Researchers from King Fahd University of Petroleum & Minerals used artificial neural network (ANN) to develop new models to predict the rheological properties of calcium chloride brine-based mud
Drilling fluid is used to aid the drilling of boreholes into the earth. Conventional drilling fluids are classified as water-based, oil-based, and synthetic-based fluid systems. These fluids aid in enhancing performance of drilling process under certain temperatures and pressures. Special measures are taken to drill reservoir section to avoid damaging the reservoir and plugging the reservoir pores. Reservoir drill-in fluids (RDFs) are specially formulated to maximize drilling experience and protect the reservoir from being damaged. ANN is one of the most common Artificial Intelligence (AI) techniques that has the ability to deal with different engineering problems. Now, a team of researchers from King Fahd University of Petroleum & Minerals used artificial neural network (ANN) to develop new models to predict the rheological properties of calcium chloride brine-based mud.
The team used the models to predict the rheological properties of CaCL2 brine-based drill-in fluid in a real-time. The team also predicted plastic viscosity (PV), apparent viscosity (AV), yield point (Yp), flow behavior index (n), and flow consistency index (k). 515 field data measurements of mud weight (MW) and Marsh funnel viscosity (MF) in ratios 70:30 were used to train and validate the ANN models respectively. The team found that the developed correlations can offer the ability for real-time monitoring of the hole cleaning performance. They can also detect any abnormal changes in the normal trends to elude interrupting problems, thereby aiding to predict several drilling problems. This in turn can help to save the drilling cost while enhancing drilling operations.
Moreover, the models were helpful in the calculations of rig hydraulics, surge and swab pressures, and equivalent circulating density. The extracted links from the developed ANN models help to predict the rheological properties of CaCL2 brine-based mud directly without the need to run the models. Moreover, the team found that the optimized parameters produced the highest accuracy and the lowest error. The team concluded that the ANN models can predict the rheological parameters in real time based on MW and MF with high accuracy. The research was published in the journal MDPI Energies on May 17, 2019.
Subscribe to our newsletter to get notification about new updates,information, discount, etc..