Shear, Compressional and Density logs from horizontal wells are often collected in the field with inadequate quality control (QC). In some cases (most notably with Shear) the data is so unreliable that it may be ignored by E&P companies in their geologic and geomechanical evaluations. For Quantico, this situation creates a challenge to generate accurate models for its proprietary QLog product because corrupt input data limits the ability to create reliable output.
Quantico leveraged its experience with a wide variety of geological settings and its vast library of well data – with more horizontal logging and drilling data than possessed by virtually any single oil and gas company – to develop a highly effective QC workflow. This workflow is based on geologically specific QC criteria for input log data and allows for the swift identification and elimination of “bad” data from its machine learning process. Additionally, because the machine learning software is trained only with “good” data, the product supplied to the customer can be used not only to generate more accurate QLogs, but QLogs can also be used to replace logs with corrupt field data over entire well intervals or portions of wells.
Quantico customers can be assured that only truly valid data is used in the calibration and generation of its proprietary QLog product. Further, Quantico can provide customers an assessment and QC/highgrade of their inventory of horizontal log data. Using QLog for wells without log data and QC’s for actual logs, customers can now quickly build the most reliable and accurate Earth models possible for use in geologic and geomechanical evaluation. Such models can be reliably updated using QLog in new horizontal wells without the expense collecting actual log data at high expense and wellbore risk.