Alireza Golsefatan - Student, Petroleum Engineering Department, Petroleum University of Technology
Khalil Shahbazi - Associate Professor, Petroleum Engineering Department, Petroleum University of Technology,

Water-based drilling fluid mainly consists of water as base fluid, inert and reactive solids as additives which still has many disadvantages including formation damage due to fluid losses. Filtration loss of the drilling fluid in the reservoir is one of the main causes of crack propagation and borehole instability. Filtration losses significantly increase the costs and risks of drilling around the world. Recently, using nanoparticles as fluid loss additive is one of the new methodologies in drilling industry for overcoming related issues in drilling fluid such as minimizing formation damage, controlling fluid loss, improving wellbore stability and subsequently improving drilling performance. Performing the filtration loss experiments is costly and time consuming; therefore, proposing a model for predicting the performance of nanoparticles on filtration volume is highly suggested. For this purpose, an adaptive neuro-fuzzy interference system (ANFIS) model is developed to model the drilling fluid filtration volume as a function of drilling fluid properties, SiO2 nanoparticles and KCl salt concentration. The results show that the predicted filtration volumes by ANFIS are in agreement with the actual measured values. The presented model fits the experimental results with correlation coefficient (R2) of 0.996 and average absolute relative error (AARE) of 6.05% which indicates that the ANFIS model is a good model for predicting the nanoparticles performance on filtration volume.

Filtration volume, SiO2 Nanoparticles, KCl salt, Modeling