Decoupling tensor components without matrix inversion

Patent metadata

Field	Value
Publication number	US-11035976-B2
Application number	US-201916633830-A
Country	US
Kind code	B2
Filing date	Mar 6, 2019
Priority date	Mar 6, 2019
Publication date	Jun 15, 2021
Grant date	Jun 15, 2021

Abstract

Numerical and/or semi-analytical methods are leveraged to decouple a complete set of nonzero electromagnetic field tensor components (118) from detected signal data (119). Nine nonzero components can serve as inputs for a three-dimensional inversion process to determine formation properties. A resistivity tool (100) containing at least one transmitter (111) and at least one receiver (108, 109) at tilted angles receives an electromagnetic signal throughout a rotation. A difference in the azimuthal positions of the transmitter(s) and receiver(s) during rotation of the resistivity tool can result in an azimuthal offset between resistivity tool subs. The components (118) are decoupled from the detected signal data (119) numerically or semi-analytically according to whether the azimuthal offset angle is known. If the azimuthal offset angle is known, the nine components are determined numerically through curve fitting. If the azimuthal offset angle is unknown, a semi-analytical process is used to solve for the nine components.

First claim

What is claimed is:

1. A method comprising:
detecting, at a first receiver of a set of one or more receivers a formation response to electromagnetic radiation emitted from a first transmitter of a set of one or more transmitters, wherein the first receiver is tilted at a first angle of tilt and the first transmitter is tilted at a second angle of tilt;
determining complex azimuthal electromagnetic field data corresponding to the formation response;
decoupling a first plurality of nonzero tensor components from the complex azimuthal electromagnetic field data without matrix inversion,
wherein the decoupling comprises numerically decoupling if an azimuthal offset between the first transmitter and the first receiver is known and semi-analytically decoupling if the azimuthal offset between the first transmitter and the first receiver is unknown,
wherein the decoupling is performed downhole; and

communicating the first plurality of nonzero tensor components uphole.

2. The method of claim 1 , wherein semi-analytically decoupling comprises:
calculating a minimum for the azimuthal offset if the azimuthal offset is unknown;
determining a parallel signal value and a perpendicular signal value corresponding to pairing of the set of one or more transmitters and set of one or more receivers;
calculating a sum of the parallel signal value and the perpendicular signal value and a difference of the parallel signal value and the perpendicular signal value; and
determining a first subset of the first plurality of nonzero tensor components based, at least in part, on the calculated azimuthal offset minimum and at least one of the sum and the difference.

3. The method of claim 2 , further comprising performing a tensor rotation on the first subset of nonzero tensor components to determine a second subset of the first plurality of nonzero tensor components, wherein the first and the second subset of nonzero tensor components form the first plurality of nonzero tensor components.

4. The method of claim 1 , wherein numerically decoupling the first plurality of nonzero tensor components comprises:
fitting a plurality of curve representations corresponding to azimuthal angles at which measurements have been taken corresponding to the formation response,
wherein fitting the plurality of curve representations comprises fitting each curve representation based on the known azimuthal offset, fitting coefficients, and formation response measurements at the corresponding one of the azimuthal angles;

determining the fitting coefficients across the plurality of curve representations for each pairing between the set of one or more transmitters and the set of one or more receivers; and
based, at least in part, on the fitting coefficients, calculating the first plurality of nonzero tensor components.

5. The method of claim 4 , wherein calculating the first plurality of nonzero tensor components based on the fitting coefficients comprises solving for each receiver of the set of one or more receivers,
H r (β i )= A cos(2β i +β ref )+ B sin(2β i +β ref )+ C cos β i +D sin β i +E,

wherein β i represents the formation response measurement at an azimuthal angle bin i for receiver r, β ref indicates the known azimuthal offset corresponding to the receiver, and A, B, C, D, and E are the fitting coefficients.

6. The method of claim 5 , wherein solving comprises regression analysis.

7. The method of claim 6 , wherein the regression analysis comprises a least squares method.

8. The method of claim 1 further comprising:
determining whether a dogleg angle exists between a transmitter sub and a receiver sub;
based on a determination that the dogleg angle exists, determining angles of tilt of those of the set of one or more receivers and set of one or more transmitters corresponding to the transmitter sub and the receiver sub; and
correcting the angles of tilt based, at least in part, on a measurement of the dogleg angle, wherein the decoupling uses the corrected angles of tilt.

9. One or more non-transitory machine-readable media comprising program code executable by a processor to perform operations comprising:
determining complex azimuthal electromagnetic field data corresponding to a formation response to electromagnetic radiation emitted from at least a first transmitter of a set of one or more transmitters and detected by a set of one or more tilted receivers, wherein the first transmitter is tilted at a first angle of tilt;
decoupling a first plurality of nonzero tensor components from the complex azimuthal electromagnetic field data without matrix inversion,
wherein the decoupling comprises numerically decoupling if an azimuthal offset between the first transmitter and the set of one or more titled receivers is known and semi-analytically decoupling if the azimuthal offset is unknown,
wherein the decoupling is performed downhole; and

communicating the first plurality of nonzero tensor components uphole.

10. The non-transitory machine-readable media of claim 9 , wherein semi-analytically decoupling comprises:
calculating a minimum for the azimuthal offset if the azimuthal offset is unknown;
determining a parallel signal value and a perpendicular signal value corresponding pairing of the set of one or more transmitters and set of one or more receivers;
calculating a sum of the parallel signal value and the perpendicular signal value and a difference of the parallel signal value and the perpendicular signal value; and
determining a first subset of the first plurality of nonzero tensor components based, at least in part, on the calculated azimuthal offset minimum and at least one of the sum and the difference.

11. The non-transitory machine-readable media of claim 10 , wherein the operations further comprise performing a tensor rotation on the first subset of nonzero tensor components to determine a second subset of the first plurality of nonzero tensor components, wherein the first and the second subset of nonzero tensor components form the first plurality of nonzero tensor components.

12. The non-transitory machine-readable media of claim 9 , wherein numerically decoupling the first plurality of nonzero tensor components comprises:
fitting a plurality of curve representations corresponding to azimuthal angles at which measurements have been taken corresponding to the formation response,
wherein fitting the plurality of curve representations comprises fitting each curve representation based on the known azimuthal offset, fitting coefficients, and formation response measurements at the corresponding one of the azimuthal angles;

determining the fitting coefficients across the plurality of curve representations for each pairing between the set of one or more transmitters and the set of one or more tilted receivers; and
based, at least in part, on the fitting coefficients, calculating the first plurality of nonzero tensor components.

13. The non-transitory machine-readable media of claim 12 , wherein calculating the first plurality of nonzero tensor components based on the fitting coefficients comprises solving for each receiver of the set of one or more tilted receivers
H r (β i )= A cos(2β i +β ref )+ B sin(2β i +β ref )+ C cos β i +D sin β i +E,

wherein β i represents the formation response measurement at an azimuthal angle bin i for receiver r, β ref indicates the known azimuthal offset corresponding to the receiver, and A, B, C, D, and E are the fitting coefficients.

14. The non-transitory machine-readable media of claim 9 , wherein the operations further comprise:
determining whether a dogleg

Assignees

• Halliburton Energy Services Inc

Inventors

• Hou Junsheng
• Bittar Michael S
• Fan Yijing
• Pan Li
• Ma Jin

Classifications

• G01V3/38

  Processing data, e.g. for analysis, for interpretation, for correction · CPC title

• G01V3/26 Primary

  operating with magnetic or electric fields produced or modified either by the surrounding earth formation or by the detecting device (with electromagnetic waves G01V3/30) · CPC title

• E21B7/04

  Directional drilling · CPC title

• E21B7/022

  Control of the drilling operation; Hydraulic or pneumatic means for activation or operation (control circuits for drilling masts E21B15/045) · CPC title

• E21B2200/20

  Computer models or simulations, e.g. for reservoirs under production, drill bits · CPC title