An Algorithm to Automatically Zero Weight on Bit and Differential Pressure and Resulting Improvements in Data Quality

Abstract

Summary Weight on bit (WOB) and differential pressure (DIFP) are two essential parameters derived from surface sensors during the drilling process. However, there can be significant errors in these measurements due to improper zeroing of these traces. Regular zeroing of WOB is important to ensure that it is calibrated for additional stands that have been added to the drillstring. Similarly, regular zeroing of DIFP is important to ensure that added hydrostatic pressure, which increases with depth, is taken into account, and to ensure that DIFP has been calibrated to the correct pump rate. In this paper, we quantify the errors due to forgotten and incorrect zeroing, and discuss an algorithm that was developed to automatically zero WOB and DIFP to eliminate these errors. This study has two main sections. In the first section, the current practice of zeroing WOB (prior to rotary drilling) is analyzed in 40 onshore wells. It is found for 86% of all stands that WOB is either zeroed incorrectly or not at all. An algorithm is developed to determine the appropriate time to perform the zero WOB operation. Using this algorithm, it is found that the average WOB error per stand due to improper zeroing is 16.8 and 17.6% in the vertical and lateral wellbore sections, respectively. Further, it was found that repeated forgotten zeroes could result in large errors, particularly in the vertical where 8% of stands have WOB inaccuracies of more than 30%. In the second section, the analysis is repeated for DIFP and it is found that zeroing DIFP is forgotten in 51% of all stands, resulting in errors of 17.5 and 8.9% in the vertical and lateral, respectively. In the vertical, 9.9% of stands have DIFP inaccuracies of more than 30% due to forgotten zeroing. Applying these algorithms to historic data can eliminate these errors, and improve the effectiveness of data-based drilling optimization and analytics. Further, these algorithms could be implemented in an electronic drilling recorder (EDR) to improve the quality of real-time data at the rig.