Real-Time Optimization of Drilling Parameters by Autonomous Empirical Methods

Abstract

Abstract In the past half century, many techniques have evolved for the purposes of optimizing key parameters while drilling to achieve performance gains and economic improvements (Lubinski 1958; Speer 1958). More recently, with the advent of computer systems and controls, different approaches to real-time optimization of drilling performance have been developed in attempts to improve overall profitability, with and without success (Brett et al. 1990; Dupriest 2006; Dupriest and Koederitz 2005; Young 1969). The reason being, the successful application of a vast majority of these drilling optimization methods requires substantial company resources in terms of technical expertise and rig time, which could impact well economics. Moreover, their results can be specific to the particular drilling condition currently being optimized. It was recognized that drilling optimization research was needed to address these limitations, which had a two-fold goal: 1) to investigate the potential for autonomous drilling control with no drilling-specific knowledge, and 2) to develop a prototype platform for future drilling automation efforts. The research effort led to the development of a drilling optimization system, called the optimizer in this paper, and three elements can be used to describe it: 1) system architecture, 2) an optimization concept, and 3) software components comprised of a test process, a search method, and user interfaces. The system architecture supports three different modes of operation, namely advisory, semi-autonomous, and autonomous. The concept for the optimization logic is essentially to mimic human search techniques that are generally observed in the field. The software system uses a test process to evaluate and quantify the drilling performance for a given set of target setpoints. A search method is used to identify these setpoints to be tested; its development was based on early work in the application of real-time mechanical specific energy (MSE) displays (Dupriest and Koederitz 2005; Koederitz and Weis 2005). Key to the development of both the optimizer's test process and search method is their ability to automatically adapt to the behavior of the complete drilling system, including formation changes, current bit and bottomhole assembly (BHA) characteristics, and the existing preciseness (or lack thereof) of drilling parameter control being achieved. The system's user interfaces consist of screen sets for the optimizer application that include pop-up alerts to advise the driller of the need for a setpoint change. This paper describes the development and field testing of this autonomous drilling system, the optimizer, and provides insight into its future potential. Overall, the field testing results were favorable, displaying that the potential for autonomous drilling optimization without drilling knowledge is practical, flexible, and economical, exhibiting promise in a range of cost-effective applications. More effective approaches will undoubtedly develop, and using drilling-specific know-how to enhance optimization logic will most likely increase its effectiveness. While the greatest potential for success operationally are in the areas of fully autonomous automated control and fully electronic control systems, the optimizer also is believed to offer cost-effective drilling optimization in retrofitted control systems.