining physical properties
    Gottschalk et al.2020|Coarsening in Algebraic Multigrid using Gaussian Processes
    Sochala et al.2020|Model reduction for large-scale earthquake simulation in an uncertain 3D medium
    Fraysse2012|Numerical Error Prediction and its Applications in CFD using tau-estimation
    US11255995B2|2022-02-22|Methods and systems for determining subsurface feature using broadband full waveform inversion
    Gottschalk et al.2021|Coarsening in algebraic multigrid using Gaussian processes
    Symes2012|Transfer of approximation and numerical homogenization of hyperbolic boundary value problems with a continuum of scales
    JP2022001861A|2022-01-06|Full Waveform Inversion Method, Equipment and Electronics
    Fernández2016|Automated calibration for numerical models of riverflow
    Dunbar et al.2018|Binary recovery via phase field regularization for first traveltime tomography
    CN109946518A|2019-06-28|Electric harmonic signal analysis method and analytical equipment based on bayes method
    同族专利:
    公开号 | 公开日
    AU2011233680A1|2012-10-11|
    EA028104B1|2017-10-31|
    SG183794A1|2012-10-30|
    EP2553601A1|2013-02-06|
    US20110238390A1|2011-09-29|
    CA2789714A1|2011-10-06|
    KR101797450B1|2017-11-14|
    EP2553601B1|2019-05-01|
    CN102918521B|2016-05-18|
    US8223587B2|2012-07-17|
    EA201290773A1|2013-03-29|
    CN102918521A|2013-02-06|
    CA2789714C|2016-07-26|
    KR20130054236A|2013-05-24|
    WO2011123197A1|2011-10-06|
    MY156075A|2016-01-15|
    EP2553601A4|2017-11-22|
    AU2011233680B2|2016-01-07|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US3812457A|1969-11-17|1974-05-21|Shell Oil Co|Seismic exploration method|
    US3864667A|1970-09-11|1975-02-04|Continental Oil Co|Apparatus for surface wave parameter determination|
    US3984805A|1973-10-18|1976-10-05|Daniel Silverman|Parallel operation of seismic vibrators without phase control|
    US4168485A|1974-08-12|1979-09-18|Continental Oil Company|Simultaneous use of pseudo-random control signals in vibrational exploration methods|
    US4041443A|1976-06-01|1977-08-09|Western Geophysical Co. Of America|Seismic recording apparatus having a time-varying sample|
    US4675851A|1982-09-09|1987-06-23|Western Geophysical Co.|Method for seismic exploration|
    US4545039A|1982-09-09|1985-10-01|Western Geophysical Co. Of America|Methods for seismic exploration|
    US4575830A|1982-10-15|1986-03-11|Schlumberger Technology Corporation|Indirect shearwave determination|
    US4594662A|1982-11-12|1986-06-10|Schlumberger Technology Corporation|Diffraction tomography systems and methods with fixed detector arrays|
    US4562540A|1982-11-12|1985-12-31|Schlumberger Technology Corporation|Diffraction tomography system and methods|
    FR2543306B1|1983-03-23|1985-07-26|Elf Aquitaine|METHOD AND DEVICE FOR OPTIMIZING SEISMIC DATA|
    JPS59189278A|1983-03-23|1984-10-26|Hashimoto Denki Co Ltd|Wicket type plate drier|
    JPS606032A|1983-06-22|1985-01-12|Honda Motor Co Ltd|Control method of operating condition of internal- combustion engine|
    US4924390A|1985-03-04|1990-05-08|Conoco, Inc.|Method for determination of earth stratum elastic parameters using seismic energy|
    FR2589587B1|1985-10-30|1988-02-05|Inst Francais Du Petrole|MARINE SEISMIC PROSPECTION METHOD USING A CODE VIBRATORY SIGNAL AND DEVICE FOR IMPLEMENTING SAME|
    US4707812A|1985-12-09|1987-11-17|Atlantic Richfield Company|Method of suppressing vibration seismic signal correlation noise|
    US4823326A|1986-07-21|1989-04-18|The Standard Oil Company|Seismic data acquisition technique having superposed signals|
    US4686654A|1986-07-31|1987-08-11|Western Geophysical Company Of America|Method for generating orthogonal sweep signals|
    US4715020A|1986-10-29|1987-12-22|Western Atlas International, Inc.|Simultaneous performance of multiple seismic vibratory surveys|
    US4766574A|1987-03-31|1988-08-23|Amoco Corporation|Method for depth imaging multicomponent seismic data|
    CN1016999B|1987-11-01|1992-06-10|大庆石油管理局地球物理勘探公司|Hydrocarbons detecting method of prospecting earthquake|
    US4953657A|1987-11-30|1990-09-04|Halliburton Geophysical Services, Inc.|Time delay source coding|
    US4969129A|1989-09-20|1990-11-06|Texaco Inc.|Coding seismic sources|
    US4982374A|1989-10-23|1991-01-01|Halliburton Geophysical Services, Inc.|Method of source coding and harmonic cancellation for vibrational geophysical survey sources|
    GB9011836D0|1990-05-25|1990-07-18|Mason Iain M|Seismic surveying|
    US5469062A|1994-03-11|1995-11-21|Baker Hughes, Inc.|Multiple depths and frequencies for simultaneous inversion of electromagnetic borehole measurements|
    GB2322704B|1994-07-07|1998-12-09|Geco As|Method of Processing seismic data|
    US5583825A|1994-09-02|1996-12-10|Exxon Production Research Company|Method for deriving reservoir lithology and fluid content from pre-stack inversion of seismic data|
    WO1996033425A1|1995-04-18|1996-10-24|Western Atlas International, Inc.|Uniform subsurface coverage at steep dips|
    US5924049A|1995-04-18|1999-07-13|Western Atlas International, Inc.|Methods for acquiring and processing seismic data|
    US5721710A|1995-09-29|1998-02-24|Atlantic Richfield Company|High fidelity vibratory source seismic method with source separation|
    US5719821A|1995-09-29|1998-02-17|Atlantic Richfield Company|Method and apparatus for source separation of seismic vibratory signals|
    US5715213A|1995-11-13|1998-02-03|Mobil Oil Corporation|High fidelity vibratory source seismic method using a plurality of vibrator sources|
    US5790473A|1995-11-13|1998-08-04|Mobil Oil Corporation|High fidelity vibratory source seismic method for use in vertical seismic profile data gathering with a plurality of vibratory seismic energy sources|
    US5822269A|1995-11-13|1998-10-13|Mobil Oil Corporation|Method for separation of a plurality of vibratory seismic energy source signals|
    US5838634A|1996-04-04|1998-11-17|Exxon Production Research Company|Method of generating 3-D geologic models incorporating geologic and geophysical constraints|
    US5798982A|1996-04-29|1998-08-25|The Trustees Of Columbia University In The City Of New York|Method for inverting reflection trace data from 3-D and 4-D seismic surveys and identifying subsurface fluid and pathways in and among hydrocarbon reservoirs based on impedance models|
    GB9612471D0|1996-06-14|1996-08-14|Geco As|Method and apparatus for multiple seismic vibratory surveys|
    US5878372A|1997-03-04|1999-03-02|Western Atlas International, Inc.|Method for simultaneous inversion processing of well log data using a plurality of earth models|
    US6014342A|1997-03-21|2000-01-11|Tomo Seis, Inc.|Method of evaluating a subsurface region using gather sensitive data discrimination|
    US5999489A|1997-03-21|1999-12-07|Tomoseis Inc.|High vertical resolution crosswell seismic imaging|
    US5920828A|1997-06-02|1999-07-06|Baker Hughes Incorporated|Quality control seismic data processing system|
    US5920838A|1997-06-02|1999-07-06|Carnegie Mellon University|Reading and pronunciation tutor|
    FR2765692B1|1997-07-04|1999-09-10|Inst Francais Du Petrole|METHOD FOR 3D MODELING THE IMPEDANCE OF A HETEROGENEOUS MEDIUM|
    GB2329043B|1997-09-05|2000-04-26|Geco As|Method of determining the response caused by model alterations in seismic simulations|
    US5999488A|1998-04-27|1999-12-07|Phillips Petroleum Company|Method and apparatus for migration by finite differences|
    US6219621B1|1998-06-30|2001-04-17|Exxonmobil Upstream Research Co.|Sparse hyperbolic inversion of seismic data|
    US6388947B1|1998-09-14|2002-05-14|Tomoseis, Inc.|Multi-crosswell profile 3D imaging and method|
    US6574564B2|1998-10-01|2003-06-03|Institut Francais Du Petrole|3D prestack seismic data migration method|
    FR2784195B1|1998-10-01|2000-11-17|Inst Francais Du Petrole|METHOD FOR PERFORMING IN 3D BEFORE SUMMATION, A MIGRATION OF SEISMIC DATA|
    US6225803B1|1998-10-29|2001-05-01|Baker Hughes Incorporated|NMR log processing using wavelet filter and iterative inversion|
    US6021094A|1998-12-03|2000-02-01|Sandia Corporation|Method of migrating seismic records|
    US6754588B2|1999-01-29|2004-06-22|Platte River Associates, Inc.|Method of predicting three-dimensional stratigraphy using inverse optimization techniques|
    AU3229900A|1999-02-12|2000-08-29|Prange, Michael|Uncertainty constrained subsurface modeling|
    US6826486B1|2000-02-11|2004-11-30|Schlumberger Technology Corporation|Methods and apparatus for predicting pore and fracture pressures of a subsurface formation|
    US6058073A|1999-03-30|2000-05-02|Atlantic Richfield Company|Elastic impedance estimation for inversion of far offset seismic sections|
    FR2792419B1|1999-04-16|2001-09-07|Inst Francais Du Petrole|METHOD FOR OBTAINING AN OPTIMAL MODEL OF A PHYSICAL CHARACTERISTICS IN A HETEROGENEOUS ENVIRONMENT, SUCH AS THE BASEMENT|
    GB9927395D0|1999-05-19|2000-01-19|Schlumberger Holdings|Improved seismic data acquisition method|
    US6327537B1|1999-07-19|2001-12-04|Luc T. Ikelle|Multi-shooting approach to seismic modeling and acquisition|
    FR2798197B1|1999-09-02|2001-10-05|Inst Francais Du Petrole|METHOD FOR FORMING A MODEL OF A GEOLOGICAL FORMATION, CONSTRAINED BY DYNAMIC AND STATIC DATA|
    EP1746443B1|1999-10-22|2014-01-08|Fugro N.V.|Method of estimating elastic parameters and rock composition of underground formations using seismic data|
    US6480790B1|1999-10-29|2002-11-12|Exxonmobil Upstream Research Company|Process for constructing three-dimensional geologic models having adjustable geologic interfaces|
    FR2800473B1|1999-10-29|2001-11-30|Inst Francais Du Petrole|METHOD FOR 2D OR 3D MODELING A HETEROGENEOUS MEDIUM SUCH AS THE BASEMENT DESCRIBED BY SEVERAL PHYSICAL PARAMETERS|
    WO2001053853A1|2000-01-21|2001-07-26|Schlumberger Holdings Limited|System and method for estimating seismic material properties|
    EP1254383B1|2000-01-21|2005-08-24|Schlumberger Holdings Limited|System and method for seismic wavefield separation|
    FR2805051B1|2000-02-14|2002-12-06|Geophysique Cie Gle|METHOD OF SEISMIC MONITORING OF AN UNDERGROUND AREA BY SIMULTANEOUS USE OF MULTIPLE VIBROISISMIC SOURCES|
    GB2359363B|2000-02-15|2002-04-03|Geco Prakla|Processing simultaneous vibratory seismic data|
    US6687659B1|2000-03-24|2004-02-03|Conocophillips Company|Method and apparatus for absorbing boundary conditions in numerical finite-difference acoustic applications|
    WO2002047011A1|2000-12-08|2002-06-13|Ortoleva Peter J|Methods for modeling multi-dimensional domains using information theory to resolve gaps in data and in theories|
    US6317695B1|2000-03-30|2001-11-13|Nutec Sciences, Inc.|Seismic data processing method|
    CN1254693C|2000-10-17|2006-05-03|维斯特恩格科地震控股有限公司|Method of using cascaded sweeps for source coding and harmonic cancellation|
    FR2818753B1|2000-12-21|2003-03-21|Inst Francais Du Petrole|METHOD AND DEVICE FOR SEISMIC PROSPECTION BY SIMULTANEOUS EMISSION OF SEISMISK SIGNALS OBTAINED BY CODING A SIGNAL BY RANDOM PSEUDO SEQUENCES|
    FR2821677B1|2001-03-05|2004-04-30|Geophysique Cie Gle|IMPROVEMENTS TO TOMOGRAPHIC INVERSION PROCESSES OF POINTED EVENTS ON MIGREE SEISMIC DATA|
    US6751558B2|2001-03-13|2004-06-15|Conoco Inc.|Method and process for prediction of subsurface fluid and rock pressures in the earth|
    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|
    US6882958B2|2001-06-28|2005-04-19|National Instruments Corporation|System and method for curve fitting using randomized techniques|
    GB2379013B|2001-08-07|2005-04-20|Abb Offshore Systems Ltd|Microseismic signal processing|
    US6593746B2|2001-08-27|2003-07-15|Larry G. Stolarczyk|Method and system for radio-imaging underground geologic structures|
    US6927698B2|2001-08-27|2005-08-09|Larry G. Stolarczyk|Shuttle-in receiver for radio-imaging underground geologic structures|
    US7672824B2|2001-12-10|2010-03-02|Westerngeco L.L.C.|Method for shallow water flow detection|
    US7069149B2|2001-12-14|2006-06-27|Chevron U.S.A. Inc.|Process for interpreting faults from a fault-enhanced 3-dimensional seismic attribute volume|
    US7330799B2|2001-12-21|2008-02-12|Société de commercialisation des produits de la recherche appliquée-Socpra Sciences et Génie s.e.c.|Method and algorithm for using surface waves|
    US6842701B2|2002-02-25|2005-01-11|Westerngeco L.L.C.|Method of noise removal for cascaded sweep data|
    GB2387226C|2002-04-06|2008-05-12|Westerngeco Ltd|A method of seismic surveying|
    FR2839368B1|2002-05-06|2004-10-01|Total Fina Elf S A|METHOD OF DECIMATING SEISMIC TRACES PILOTED BY THE SEISMIC PATH|
    US6832159B2|2002-07-11|2004-12-14|Schlumberger Technology Corporation|Intelligent diagnosis of environmental influence on well logs with model-based inversion|
    FR2843202B1|2002-08-05|2004-09-10|Inst Francais Du Petrole|METHOD FOR FORMING A REPRESENTATIVE MODEL OF THE DISTRIBUTION OF A PHYSICAL QUANTITY IN AN UNDERGROUND AREA, FREE OF THE EFFECT OF CORRECTED NOISES BINDING EXPLORATION DATA|
    AU2003279870A1|2002-10-04|2004-05-04|Paradigm Geophysical Corporation|Method and system for limited frequency seismic imaging|
    GB2396448B|2002-12-21|2005-03-02|Schlumberger Holdings|System and method for representing and processing and modeling subterranean surfaces|
    US7027927B2|2002-12-23|2006-04-11|Schlumberger Technology Corporation|Methods for determining formation and borehole parameters using fresnel volume tomography|
    US6735527B1|2003-02-26|2004-05-11|Landmark Graphics Corporation|3-D prestack/poststack multiple prediction|
    US6999880B2|2003-03-18|2006-02-14|The Regents Of The University Of California|Source-independent full waveform inversion of seismic data|
    WO2004095072A2|2003-03-27|2004-11-04|Exxonmobil Upstream Research Company|Method to convert seismic traces into petrophysical property logs|
    US7072767B2|2003-04-01|2006-07-04|Conocophillips Company|Simultaneous inversion for source wavelet and AVO parameters from prestack seismic data|
    AT535826T|2003-04-01|2011-12-15|Exxonmobil Upstream Res Co|SHAPED HIGH FREQUENCY VIBRATION SOURCE|
    NO322089B1|2003-04-09|2006-08-14|Norsar V Daglig Leder|Procedure for simulating local preamp deep-migrated seismic images|
    GB2400438B|2003-04-11|2005-06-01|Westerngeco Ltd|Determination of waveguide parameters|
    US6970397B2|2003-07-09|2005-11-29|Gas Technology Institute|Determination of fluid properties of earth formations using stochastic inversion|
    US6882938B2|2003-07-30|2005-04-19|Pgs Americas, Inc.|Method for separating seismic signals from two or more distinct sources|
    US6944546B2|2003-10-01|2005-09-13|Halliburton Energy Services, Inc.|Method and apparatus for inversion processing of well logging data in a selected pattern space|
    US6901333B2|2003-10-27|2005-05-31|Fugro N.V.|Method and device for the generation and application of anisotropic elastic parameters|
    US7046581B2|2003-12-01|2006-05-16|Shell Oil Company|Well-to-well tomography|
    US20050128874A1|2003-12-15|2005-06-16|Chevron U.S.A. Inc.|Methods for acquiring and processing seismic data from quasi-simultaneously activated translating energy sources|
    FR2872584B1|2004-06-30|2006-08-11|Inst Francais Du Petrole|METHOD FOR SIMULATING THE SEDIMENT DEPOSITION IN A BASIN RESPECTING THE SEDIMENT SEQUENCE THICKNESS|
    US7646924B2|2004-08-09|2010-01-12|David Leigh Donoho|Method and apparatus for compressed sensing|
    US7480206B2|2004-09-13|2009-01-20|Chevron U.S.A. Inc.|Methods for earth modeling and seismic imaging using interactive and selective updating|
    GB2422433B|2004-12-21|2008-03-19|Sondex Wireline Ltd|Method and apparatus for determining the permeability of earth formations|
    US7373251B2|2004-12-22|2008-05-13|Marathon Oil Company|Method for predicting quantitative values of a rock or fluid property in a reservoir using seismic data|
    US7230879B2|2005-02-12|2007-06-12|Chevron U.S.A. Inc.|Method and apparatus for true relative amplitude correction of seismic data for normal moveout stretch effects|
    WO2006090374A2|2005-02-22|2006-08-31|Paradigm Geophysical Ltd.|Multiple suppression in angle domain time and depth migration|
    US7840625B2|2005-04-07|2010-11-23|California Institute Of Technology|Methods for performing fast discrete curvelet transforms of data|
    US7271747B2|2005-05-10|2007-09-18|Rice University|Method and apparatus for distributed compressed sensing|
    US7376517B2|2005-05-13|2008-05-20|Chevron U.S.A. Inc.|Method for estimation of interval seismic quality factor|
    US20090164756A1|2005-10-18|2009-06-25|Tor Dokken|Geological Response Data Imaging With Stream Processors|
    AU2006235820B2|2005-11-04|2008-10-23|Westerngeco Seismic Holdings Limited|3D pre-stack full waveform inversion|
    FR2895091B1|2005-12-21|2008-02-22|Inst Francais Du Petrole|METHOD FOR UPDATING A GEOLOGICAL MODEL WITH SEISMIC DATA|
    GB2436626B|2006-03-28|2008-08-06|Westerngeco Seismic Holdings|Method of evaluating the interaction between a wavefield and a solid body|
    US7620534B2|2006-04-28|2009-11-17|Saudi Aramco|Sound enabling computerized system for real time reservoir model calibration using field surveillance data|
    US20070274155A1|2006-05-25|2007-11-29|Ikelle Luc T|Coding and Decoding: Seismic Data Modeling, Acquisition and Processing|
    US7725266B2|2006-05-31|2010-05-25|Bp Corporation North America Inc.|System and method for 3D frequency domain waveform inversion based on 3D time-domain forward modeling|
    US7599798B2|2006-09-11|2009-10-06|Westerngeco L.L.C.|Migrating composite seismic response data to produce a representation of a seismic volume|
    CN101681394B|2006-09-28|2012-11-21|埃克森美孚上游研究公司|Iterative inversion of data from simultaneous geophysical sources|
    AT543109T|2007-01-20|2012-02-15|Spectraseis Ag|TIME REVERSE RESERVOIR LOCATION|
    JP2009063942A|2007-09-10|2009-03-26|Sumitomo Electric Ind Ltd|Far-infrared camera lens, lens unit, and imaging apparatus|
    US20090070042A1|2007-09-11|2009-03-12|Richard Birchwood|Joint inversion of borehole acoustic radial profiles for in situ stresses as well as third-order nonlinear dynamic moduli, linear dynamic elastic moduli, and static elastic moduli in an isotropically stressed reference state|
    US20090083006A1|2007-09-20|2009-03-26|Randall Mackie|Methods and apparatus for three-dimensional inversion of electromagnetic data|
    US20090164186A1|2007-12-20|2009-06-25|Bhp Billiton Innovation Pty Ltd.|Method for determining improved estimates of properties of a model|
    US8577660B2|2008-01-23|2013-11-05|Schlumberger Technology Corporation|Three-dimensional mechanical earth modeling|
    ES2651923T3|2008-03-21|2018-01-30|Exxonmobil Upstream Research Company|An efficient method for geophysical data inversion|
    EP2105765A1|2008-03-28|2009-09-30|Schlumberger Holdings Limited|Simultaneous inversion of induction data for dielectric permittivity and electric conductivity|
    US8494777B2|2008-04-09|2013-07-23|Schlumberger Technology Corporation|Continuous microseismic mapping for real-time 3D event detection and location|
    US8345510B2|2008-06-02|2013-01-01|Pgs Geophysical As|Method for aquiring and processing marine seismic data to extract and constructively use the up-going and down-going wave-fields emitted by the source|
    WO2010019957A1|2008-08-15|2010-02-18|Bp Corporation North America Inc.|Method for separating independent simultaneous sources|
    US20100054082A1|2008-08-29|2010-03-04|Acceleware Corp.|Reverse-time depth migration with reduced memory requirements|
    US20100142316A1|2008-12-07|2010-06-10|Henk Keers|Using waveform inversion to determine properties of a subsurface medium|KR101167715B1|2010-04-30|2012-07-20|서울대학교산학협력단|Apparatus and method for seismic imaging using waveform inversion solved by Conjugate Gradient Least Square Method|
    US8694299B2|2010-05-07|2014-04-08|Exxonmobil Upstream Research Company|Artifact reduction in iterative inversion of geophysical data|
    US8756042B2|2010-05-19|2014-06-17|Exxonmobile Upstream Research Company|Method and system for checkpointing during simulations|
    WO2012041834A1|2010-09-28|2012-04-05|Shell Internationale Research Maatschappij B.V.|Earth model estimation through an acoustic full waveform inversion of seismic data|
    US9235761B2|2010-11-17|2016-01-12|The Boeing Company|Zero phase real time filtering with a recurrence matrix|
    US9921187B2|2011-01-20|2018-03-20|Northeastern University|Real-time pavement profile sensing system using air-coupled surface wave|
    US8614930B2|2011-03-23|2013-12-24|Chevron U.S.A. Inc.|System and method for seismic data modeling and migration|
    US8892413B2|2011-03-30|2014-11-18|Exxonmobil Upstream Research Company|Convergence rate of full wavefield inversion using spectral shaping|
    US9158018B2|2011-04-05|2015-10-13|Westerngeco L.L.C.|Waveform inversion using a response of forward modeling|
    US9176930B2|2011-11-29|2015-11-03|Exxonmobil Upstream Research Company|Methods for approximating hessian times vector operation in full wavefield inversion|
    US10012745B2|2012-03-08|2018-07-03|Exxonmobil Upstream Research Company|Orthogonal source and receiver encoding|
    US20130311149A1|2012-05-17|2013-11-21|Yaxun Tang|Tomographically Enhanced Full Wavefield Inversion|
    JP2014063551A|2012-09-21|2014-04-10|Toshiba Corp|Semiconductor memory device|
    EP2926170A4|2012-11-28|2016-07-13|Exxonmobil Upstream Res Co|Reflection seismic data q tomography|
    US9103935B2|2013-02-04|2015-08-11|King Fahd University Of Petroleum And Minerals|Method of first arrival picking of seismic refraction data|
    US9823369B2|2013-02-28|2017-11-21|Cgg Services Sas|System and method for correcting near surface statics by using internal multiples prediction|
    CN103207409B|2013-04-17|2016-01-06|中国海洋石油总公司|A kind of frequency domain full-waveform inversion seismic velocity modeling method|
    EP3004938A1|2013-05-24|2016-04-13|Exxonmobil Upstream Research Company|Multi-parameter inversion through offset dependent elastic fwi|
    US10459117B2|2013-06-03|2019-10-29|Exxonmobil Upstream Research Company|Extended subspace method for cross-talk mitigation in multi-parameter inversion|
    US9702998B2|2013-07-08|2017-07-11|Exxonmobil Upstream Research Company|Full-wavefield inversion of primaries and multiples in marine environment|
    BR112015030104A2|2013-08-23|2017-07-25|Exxonmobil Upstream Res Co|simultaneous acquisition during both seismic acquisition and seismic inversion|
    US10036818B2|2013-09-06|2018-07-31|Exxonmobil Upstream Research Company|Accelerating full wavefield inversion with nonstationary point-spread functions|
    GB2509223B|2013-10-29|2015-03-18|Imp Innovations Ltd|Method of, and apparatus for, full waveform inversion|
    US11175421B2|2014-01-10|2021-11-16|Cgg Services Sas|Device and method for mitigating cycle-skipping in full waveform inversion|
    US10345470B2|2014-01-13|2019-07-09|Cgg Services Sas|Device and method for deghosting seismic data using sparse tau-p inversion|
    US9910189B2|2014-04-09|2018-03-06|Exxonmobil Upstream Research Company|Method for fast line search in frequency domain FWI|
    EP3140675A1|2014-05-09|2017-03-15|Exxonmobil Upstream Research Company|Efficient line search methods for multi-parameter full wavefield inversion|
    US10185046B2|2014-06-09|2019-01-22|Exxonmobil Upstream Research Company|Method for temporal dispersion correction for seismic simulation, RTM and FWI|
    SG11201608179PA|2014-06-17|2017-01-27|Exxonmobil Upstream Res Co|Fast viscoacoustic and viscoelastic full-wavefield inversion|
    US10838092B2|2014-07-24|2020-11-17|Exxonmobil Upstream Research Company|Estimating multiple subsurface parameters by cascaded inversion of wavefield components|
    US10422899B2|2014-07-30|2019-09-24|Exxonmobil Upstream Research Company|Harmonic encoding for FWI|
    US10386511B2|2014-10-03|2019-08-20|Exxonmobil Upstream Research Company|Seismic survey design using full wavefield inversion|
    CA2961572C|2014-10-20|2019-07-02|Exxonmobil Upstream Research Company|Velocity tomography using property scans|
    EP3209859B1|2014-10-24|2021-04-28|Schlumberger Technology B.V.|Travel-time objective function for full waveform inversion|
    US11163092B2|2014-12-18|2021-11-02|Exxonmobil Upstream Research Company|Scalable scheduling of parallel iterative seismic jobs|
    US10520618B2|2015-02-04|2019-12-31|ExxohnMobil Upstream Research Company|Poynting vector minimal reflection boundary conditions|
    US10317546B2|2015-02-13|2019-06-11|Exxonmobil Upstream Research Company|Efficient and stable absorbing boundary condition in finite-difference calculations|
    MX2017007988A|2015-02-17|2017-09-29|Exxonmobil Upstream Res Co|Multistage full wavefield inversion process that generates a multiple free data set.|
    GB2538807B|2015-05-29|2019-05-15|Sub Salt Solutions Ltd|Method for improved geophysical investigation|
    RU2017145541A3|2015-06-04|2019-07-17|
    US10838093B2|2015-07-02|2020-11-17|Exxonmobil Upstream Research Company|Krylov-space-based quasi-newton preconditioner for full-wavefield inversion|
    EP3341850A1|2015-08-25|2018-07-04|Saudi Arabian Oil Company|Three-dimensional elastic frequency-domain iterative solver for full waveform inversion|
    RU2693495C1|2015-10-02|2019-07-03|Эксонмобил Апстрим Рисерч Компани|Complete wave field inversion with quality factor compensation|
    CN108139498B|2015-10-15|2019-12-03|埃克森美孚上游研究公司|FWI model domain angular stack with amplitude preservation|
    US10768324B2|2016-05-19|2020-09-08|Exxonmobil Upstream Research Company|Method to predict pore pressure and seal integrity using full wavefield inversion|
    US10908305B2|2017-06-08|2021-02-02|Total Sa|Method for evaluating a geophysical survey acquisition geometry over a region of interest, related process, system and computer program product|
    US11048001B2|2018-03-30|2021-06-29|Cgg Services Sas|Methods using travel-time full waveform inversion for imaging subsurface formations with salt bodies|
    CN110415721B|2018-04-28|2022-02-01|华为技术有限公司|Method and device for calculating cut-off frequency|
    WO2021040763A1|2019-08-26|2021-03-04|Landmark Graphics Corporation|Building scalable geological property models using machine learning algorithms|
    法律状态:
    2020-09-01| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|
    2020-09-08| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|
    2020-10-20| B08F| Application dismissed because of non-payment of annual fees [chapter 8.6 patent gazette]|Free format text: REFERENTE A 9A ANUIDADE. |
    2020-12-22| B11B| Dismissal acc. art. 36, par 1 of ipl - no reply within 90 days to fullfil the necessary requirements|
    2021-11-03| B350| Update of information on the portal [chapter 15.35 patent gazette]|
    优先权:
    申请号 | 申请日 | 专利标题
    US31856110P| true| 2010-03-29|2010-03-29|
    US61/318,561|2010-03-29|
    PCT/US2011/025616|WO2011123197A1|2010-03-29|2011-02-21|Full wavefield inversion using time varying filters|
    [返回顶部]