Reza Amanda, Irsan (2023) Studi Sensitivitas Fracture Half Length, Effective Permeability Dan Fracture Width Pada Hydraulic Fracturing Menggunakan X - Gradient Boosting Algorithm. Other thesis, Fakultas Teknik Perminyakan.

Text 183210874.pdf - Submitted Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Implementation of Hydraulic fracturing done by increasing the permeability value of the shale gas that has range low value. The parameters tested in this study include : Pad Volume, Fracture Propant Concentration, Injecton Rate, Fluid Volume, and Fracture Width. Of the five parameters above carried out generate use FRACCADE 7.0 and found results in the form of values Half Length, Effective Permeability and Fracture width which will then be analyzed using CMOST and Machine Learning. Research using X – Gradient Boosting stated that Half Length is the most influential parameter in hydraulic fracturing with rate importance of 2.995416, greater than Effective Permeability of 1.879223 and fracture width of 0.414047 then the results are displayed predictive modelling with 3 variations ratio comparison training and testing form data ratio Training (0.7) : Testing (0.3) with R value2 worth 0.852, MAE 0.00 and MSE 20.03,Training (0.8) : Testing (0.2) with R value2 worth 0.999, MAE 0.00 and MSE 9.05,ratio Training (0.9) : Testing (0.1) with R2 value worth 0.718, MAE 0.00 and MSE 9.05. of results training and testing From these data it can be concluded that’s tagging with ratio Training (0.8) : Testing (0.2) is the predictive model with the best ratio with half length being the most sensitive parameter compared to effective permeability and fracture width.

Actions (login required)

View Item