ABSTRACT

Over the past few years, a large number of wells were logged with casedhole resistivity service. The wealth of information in the logs has helped us with improved hydrocarbon production. On the other hand, there are still several issues that make proper interpretation in some wells difficult. This paper analyzes the sources of the problems and explores the proper ways of identifying and minimizing the anomalies which are unrelated to the formation. We encountered three main classes of anomalies in some of the logs. The first kind is from localized casing inhomogeneities. These are mostly from the casing perforations, casing collars, and casing centralizers. The second type is from the k-factor or the normalization factor. The casedhole resistivity so far is still a relative measurement because the voltage with respect to a reference electrode at "infinite" cannot be made easily. The log is thus adjusted to overlay with openhole resistivity in an impermeable zone or in a zone where no change has happened over the time (e.g. in a shale). Therefore, this step also requires accurate openhole resistivity in shale, which may not always readily available. The third one comes from the cement effect. Since casedhole resistivity still cannot achieve multiple depths of investigation, if this relatively rare cement effect becomes non-negligible, we cannot separate the contribution of cement from that of formation without constraints from other sources. Through extensive numerical modeling, a better understanding of the tool-response characteristics is achieved. It is found that an individual casing collar, a casing centralizer, or a casing perforation creates a localized anomaly within an interval of about six feet or less. At the sampling rate of 2ft, the interval is equivalent to about 3 measurement points or less. The anomalous points have both higher and lower values than the actual formation resistivity. Based on these findings, a numerical processing algorithm is developed to filter out the localized casing anomalies. In the case of log normalization, one needs to make sure that the openhole log in shale has an accurate value by applying all necessary corrections properly. The processing methods and interpretation procedures have been applied to a series of wells and significant improvement in hydrocarbon saturation evaluation is achieved. Over the past few years, a large number of wells were logged with casedhole resistivity service. The wealth of information in the logs has helped us with improved hydrocarbon production. On the other hand, there are still several issues that make proper interpretation in some wells difficult. This paper analyzes the sources of the problems and explores the proper ways of identifying and minimizing the anomalies which are unrelated to the formation. We encountered three main classes of anomalies in some of the logs. The first kind is from localized casing inhomogeneities. These are mostly from the casing perforations, casing collars, and casing centralizers. The second type is from the k-factor or the normalization factor. The casedhole resistivity so far is still a relative measurement because the voltage with respect to a reference electrode at "infinite" cannot be made easily. The log is thus adjusted to overlay with openhole resistivity in an impermeable zone or in a zone where no change has happened over the time (e.g. in a shale). Therefore, this step also requires accurate openhole resistivity in shale, which may not always readily available. The third one comes from the cement effect. Since casedhole resistivity still cannot achieve multiple depths of investigation, if this relatively rare cement effect becomes non-negligible, we cannot separate the contribution of cement from that of formation without constraints from other sources. Through extensive numerical modeling, a better understanding of the tool-response charact

This content is only available via PDF.
You can access this article if you purchase or spend a download.