Mineral Inversion Module

Objective

The Inversion module provides a means to calculate mineral volumes based on a user specified mineral model, the mineral properties, and the input data. The primary results are a total porosity (from a sum of the fluids) and a grain density (based on the mineral volumes). The mineral volumes can also be used or exported.

Key Parameters

Displaying a matrix layout for each mineral model necessitates a different layout for this module than other modules. As such, when users are setting up mineral models they have the option of setting up a “Default” model and to then make customizations to that model for each zone.

Mineral matrix:

• Zone: Select the zone (or default) for modification.
• Mineral: The name of the mineral in the model.
• Enabled: Checkbox to specify if the mineral will be used in the inversion for the active zone.
• Grain: Specify if the mineral is part of the matrix (checked) or a fluid (unchecked).

Note that the mineral properties given on each row correspond to the curve listed in the Curves matrix.

Curves matrix:

• Curve: Dropdown menu with available curves. User can also type in a curve name.
• Mode: Specifies how the curve will be used in the inversion.
• Conf. Int: Places a weight on the curve in the inversion. Smaller weights increase importance. Should be set relative to curve value range.

Output Curves

Most important curves shown in bold.

• MINV_GRAIN_DENSITY: Grain density from the mineral inversion.
• MINV_POROSITY_STRICT: Sum of the fluids from the mineral inversion.
• RHOB_PRED: Predicted density from the mineral inversion for model QC.
• NPHI_PRED: Predicted neutron from the mineral inversion for model QC.
• PE_PRED: Predicted photoelectric factor from the mineral inversion for model QC.
• SONIC_PRED: Predicted sonic from the mineral inversion for model QC.
• COND_PRED: Predicted conductivity from the mineral inversion for model QC.
• VWCL_PRED: Predicted clay volume from the mineral inversion for model QC.
• V_KER_PRED: Predicted kerogen volume from the mineral inversion for model QC.
• MINV_MODEL_TOTAL_ERROR: Sum of errors between predicted curves from inversion and actual curves.
• MINV_MODEL_AVG_ERROR: Average error between predicted curves from inversion and actual curves.

CPI Config Reference

The relevant configuration file is 07_inversion.yaml.

Display

1. Mineralogy track.
2. Grain density track.
3. Total porosity track.
4. Bulk density inversion QC track.
5. Neutron inversion QC track.
6. Photoelectric inversion QC track.
7. Compressional sonic inversion QC track.
8. Conductivity inversion QC track.
9. Clay volume inversion QC track.
10. Kerogen volume inversion QC track.
11. Inversion error track.

Associated cross-plots

• Rt-Clay cross-plot.
• DT-Clay cross-plot.

Associated maps

None.

Recommendations

• When setting the parameters expand the window so that the mineral and curve matrices are aligned.
• Set the default mineral model first and then make zone-by-zone modifications as necessary.
• Use the minimum number of minerals possible to prevent / minimize non-unique solutions due to being underdetermined.
• Remember that most mineral properties are well established. The clay and fluid properties have the most potential variation from the defaults.