- Dennis Denney (JPT Technology Editor)
- Document ID
- Society of Petroleum Engineers
- Journal of Petroleum Technology
- Publication Date
- May 2006
- Document Type
- Journal Paper
- 49 - 50
- 2006. Society of Petroleum Engineers
- 0 in the last 30 days
- 149 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||Free|
|SPE Non-Member Price:||USD 17.00|
This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 95611, "Quantification of Depth Accuracy," by A. Brooks, SPE, and H. Wilson, SPE, Inteq; A. Jamieson, SPE, Tech 21; D. McRobbie, SPE, Sperry Drilling Services; and S.G. Holehouse, SPE, Total S.A., prepared for the 2005 SPE Annual Technical Conference and Exhibition, Dallas, 9-12 October.
Depth is a critical measurement in the economic development of a hydrocarbon asset. Almost all downhole activities, from making petrophysical measurements to setting packers, are performed remotely from surface. The common reference for all such activities is depth. A vertical-depth error of less than 1 m can have a financial effect in the millions of dollars. Although the physical measurement made at the rigsite normally is along-hole depth, vertical depth defines the relationship between subsurface features. The quantification of measured along-hole-depth uncertainty is, therefore, only a partial solution; it also is necessary to estimate vertical uncertainty.
There are frequent calls from the end users of formation-evaluation (FE) logs for improved depth accuracy. Zones of interest identified from FE logs (e.g., zones targeted for production or injection) are subsequently exploited with tools and procedures that also are applied at specified depths. Therefore, improvements made to the measurement and management of FE depths should be applied to all other depth measurements.
It has been proposed that rational improvement in depth-measurement accuracy is not possible until current performance is better understood and properly quantified and that the directional-survey-tool error models commonly used in the industry to predict wellbore-position uncertainty offer a starting point for modeling the performance of depth-measurement systems. Survey-tool error models quantify accuracy largely in terms of uncertainty or probability. The output consists of position bias and position uncertainty, but these values are derived from estimates of the biases and uncertainties associated with the measured values of along-hole depth, inclination, and azimuth. Along-hole depth is more commonly referred to as measured depth (MD).
Several directional-survey-tool error models are described in the literature. These models include MD terms that can be extracted, revised, and added to producing a dedicated MD error model. The most recent papers on the subject were written under the auspices of the Industry Steering Committee for Wellbore Survey Accuracy (ISCWSA). The models described in these papers are widely adopted within the industry and are likely to become de facto standards. In 2004, the ISCWSA was assimilated into the SPE Wellbore Positioning Technical Section.
This new technical section understood the development of a comprehensive depth-error model as a natural extension of the earlier error-modeling work of the ISCWSA and as something that might benefit the wider wellbore-construction community. This paper is a first step in meeting the objective of providing a standard depth-error model. It should be noted that an error model is a necessary component of, not a substitute for, good depth-management practices.
Basic Survey-Tool Error Model
A survey-tool error model outputs position uncertainty in three dimensions. An MD model needs to deal in only one dimension, resulting in a scalar uncertainty acting along the wellbore’s trajectory at the measurement point.
It is possible to construct a general error model that includes sufficient terms to model all common depth-measurement systems. By definition, the values of the error terms in such a general model are unspecified. System-specific error models are defined by inserting appropriate values for selected terms.
|File Size||69 KB||Number of Pages||2|