Real-time control and monitoring of drilling fluids density by fuzzy-based control system

Abstract

The drilling fluids play a significant role in the oil drilling process, being its physicochemical properties the focus of the operation control. Nowadays, the control methods are usually based on manual dosing of additives by the specialized operators, where the recipes are obtained by pilot or bench tests, demanding a greater amount of raw material, increasing the operation time and costs. Control strategies of drilling fluids properties based on fuzzy logic may contribute to a high reliability and safety for human resources in drilling operations, a higher process efficiency and the mitigation of environmental impacts. This research aimed to propose and develop an automated flow loop system for fluid preparation with real-time monitoring, control, and self-correction of the drilling fluid density. The control strategy was based on fuzzy logic and used the Mamdani-type inference rules as the basis for developing a Fuzzy-PI controller. The performance of the proposed Fuzzy-PI controller was compared to a classic PI controller in disturbance-rejection and setpoint-tracking control problems. Beyond efficient control, the Fuzzy-PI controller enabled a reduction in raw material consumption of 63% for barite and 38% for water, in the disturbance-rejection control scenario; and 21% for barite and 47% for water, in the setpoint-tracking control scenario.