Abstract

The objective of this paper is to apply machine learning to predict a synthetic caliper log using available Logging While Drilling (LWD) well logs. LWD logs and wireline (WL) mechanical caliper logs for two well datasets, one in tight carbonate and the other in tight sandstone, were used to train a machine learning model for each dataset. LWD logs included gamma ray, sonic, density, azimuthal density, porosity, photoelectric log, and mineralogy. In both models, the LWD logs were used as input parameters, and the maximum borehole diameter from the mechanical caliper log was the output parameter. The predicted synthetic borehole diameter was compared with the measured caliper log. The models were trained with XGBoost, which is an ensemble machine learning algorithm. In the training process, Leave-One-Out method was implemented where all wells in the dataset, except one, were used for training to predict the maximum borehole diameter in the remaining blind well. The results of the carbonate and sandstone datasets showed good prediction accuracy with an average absolute percentage error of less than 6%. This proposed methodology for caliper prediction was applied in the field to make completion design in several wells with multi-stage frac completions, which were subsequently stimulated successfully without completion integrity issues.

Introduction

Stress-induced wellbore failure during oil and gas well drilling can lead to over-gauged wellbore. Borehole instability leading to excessively enlarged wellbore can increase the risk of drilling difficulties, as well as cause completions integrity issues (Ottesen and Kwakwa, 1991; Chen et al., 2002; Kiran et al., 2017). For instance, there is a tendency for poor cementing or compromised zonal isolation across over-gauged wellbore sections in cemented-liner plug-n-perf completions. Enlarged wellbore sections can also result in loss of packers’ sealing in openhole multi-stage frac (MSF) completions. Such inadequacies could jeopardize well frac operations. In addition, borehole diameter data is also used to calculate cement volume for well completion. Moreover, other measurements such as formation density that are affected by wellbore enlargements also need borehole diameter for applying data corrections. Hence, knowing accurate wellbore diameter is critical for several operational requirements in oil and gas well construction.

While WL multi-arm mechanical caliper tools can be run in the open hole after drilling to get direct measurements of hole size, such measurements are not always possible due to prevailing drilling difficulties. Under such circumstances, making completions decisions become even more difficult with higher risk of failure. Moreover, a solo wireline logging run to acquire caliper data requires additional cost including rig time.

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