Real-Time Borehole Correction of Multicomponent Induction Data Acquired in OBM Wells: Algorithm and Applications1
- Junsheng Hou (Halliburton) | Luis Sanmartin (Halliburton) | Dagang Wu (Halliburton) | F. Turker Celepcikay (Halliburton) | David Torres (Halliburton)
- Document ID
- Society of Petrophysicists and Well-Log Analysts
- Publication Date
- April 2013
- Document Type
- Journal Paper
- 128 - 141
- 2013. Society of Petrophysicists & Well Log Analysts
- 3 in the last 30 days
- 138 since 2007
- Show more detail
Multicomponent induction (MCI) logging has been successfully used for the evaluation of anisotropic formations in recent years. In MCI applications, it is commonly known that a borehole signifi cantly affects MCI tool measurements, even in oil-based mud (OBM) environments, and these effects get much stronger for shorter-spacing subarrays operated at higher frequencies. Consequently, the MCI measurements must be corrected for borehole effects before the measured data can be used to obtain desired reservoir properties by using other advanced data processing such as vertically 1-dimensional (V1D) inversion.
In this paper, the authors present a real-time borehole correction (BHC) algorithm and its processing system designed to correct for the borehole effects of the measurements acquired by a newly developed MCI tool when logging in OBM or air-drilled wells. The proposed BHC methodology uses a model-based multistep inversion method to determine all unknown formation, borehole, and tool-position parameters. Once these inverted parameters are obtained, they are used in a forward model to compute the borehole effects for each subarray and frequency, and to apply these corrections to the logging data after calibration and temperature correction. The BHC logs are fi nally used as inputs for the ensuing advanced data processing that provides the desired product answers such as formation dip and anisotropic resistivity.
The key in this real-time BHC algorithm/system is the accurate and fast inversion for the unknown parameters from MCI measurements. The multistep algorithm of the inversion based on a radially 1-dimensional (R1D) model is implemented by splitting the inversion problem of one high-dimension unknown vector into several lowerdimension ones, based on their sensitivity to different components of measured conductivity tensors for different subarrays operated at different frequencies. This reduction in dimensionality makes the BHC nonlinear inversion considerably faster, more reliable, and robust. Another critical issue for this accurate and fast BHC processing is to use the prebuilt lookup table/library as the forward engine to replace the time-consuming 3-dimensional (3D) simulation in the inversion. As its by-products, the BHC system also provides real-time inverted formation parameters, tool location in the borehole, and borehole size (if they are not available from other independent measurements). To test the new BHC method and system, MCI synthetic and fi eld log data have been processed for the borehole correction by using the new algorithm/system. The BHC results show that the BHC based on the multilevel processing and multistep inversion approach is fast, reliable, and robust, enabling it to be accurately used for real-time MCI data processing in OBM and air-drilled wells.
|File Size||17 MB||Number of Pages||14|