This paper describes the mechanical, electrical, computer, and software systems of a prototype computer controlled drilling rig. The rig has prototype computer controlled drilling rig. The rig has operated under computer control for over three years on a project investigating optimized turbine control. In addition to describing the advantages and the basic computer control features of the rig, this paper presents the practical problems for computer paper presents the practical problems for computer controlled drilling and suggests possible solutions to these difficulties.
In the past, several attempts at full computer control of the drilling process have been made, but all have ultimately proven to be short lived. Recent advances in the power and reliability of sensors and computer control hardware, as well as reductions in their cost, have brought computerized rig control closer to reality. However, some important hurdles remain.
It is the authors' experience that most of the hurdles lie on the electrical and mechanical side of the system. Computer hardware and software capabilities have improved to the point that almost any function can be economically programmed. However, problems with sensor reliability and calibration, problems with sensor reliability and calibration, and actuation still exist. Each of these problem areas is discussed in this paper.
In summary, the purpose of this paper is:
to describe the use of a computer controlled rig,
to describe the problems the authors have experienced during three years of computer controlled drilling,
to present possible solutions to these problems, and
to discuss the future for field problems, and
to discuss the future for field use of computer controlled drilling.
The computer controlled rig discussed here was one element of a project aimed at reducing costs in drilling slower formations (those that drill at less than about 25ft/hr). The rig has been used in dozens of experimental wells to drill more than 40,000 feet under complete computer control. This effort should be distinguished from recent efforts to automate/computer control the tripping process. The computer controlled rig discussed here does have a mechanized tripping system, but no attempts were made to automate the process. Rather, the effort focused on computer control of the drilling process itself (i.e., when the bit was actually drilling ahead). Consequently, the points made here will specifically apply to computerization of the drilling process. To the extent that the environment and requirements are similar, however, many of the points made here may also apply to computer points made here may also apply to computer automated tripping systems.
There have been many notable attempts at computerization of the drilling process. Generally, these past attempts combined fine control of the weight and rotary with drilling models to prevent hydraulic flounder and to minimize cost per foot. It was intended that by taking the human out of the control loop, there would also be increased reliability, improved vigilance, safer operation, and faster response to anomalous situations. All of these projects have been short lived, and the closest that most drilling operations come to automation is the use of a pneumatic automatic driller. Current automation project are mostly directed at automating the tripping process.
The short lives of these past drilling control projects is curious, especially in light of the fact projects is curious, especially in light of the fact that most of the literature seems to indicate that these projects held great promise. Here we sit ten or twenty years later wondering what went wrong.