AI-Enabled, Automated Digital Dull Bit Analysis - Field Experience

Abstract

Background Digitization and automation have been areas of increasing focus in the drilling industry in recent years. One critical area of drilling operations that has been largely overlooked in the drive to digitization is the assessment of drill bit wear and damage. The International Association of Drilling Contractors (IADC) drill bit dull grading standard is the current industry reference to document the condition of a dull drill bit. Since these protocols rely on human interaction and judgement, the resulting data is limited in terms of its accuracy, its consistency, and its comparability. As a result, the usefulness of this data in improving how bits are designed and operated is also limited. This paper describes an experience of a new system, which involves scanning a drill bit, and digitally analysing the results, thereby overcoming the limitations of the current protocols. The implementation of the drill bit scanner and dull bit grading software services will contribute greatly to improve the inspection, and classification of drill bits. Furthermore, it will enable to monitor drill bit performance, and optimize drilling processes by utilizing the data provided by the system. The system described incorporates the automated generation of a digital three-dimensional visualization of a dull bit, which is then analysed digitally to assess wear and damage, on an individual cutter basis, as well as on an overall bit basis. Since the process is automated and digital in nature, the uncertainties related to human interaction and judgement in the process typically used today are eliminated. This data can then be used to identify drilling dysfunctions, and modify drilling procedures accordingly to optimize performance, as well as to identify potential improvements in drill bit design. Examples of digital dull bit analyses demonstrate that the bit wear data obtained from the system is much more detailed, more accurate, more consistent, and more comparable than the methods employed today. The resulting data is also much more suited to analytics, as well as other types of analyses, with a view to improving bit designs, identifying drilling dysfunctions causing bit damage, and optimizing drilling operating parameters to improve performance.