I am also grateful to the faculty and staff of the School of Mining and Geosciences at Nazarbayev University, who have provided me with resources and opportunities to pursue my academic interests. Any contribution to the research from others with whom I have worked at NU or elsewhere is expressly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that the assistance of others in the conception and design of the project or in style, presentation, and linguistic expression is acknowledged.
Filter cake formation is an integral part of the drilling process and provides wellbore stability and eliminates mud circulation losses. Due to unique properties such as small size (nanoscale) and large surface area, nanoparticles show high penetration rate and filter cake removal efficiency. This study aims to estimate the effectiveness of two nanoparticle-based breakers, namely silica and titanium dioxide, using numerical simulation by studying the size distribution and return permeability.
Additionally, a statistical analysis was performed to predict the reverse permeability after nanocrusher treatment using MATLAB Machine Learning tools. An extension of the reverse permeability study could consider formation damage before and after nano-based breaker treatment.
Introduction
- Background
- Problem Statement
- Aims and Objectives
- Thesis Structure
The use of nanoparticles (NPs) to remove filter debris is an emerging technology in the petroleum industry. For example, in a study done by (Prakash et al., 2021), researchers tested the effectiveness of silica nanoparticles in removing filter cake from a well. Nanoparticles can be used in a variety of ways to remove filter cake from wells, including as additives to drilling fluids, as preflush or postflush treatments, or as stand-alone treatments.
Selection of the appropriate nanoparticle and treatment strategy depends on the specific properties of the filter cake, drilling conditions, and well cleanup objectives. Therefore, this thesis answers the following questions: how effective are nanoparticle-based breakers in removing synthetic filters from the well and what is the best way to optimize their flow using numerical analysis. The research focuses on using numerical analysis to validate empirical results and optimize decision-making towards the application of nanoparticle-based switches.
The structure of the dissertation includes five chapters, with the first chapter introducing the background of the topic, the problem statement and the objectives of the research. The materials, experimental design, and methodology used to achieve the objectives of the study are developed in Chapter 3.
Literature Review
- Synthetic Drilling Muds
- Filter Cake Formation
- Nanoparticle-based Breakers
- Numerical Simulation
2021) proposed a new formulation of water- and oil-based drilling fluids to remove barite filter cake. Filter cake thickness after filtration of conventional (old) and new formulation of drilling fluids (Tariq et al., 2021). Similar to Tariq et al. 2021), Rostami & Nasr-El-Din (2010) investigated the self-destruction of filter cake formed by water-based drilling fluid.
The researchers evaluated the effect of temperature, soaking time, and particle size on filtration loss and filter cake removal. Filter cake removal efficiency increased with temperature, however, increasing temperature also caused drilling fluid thickening due to enhanced PLA hydrolysis reaction and subsequent mud dehydration. Additionally, soaking tests showed that PLA required considerable time (20 hours) to remove 80% of the filter cake.
2015) tested two chelating agents, namely NTA and EDTA, in removing filter cake formed by oil-based mud. When a liquid containing suspended particles or contaminants is passed through a porous medium, the particles can accumulate on the surface of the medium, creating a layer of material known as the filter cake. The inlet velocity also had an effect on the interference factor and porosity of the filter cake.
The lower the mean particle size, the thicker the filter cake formed. Furthermore, lower filtration losses were observed in drilling fluid with SiO2 nanoparticles, and therefore a denser filter cake was formed in this case. The researchers also mentioned that surface modification improved the filter cake removal efficiency and the stability of nanodispersions.
The properties and performance of the filter cake were found to be affected by the applied pressure, the coefficient of friction and the cohesive energy density. When the applied pressure was higher, the filter cake became more compressed, resulting in a greater pressure drop. The study also showed that thick filter cake is formed by larger particles at high pressure due to compaction of particles in the filter cake structure.
It was found that the higher the porosity of the sandstone sample, the wider the pores formed in the filter cake. When attractive forces prevailed, flocculation occurred, resulting in a loose packing of the filter cake with high compressibility and increased porosity.
Methodology
- Modeling parameters
- Experimental setup
- Synthetic Data
- Nanoparticle Density and Size Distribution
- Assumptions
- Numerical Investigation
- Artificial Neural Network Method
- Multiple Linear Regression Analysis
The physics-driven synthetic data set consisting of 500 samples (iterations) as presented in Table 2 explains the induced pressure for filtration, the density of nanoparticles under fluidity and permeability return based on the chemical treatment of the filter cakes. Nanoparticles are highly influential in determining the filtration and rheological properties of drilling muds, and this study focuses on the permeability return of filter cakes treated with silica and titanium dioxide breakers. However, the density of 1.87 g/cm3 of the nanoparticles (Kimoto et al., 2017) under investigation improves optimal fluid-particle dispersion.
However, the size distribution of the nanoparticles can influence the structure and properties of the resulting filter cake. If the nanoparticles have a narrow size distribution, the resulting filter cake can have a more uniform structure with fewer defects. On the other hand, if the nanoparticles have a broad size distribution, the resulting filter cake may have a more heterogeneous structure.
In addition, the size of the nanoparticles can also affect the porosity and permeability of the filter cake. The size and distribution of nanoparticles were spherical and uniformly distributed over the surface of the filter cake. The initial permeability of K1, which consists of oil and ceramic disc under pressure, is 20 D.
It is a value between 0 and 1 and is often expressed as a percentage where 1 indicates that the model explains all the variation. Network architecture: The next step is to define the architecture of the neural network, including the number of layers, nodes and activation functions. During training, the weights of the network are adjusted to minimize the difference between the predicted output and the actual output.
The intercept represents the expected value of the dependent variable when all independent variables are zero. This is typically done using a method called ordinary least squares regression, which minimizes the sum of the squared errors between the predicted values and the actual values of the dependent variable. In forecasting, the goal is to use the model to predict the value of the dependent variable for new observations based on the values of the independent variables.
Results and Discussion
Prediction of Return Permeabilty Using Artificial Neural Network
XN are the independent variables (also called the explanatory variables or predictors), b0 is the intercept and b1, b2. The regression coefficients represent the change in the dependent variable that occurs when the corresponding independent variable changes by one unit, while all other independent variables are held constant. The process of fitting a multiple linear regression model involves estimating the regression coefficients and the intercept that best fits the data.
In inference, the purpose is to use the model to test hypotheses about the relationship between the dependent variable and the independent variables.
Prediction of Return Permeabilty Using Multiple Linear Regression
Return Permeability Optimization
The physics behind this is as follows: as the flow rate increases, more concentration of nanoparticle breakers is introduced into the pores of the filter cake driving the chemical reaction between the nanobreaker particles and the filter particles. Moreover, the statistical analyzes (ANN and MLR) presented in the research proposal serve as an alternative way to conventional means to evaluate the removal efficiency of nanoparticle-based switches by predicting the return permeability of media after treatment. Both ANN and MLR models showed a good performance in predicting return permeability from synthetic data.
The study also showed the effect of pressure and flow rate on return permeability and provided insight into the physics behind the phenomenon. For further research purposes, the influence of heterogeneous pore geometry on return permeability and accuracy of applied models can be investigated. Evaluation of rheological and filtering properties of a water-based polymeric drilling mud in the presence of nano additives at different temperatures.
Evaluation of the effects of SiO2, ZnO and TiO2 nanoparticles on the rheological properties and clay inhibition of water-based drilling mud. Effect of high temperature aging on TiO2 nanoparticle enhanced drilling fluids: A rheological and filtration study. Retrieved April 8, 2023, from https://www.ofite.com/products/drilling-fluids/filtration/product/1508-ceramic-filter-disk-55-micron.
An experimental investigation into the enhanced oil recovery and improved performance of drilling fluids using titanium dioxide and fumed silica nanoparticles. Filter cake formation on the vertical well at high temperature and high pressure: computational fluid dynamics modeling and simulations. Removal of barite scale and barite-weighted water- or oil-based drilling fluid residues in a single stage.
Effect of particle number density on rheological properties and barite subsidence in oil-based drilling fluids. Effect of silica nanoparticles on the rheological and HTHP filtration properties of environmentally friendly additive in water-based drilling fluid. Primary Evaluation of Filter Cake Breaker in Biodegradable Synthetic-Based Drilling Fluid (Issue April).
