Predicting Perofrmance of SiO2 Nanoparticles on Filtration Volume Using a Comprehensive Approach: Application in Water-Based Drilling Fluids

In drilling operation, duo to inhomogeneous and low quality of filter cake, filtration loss occurs which causes formation damage and other problems such as crack propagation and borehole instability. Filtration losses significantly increase the costs and risks of drilling in the industry. Then drilling fluid should have low filtration rate to reduce the problems mentioned above. This is achieved by reducing fluid loss by thin and low permeability filter cake on the wellbore. Recently nanoparticles are used to mitigate or solve this problem because they provide low drilling fluid loss while maintaining and also improving other rheological properties of drilling fluid. Proposing a model for predicting the performance of nanoparticles on filtration volume is preferred as the filtration loss experiments is costly and time consuming. For this purpose, an artificial neural network (ANN) and an adaptive neuro-fuzzy interference system (ANFIS) are developed to model the drilling fluid filtration volume as a function of drilling fluid properties, SiO2 nanoparticles and KCl salt concentration. These models are compared to each other using statistical parameters including correlation coefficient (R2), mean squared error (MSE), and average absolute relative error (AARE). The results show that both models predict the filtration volume well and are in agreement with the actual measured values. The ANN and ANFIS model fit the experimental results with AARE of 4.74% and 6.05%, MSE of 0.2630 and 0.3521, and R2 of 0.9926 and 0.9900 respectively. This indicates that these models are accurate enough for predicting the nanoparticles performance on filtration volume,

کلیدواژه ها:

Filtration volume ، SiO2 Nanoparticles ، KCl salt ، ANN ، ANFIS ، Modeling