Non-linear statistically constrained AVO inversion

Details Non-linear statistically constrained AVO inversion Non-linear statistically constrained AVO inversion

2002-09-13

2004-11-16

Hoff, Marc S. (Department: 2857)

Metal deforming

With use of control means energized in response to activator...

Metal deforming by use of roller or roller-like tool element

C702S018000

Reexamination Certificate

active

06817218

ABSTRACT:

TECHNICAL FIELD OF THE INVENTION
The present invention relates to methods of AVO Inversion of seismic data
BACKGROUND OF THE INVENTION
Seismic data is useful in addressing certain petroleum exploration and exploitation problems associated with determining quality of hydrocarbon reservoirs. Amplitude Variation with Offset (AVO) techniques have been applied extensively to qualitative interpretations of seismic data. Efforts have increasingly been made to go beyond the qualitative analysis and to try to estimate parameters of the subsurface target layers using AVO inversion. Use of linear equations for AVO inversion results in extensive complex calculations that are extremely time consuming. The use of true non-linear equations becomes unmanageable due to computational limitations. The use of linear equations reduces the complexity of the calculations to computer manageable levels. However, noise-induced errors that occur when linear equations are used for the AVO inversion have limited the accuracy of reservoir quality determinations.
Inversion of pre-stack seismic P-P amplitude data to yield rock property contrast values &Dgr;V
p
/V
p
, &Dgr;V
s
/V
s
, and &Dgr;&rgr;/&rgr; and the absolute rock property values V
p
, V
s
and &rgr; (density) has been a hotly pursued goal for many years. An object has been to use rock property contrasts and absolute rock property values to predict, hydrocarbon presence and location in underground and undersea formations. Accurate determination of the degree of hydrocarbon saturation, reservoir quality, and the location of bypass pay among other useful information determinations have been elusive due to the inaccuracy and insensitivity of current industry standard techniques and methods including current industry standard AVO inversion or amplitude attribute determination methods. Traditional AVO inversion methods are most commonly based upon the use of unconstrained linear AVO equations, such as Equation 1, as given by Shuey, or other equivalent equations such as those given by Bortfield, Aki or Richards, solved on an event-by-event basis.
Amp(&thgr;)=
A+B
*Sin(&thgr;)
2
+C
*Tan(&thgr;)
2
*Sin(&thgr;)
2
  Equation 1
Where
A=(1/2)*&Dgr;Vp/Vp+(1/2)*&Dgr;&rgr;/&rgr;
B=(1/2)*&Dgr;Vp/Vp−(2/g)*(2*&Dgr;Vs/Vs+&Dgr;&rgr;/&rgr;)
C=(1/2)*&Dgr;Vp/Vp
The industry standard approach is limited by instability and inaccuracy in the inverted rock property contrasts in the presence of any amount of noise in the seismic data. The inversions also have limitations in the determination of small rock property contrasts. Further, the current inversion techniques are limited to moderate maximum angles of incidence, restricting the amplitudes associated with larger angles of incidence that are commonly present in modern acquisition long cable data sets.
The above-listed limitations hamper the petroleum exploration and exploitation field in many economical and business aspects. Specifically, the limitations make it difficult, inaccurate and expensive to determine hydrocarbon saturation, to detect residual hydrocarbons, to detect bypassed pay, to determine reservoir quality and to detect fracture presence.
SUMMARY OF THE INVENTION
The invention disclosed herein reduces the above-described limitations of prior art AVO inversion methods. The present invention discloses the use of both a non-linear inversion equation in combination with a statistically constrained inversion technique. Both aspects of the present invention add stability and accuracy to the inversion process. The invention produces results with noticeably more accuracy and stability in the face of noisy data than prior art methods using unconstrained linear equations or other traditional approaches.
The present method comprises providing a set of geologically reasonable statistical constraints, such as, but not limited to, Equations 2 and 3. Equation 2 relates the average values of any two of the rock properties to each other, so that the average rock property contrasts over a large window follow the well known rock property trends. The novel approach taken by the present invention involves simultaneously inverting the rock property contrasts for multiple interfaces and constraining the average of the rock property contrasts so that they follow relationships such as Equation 3.
The present method further comprises the use of a more stable and accurate non-linear AVO equations. The use of non-linear Equation 10 in the inversion in combination with the above-mentioned statistical constraints provides a more stable and accurate result than prior art methods.


REFERENCES:
patent: 5838634 (1998-11-01), Jones et al.
patent: 5999486 (1999-12-01), DeVault
patent: 6058074 (2000-05-01), Swan et al.
patent: 6091669 (2000-07-01), Chen
patent: 6263284 (2001-07-01), Crider et al.
patent: 6421611 (2002-07-01), Kelly et al.
patent: 6618678 (2003-09-01), Van Riel
patent: 2002/0013661 (2002-01-01), Van Riel et al.
patent: 2003/0018435 (2003-01-01), Jenner et al.
Xiang-Yang Li, Hengchang Dai, Michael C. Mueller and Olav I. Barkved in “Compensating for the effects of gas clouds on C-wave imaging: A case study from Valhall” published inthe Leading Edge, Sep. 2001, pp. 1022-1028.
Folke Engelmark, in “Using converted shear waves to image reservoirs with low-impedance contrast” publishedThe Leading Edge, Jun. 2000, pp. 600-603.

Affiliated with

Also associated with

Rating

Non-linear statistically constrained AVO inversion does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Non-linear statistically constrained AVO inversion, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Non-linear statistically constrained AVO inversion will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3355687