Pratama, Fajar Rieski (2023) Estimasi Water Breakthough Time Untuk Mengantisipasi Terjadinya Coning Phenomenon Dengan Menggunakan Artificial Neural Network Algorithm. Other thesis, Universitas Islam Riau.

Text 183210022.pdf - Submitted Version Restricted to Registered users only Download (7MB) | Request a copy

Abstract

The coning phenomenon is an event or phenomenon where the fluid that is the driving fluid from a reservoir such as water and gas enter the production zone and forms a tent. In this case, of course, the phenomenon or event is something that must be avoided or anticipated before it occurs due to losses or problems that arise as a result of these phenomena, such as a decrease in the rate of production of oil and fluids which are more dominantly produced by gas or air and there are still other losses. to be obtained. Before the coning phenomenon occurs, a breakthough condition occurs which is a reservoir condition where fluid that was previously isolated or separated from the production zone gains access and will break into the productive zone. In an effort to anticipate the occurrence of these coning phenomena, of course, it is necessary to know when these conditions and events began to occur. Because it is necessary to know the time of water penetration before the coning phenomenon occurs in order to take or make efforts in anticipation to prevent this from happening. In this study, the authors will use an Artificial Neural Network Algorithm to estimate or predict the resting time of water as a function of production rate and physical model properties. The results of this method will be compared with the previous method to see and verify the results that have been obtained. The data used is a combination of literature obtained with experimental data generated in this study. The conclusions obtained in this study are based on the simulation results and discussion and analysis of the simulations that have been carried out, it can be concluded that the higher the flow rate, the faster a field reaches the break-to-eight time condition, this can be seen from the predictions generated by ANN and CMG Numerical Simulation are almost the same where an average model accuracy is obtained of 0.99 with predicted results Qo = 800 Bbl/day obtained TBT = 726 days, Qo = 1500 Bbl/day obtained TBT = 283 days, Qo = 2500 Bbl/day obtained TBT = 128 days, Qo = 3500 Bbl/day, TBT = 77 days, Qo = 5000 Bbl/day, TBT = 45 days.

Actions (login required)

View Item