The important task of managing the drilling fluid during drilling operations is traditionally performed by highly qualified mud engineers at the rig site, following the constraints of the drilling fluid program, designed prior to commencing drilling operations. Due to uncertainty in planning data, the fluids specification may in many cases be unsuitable for the actual conditions in the well, requiring manual adjustments. However, ongoing developments in mechanization, robotics and real-time sensor technology for drilling fluids processing on the rig are making possible a revolution in fluids control. Based on such new and developing technology, we here consider the possibility of computer enabled closed loop real-time optimization of the drilling fluids property management on the rig.

Automated drilling fluids optimization may be performed within the constraints of the parameter ranges of the drilling fluid program, and with respect to more complex downhole behavior such as hole cleaning efficiency, affected by the combination of multiple fluids properties. Computationally inexpensive local optimization algorithms are sufficient for our purpose, requiring input from fluid compositional models. Such models are briefly reviewed. Standard interfaces are further required in order to apply input from third party models or correlation tables. We consider mixing scenarios where the domain of the optimization problem is constrained by the type and quantity of additives available on the rig site, and where automated sensors and computer controlled mixing technologies currently becoming available are assumed applied.

Examples of drilling fluids optimization control functionality are presented. The applied methods and algorithms can take advantage of highly instrumented surface installations, but may also be adapted to a more conventional set-up with limited real-time measurements, at the cost of lower accuracy. Correlations between composition and the various optimization targets may be visualized as an aid in the supervisory decision process. Requirements with regards to modelling and methods of control are discussed, and the status of development of enabling technology is presented, indicating also perceived technology gaps. Finally, results from simulations with a software prototype are reported.

The presented methodology complements recent efforts made in the field of drilling fluid mixing automation: while previous focus has often been on robotization of the process and on supervised control, such as maintaining a specified fluid density, the described approach allows for automatic redesign of the fluid composition. It is particularly well-suited for situations where down-hole conditions change rapidly, as they are accounted for in the optimization procedure.

You can access this article if you purchase or spend a download.