New Approach to Predict the Filtrate Invasion of Nanoparticle-Based Drilling Mud Using Artificial Neural Network

Abstract During drilling of hydrocarbon reservoirs, loss of mud filtrate into the formations occurs due to the difference between mud hydrostatic and formation pressures. Filtrate invasion is a vital parameter that should be optimized to reduce formation damage. Recently, nanoparticles (NPs) — among different additives — have been thoroughly examined to minimize mud invasion and showed promising performance. Modeling the impact of NPs on the filtrate loss can fasten the process of selecting their optimum type, size, concentration, etc. to meet the drilling conditions. In this work, artificial neural network (ANN) was used to develop a model that can predict the filtrate invasion of nano-based mud under wide range of temperature and pressure up to 350 °F and 500 Psi, respectively. Seven types of nanoparticles with size and concentration ranges from 15 to 50 nm and 0 to 2.5 wt%, respectively, have been included in the model. Almost 2,863 data points were used to develop the ANN-model. Experimental work was conducted to collect 806 data points, whereas the rest were collected form the literature. The data set was divided into 70% for training and 30% for validating the model. A total of 6,750 different combinations for the model’s hyperparameters were evaluated to select the optimal combination. N-encoded method was used to convert the categorical data into numerical one. The model was evaluated through calculating the statistical parameters. The developed ANN-model showed high accuracy for predicting the filtrate loss at different pressures and temperatures. The obtained results showed that the average absolute relative error (AARE) is less than 0.5%, and coefficient of determination (R2) is more than 0.99 for the overall data. The developed ANN-model covers wide range of pressures and temperatures. Moreover, it covers various NPs’ types, concentrations, and sizes, which confirms its useability and coverability.