Casing string monitoring using electromagnetic (EM) corrosion detection tool and junction effects correction
A corrosion monitoring method includes obtaining electromagnetic (EM) log, data along a casing string. The method also includes processing the EM log data to estimate casing thickness of the casing string as a function of position, where the processing includes correcting, for junction effects in the casing string.
1. A corrosion monitoring method that comprises:
disposing an electromagnetic (EM) logging tool downhole;
generating a time-varying EM field along a casing string with a transmitter, wherein the transmitter is a coil or a solenoid;
recording a voltage induced by the EM field with a receiver, wherein the receiver is a coil or a solenoid;
constructing an EM data log with the recorded voltage; and
processing the EM data log to estimate casing thickness of the casing string as a function of position, wherein said processing comprises correcting for junction effects in the casing string, wherein the correcting for junction effects in the casing string comprises:
comparing results from a one-dimensional (1D) casing string model and a multi-dimensional casing string model to identify the junction effects; and
re-processing the EM log data using the 1D casing string model with the junction effects removed.
2. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises employing a multi-dimensional casing string model.
3. The method of claim 2 , wherein said correcting for junction effects further comprises applying a multi-stage inversion with multiple stages, where a different casing string attribute or set of attributes is inverted for each stage while at least one other casing string attribute is fixed.
4. The method of claim 3 , wherein the multi-stage inversion comprises:
a first stage that inverts a junction location while a casing thickness is fixed; and
a second stage that inverts a casing thickness while junction location is fixed using values determined in the first stage.
5. The method of claim 4 , further comprising:
selecting EM log data at or near a casing junction for the first stage; and
selecting EM log data at or near a middle region of a casing section for the second stage.
6. The method of claim 4 , wherein the multi-stage inversion further comprises a third stage that inverts a junction location and a casing thickness using values determined in the first and second stages as initial values.
7. The method of claim 4 , wherein the multi-stage inversion further comprises performing multiple iterations of the multi-stage inversion, where initial values for attributes to be determined for each stage are based on a previous iteration.
8. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises estimating junction positions directly from the EM log data, and inverting at least one other casing string attribute while a junction position attribute based on said estimating is fixed.
9. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises calculating a first forward response using the multi-dimensional model and a second forward response using the 1D model, and wherein comparing results comprises comparing the first and second forward responses.
10. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises using a one-dimensional (1D) model to identify one or more casing string attributes, and setting values for a multi-dimensional model based on the one or more casing string attributes determined using the 1D model.
11. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises applying a layer-sliding inversion.
12. The method of claim 1 , wherein said correcting for junction effects in the casing string comprises applying a constraint condition that limits an amount of variance between casing thickness results of a one-dimensional (1D) casing string model and casing thickness results of a multi-dimensional casing string model.
13. A corrosion monitoring system that comprises:
an electromagnetic (EM) logging tool to collect EM log data along a casing string, wherein the EM logging tool comprises:
a transmitter, wherein the transmitter generates a time-varying EM field, wherein the transmitter is a coil or a solenoid; and
a receiver, wherein the receiver records a voltage induced by the EM field, wherein the receiver is a coil or a solenoid; and
a processing unit in communication with the EM logging tool, wherein the processing unit processes the EM log data to estimate casing thickness of the casing string as a function of position, wherein the processing unit corrects for junction effects in the casing string by:
comparing results from two different models to identify the junction effects; and
re-processing the EM log data using one of the different models with the junction effects removed.
14. The system of claim 13 , wherein the processing unit corrects for junction effects in the casing string by performing a multi-stage inversion with multiple stages, where a different casing string attribute or set of attributes is inverted for each stage while at least one other casing string attribute is fixed.
15. The system of claim 14 , wherein the multi-stage inversion further comprises a stage that inverts a junction location and a casing thickness using values determined in previous stages as initial values.
16. The system of claim 14 , wherein the processing unit performs multiple iterations of the multi-stage inversion, where initial values for attributes to be determined for each stage are based on a previous iteration.
17. The system of claim 13 , wherein said processing unit corrects for junction effects in the casing string based on a multi-dimensional casing string model.
18. The system of claim 13 , wherein said processing unit corrects for junction effects in the casing string by applying at least one of a layer-sliding inversion, and a constraint condition that limits an amount of variance between casing thickness results of a one-dimensional (1D) model and casing thickness results of a multi-dimensional model.
19. A corrosion monitoring method that comprises:
disposing an electromagnetic (EM) logging tool downhole;
generating a time-varying EM field along a casing string with a transmitter, wherein the transmitter is a coil or a solenoid;
recording a voltage induced by the EM field with a receiver, wherein the receiver is a coil or a solenoid
constructing an EM data log with the recorded voltage; and
processing the EM data log to estimate casing thickness of the casing string as a function of position, wherein said processing comprises correcting for junction effects in the casing string, wherein said correcting for junction effects in the casing string comprises:
employing a multi-dimensional casing string model;
applying a multi-stage inversion with multiple stages, where a different casing string attribute or set of attributes is inverted for each stage while at least one other casing string attribute is fixed, wherein the multi-stage inversion comprises a first stage that inverts a junction location while a casing thickness is fixed and a second stage that inverts a casing thickness while the junction location is fixed using values determined in the first stage;
selecting EM log data at or near a casing junction for the first stage; and
selecting EM log data at or near a middle region of a casing section for the second stage.