Method for combining the results of ultrasound and x-ray and neutron cement evaluation logs through modality merging
A combining mechanism for borehole logging tool data that employs modality merging to combine the output data of various borehole logging tools to provide a combined result and automated interpretation is provided, said mechanism comprising: at least one mechanism for assigning interpretive values to individual processed data types; at least one mechanism for combining the interpretive value data sets; and, at least one mechanism for providing an interpretation. A method of combining borehole logging tool data that employs modality merging to combine the output data of various borehole logging tools to provide a combined result and automated interpretation is also provided, said method comprising: assigning interpretive values to individual processed data types; combining the interpretive value data sets; and, providing an interpretation.
1 . A method comprising:
causing first and second borehole logging tools to emit ultrasound energy and non-ultrasonic radiant energy, respectively, while the first and second borehole logging tools are positioned within a borehole defined by a casing;
discretizing ultrasonic log data as a function of depth to generate a first dataset that characterizes a first property of cement surrounding the casing with respect to depth, the ultrasonic log data formed by processing ultrasonic measurement data output by the first borehole logging tool using the ultrasound energy;
discretizing non-ultrasonic log data as a function of depth to generate a second dataset that characterizes a second property of the cement with respect to depth, wherein the first and second properties are different, and the non-ultrasonic log data formed by processing non-ultrasonic measurement data output by the second borehole logging tool using the non-ultrasonic radiant energy;
informing an inversion of the ultrasonic measurement data using the non-ultrasonic measurement data as a function of depth; and
determining integrity of the cement with respect to depth using a third dataset that comprises a weighted combination of the first and second datasets.
2 . The method of claim 1 , further comprising:
processing the ultrasonic measurement data from the first borehole logging tool to generate the ultrasonic log data.
3 . The method of claim 1 , wherein the non-ultrasonic measurement data is x-ray density data, and the method further comprises:
processing the x-ray density data from the second borehole logging tool to form the non-ultrasonic log data.
4 . The method of claim 1 , wherein the non-ultrasonic measurement data is neutron porosity data, and the method further comprises:
processing the neutron porosity data from the second borehole logging tool to form the non-ultrasonic log data.
5 . The method of claim 1 , wherein the non-ultrasonic measurement data is neutron activation data, and the method further comprises:
processing the neutron activation data from the second borehole logging tool to form the non-ultrasonic log data.
6 . The method of claim 1 , wherein the first or second borehole logging tools comprise a wireline-based tool.
7 . The method of claim 1 , wherein the first or second borehole logging tools comprise a logging-while-drilling-based tool.
8 . The method of claim 1 , further comprising:
creating a three-dimensional representation of the second property with respect to depth using the non-ultrasonic measurement data.
9 . A method comprising:
causing a first logging tool to generate ultrasonic measurement data using ultrasound energy, the first measurement data characterizing a first property of a cement volume surrounding a borehole with respect to depth;
causing a second logging tool to generate non-ultrasonic measurement data using non-ultrasonic radiant energy, the non-ultrasonic measurement data characterizing a second property of the cement volume with respect to depth, wherein the first and second properties are different;
quantizing ultrasonic log data comprising the ultrasonic measurement data by automatically assigning discrete values from a first set of discrete values for the first property to the ultrasonic log data as a function of depth to generate a first dataset that characterizes the first property of the cement volume with respect to depth;
quantizing non-ultrasonic log data comprising the non-ultrasonic measurement data by automatically assigning discrete values from a second set of discrete values for the second property to the non-ultrasonic log data as a function of depth to generate a second dataset that characterizes the second property of the cement volume with respect to depth; and
processing the first dataset and the second dataset using a modality merging technique to generate a third dataset that characterizes integrity of the cement volume with respect to depth, wherein processing the first and second datasets using the modality merging technique comprises informing an inversion of the ultrasonic measurement data using the non-ultrasonic measurement data as a function of depth.
10 . The method of claim 9 , wherein the non-ultrasonic radiant energy includes x-ray radiation or neutron radiation.
11 . The method of claim 9 , wherein processing the first dataset and the second dataset using the modality merging technique comprises correlating and matching respective values of the first and second datasets with respect to depth.
12 . The method of claim 9 , wherein processing the first dataset and the second dataset using the modality merging technique comprises assigning different weights to respective values of the first and second datasets.
13 . The method of claim 9 , further comprising:
filtering the third dataset to generate an informative log that includes continuous values for the integrity of the cement volume with respect to depth.
14 . The method of claim 9 , further comprising:
adjusting one or more weights assigned to respective values of the first and second datasets using a machine learning technique.