On the Online Calibration of Drilling Fluid Rheological Parameters Using EMD and MLE

Drilling fluid rheology is critical for maintaining wellbore stability and controlling bottom-hole pressure. Currently, manual methods such as rotational viscometers are used on drilling sites despite their susceptibility to human error. Existing tube viscometer online measurement devices suffer from poor measurement accuracy and stability due to factors like mud diaphragm pump pulsation, motor vibration, and pressure difference bias errors. To address these challenges, this article proposes an online calibration method based on empirical mode decomposition (EMD) and maximum likelihood estimation (MLE) using a double-tube differential pressure tube viscometer. First, the EMD model uses measured differential pressure sensor data to extract and eliminate interference signals. Next, a maximum likelihood model calibrates the drilling fluid parameters. Finally, an experimental platform was built for real-time monitoring of drilling fluid performance. Results indicate that calibrated apparent viscosity, plastic viscosity, and dynamic shear force show better precision with relative errors less than 5% compared to uncalibrated parameters.