#TITLE_ALTERNATIVE#

• Main Author: HASNA (NIM: 12212009), HANIYYAH
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/27704
• Institution: Institut Teknologi Bandung

Matrix acidizing is applied primarily to remove damage caused by drilling, completion, and workover fluids and solids precipitated from produced water or oil (i.e., scale or paraffin). Removal of severe can result in very large increases in well productivity and permeability. Certain acids are usually used in acidizing operation, such as hydrochloric acid, hydrofluoric acid, acetic acid, and formic acid. Laboratory measurement to calculate the porosity enlargement and permeability change is not quite practical and takes a lot of time. Also, the heterogeneity of a reservoir causes a wide range of porosity value will need a lot of core sample and porosity measurement to analyze the effect of acidizing in porosity enlargement of formation. <br /> <br /> In this paper, Artificial Neural Network (ANN) techniques is applied to estimate the volume of minerals dissolved as the porosity enlargement after acidizing in sandstone formation. The data were gathered and then imported into the ANN models. The ANN model development training requires certain parameters, which are acid density, density and molecular weight, mineral density and molecular weight, gravimetric and volumetric constant, and pore volume. Feed-forward and back-propagation algorithm is used in building the ANN to modify the fitting in order to achieve higher prediction accuracy. ANN is proven to give very good accuracy in estimating the porosity enlargement. The effect of acid density on mineral dissolved is also applied and shows a direct proportional. Furthermore, the calculation of permeability change is held using the validation results of ANN and imported into Labrid (1975), Lund-Fogler (1976), and Lambert (1981) correlations.