Back to results
US 11,378,707 B2Granted
Third interface echo (TIE) detection from flexural data for gas/liquid annulus discrimination
AssigneeSchlumberger Technology Corporation
InventorsMikhail Lemarenko, Elizaveta Rybka, Ram Sunder Kalyanraman, Josselin Kherroubi
Life of this patent
Estimating the term…
Abstract
Systems and methods for identifying a potential third interface echo (TIE) using objective criteria are provided. A system includes an acoustic logging tool that obtains measurements in a wellbore and a data processing system that has a processor that receives the measurements from the acoustic logging tool. The data processing system may identify a third interface echo (TIE) using a neural network and/or by a signal analysis method based on the behavioral characteristics of the TIE signal.
Claims23
A system comprising:
- an acoustic logging tool configured to obtain flexural measurements in a wellbore crossing a geological formation and having a casing installed inside the wellbore;
- and
- a data processing system comprising a processor configured to:
- receive the flexural measurements;
- produce a well log view image using the flexural measurements that includes flexural waveforms in a plurality of positions versus time;
- process the well log view image to identify a presence or absence of a third interface echo by sending the well log view image to a neural network, wherein the neural network is trained to identify a third interface echo, and by receiving a modified well log view image from the neural network, wherein, for a presence of a third interface echo, the modified well log view image includes an indication of the presence of the third interface echo;
- in response to identifying the presence of the third interface echo, identify an annulus between the casing and the geological formation as having a first characteristic;
- and
- in response to identifying the absence of the third interface echo, identify the annulus between the casing and the geological formation as having a second characteristic.
6 more elements
A method comprising:
- receiving, into processing circuitry, flexural measurements obtained from an acoustic logging tool in a wellbore that crosses a geological formation and has a casing installed inside the wellbore;
- producing, using the processing circuitry, a well log view image using the flexural measurements that includes flexural waveforms in a plurality of positions versus time;
- processing, using the processing circuitry, the well log view image to identify a presence or absence of a third interface echo;
- in response to identifying the presence of the third interface echo, using the processing circuitry to identify an annulus between the casing and the geological formation as having a first characteristic;
- in response to identifying the absence of the third interface echo, using the processing circuitry to identify the annulus between the casing and the geological formation as having a second characteristic;
- receiving, into the processing circuitry, pulse echo measurements obtained from the acoustic logging tool;
- processing, using the processing circuitry, the pulse echo measurements to identify that the annulus between the casing and the geological formation has the first characteristic or the second characteristic;
- receiving, into the processing circuitry, pulse-echo calibration measurements obtained by the acoustic logging tool at a calibration location corresponding to a free pipe;
- receiving, into the processing circuitry, flexural measurements obtained by the acoustic logging tool at the calibration location;
- identifying, using the processing circuitry, the annulus between the casing and the geological formation in the calibration location as having the first or the second characteristic;
- and
- calibrating, using the processing circuitry, the pulse-echo measurements based on the first or second characteristic identification and the pulse-echo calibration measurements.
9 more elements
An article of manufacture comprising one more tangible, non-transitory, machine-readable media comprising instructions that, when executed by one or more processors, cause the one or more processors to:
- receive flexural measurements obtained by an acoustic logging tool in a wellbore crossing a geological formation and having a casing installed inside the wellbore;
- produce a well log view image using the flexural measurements that includes flexural waveforms in a plurality of positions versus time;
- process the well log view image to identify a presence or absence of a third interface echo using a trained neural network to use image recognition to identify behavioral characteristics of the third interface echo among the flexural measurements;
- in response to identifying the presence of the third interface echo, identify an annulus between the casing and the geological formation as having a first characteristic;
- and
- in response to identifying the absence of the third interface echo, identify the annulus between the casing and the geological formation as having a second characteristic.
3 more elements
A method comprising:
- receiving, into processing circuitry, flexural measurements obtained from an acoustic logging tool in a wellbore that crosses a geological formation and has a casing installed inside the wellbore;
- producing, using the processing circuitry, a well log view image using the flexural measurements that includes flexural waveforms in a plurality of positions versus time;
- processing, using the processing circuitry, the well log view image using a neural network that is trained to identify a presence of a third interface echo;
- in response to identifying the presence of the third interface echo, using the processing circuitry to identify an annulus between the casing and the geological formation as having a first characteristic;
- and
- in response to not identifying the presence of the third interface echo, using the processing circuitry to identify the annulus between the casing and the geological formation as having a second characteristic.
3 more elements
Dependencies are inferred from the claim text.