Data-Driven Gas Lift Control untuk Optimasi Produksi Jaringan Sumur Minyak pada Model Dinamika Sumur Gas Lift Berbasis Neural Network

Wibowo, Ananda Cahyo (2023) Data-Driven Gas Lift Control untuk Optimasi Produksi Jaringan Sumur Minyak pada Model Dinamika Sumur Gas Lift Berbasis Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07111940000128_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (9MB) | Request a copy

Abstract

Pemerintah melalui Satuan Kerja Khusus Minyak dan Gas Bumi (SKK Migas) memiliki visi agar industri hulu migas dapat mencapai target produksi 1 juta barel per hari (BOPD) di tahun 2030 dengan strategi optimalisasi produksi lapangan existing/sumur-sumur tua. Namun teknologi artificial lift, khususnya gas lift, dalam memaksimalkan oil production rate memerlukan workflow dan proses akuisisi data yang panjang. Proses adjusting Gas Lift Injection Rate (GLIR) yang optimal berdasarkan kurva Gas Lift Performance Curve (GLPC) dari sumur minyak masih dilakukan manual dan tidak berkala – padahal dinamika dari sumur terus berubah. Pada tugas akhir ini dirancang dan disimulasikan kontrol adaptif untuk meningkatkan produksi minyak dari jaringan sumur dengan bantuan gas lift. Dinamika sumur OA – 11 dan OA – 12 dimodelkan menggunakan metode Neural Network dari data production well testing PT. Pertamina Hulu Mahakam. Penentuan GLIR optimal didapat dari estimasi kurva GLPC menggunakan algoritma regresi polynomial orde 2 dan optimisasi SLSQP. Peningkatan produksi minyak kedua sumur tersebut disimulasikan berbasis Hardware-in-the-loop (HITL) simulation menggunakan virtual instrumentation yang dihubungkan oleh protokol Modbus dan disajikan dalam Graphical User Interface (GUI). Penelitian ini berhasil mengestimasi peningkatan produksi minyak rata-rata dikisaran 13-20% lebih banyak dengan tingkat akurasi nilai R2 model sumur miyak berbasis LSTM neural network antara 88-92 %.
==================================================================
The government through the Special Task Force for Oil and Gas (SKK Migas) has a vision that the upstream oil and gas industry can achieve a production target of 1 million barrels per day (BOPD) in 2030 with a strategy of optimizing production of existing fields/old wells. However, artificial lift technology, especially gas lift, in maximizing the oil production rate requires a lengthy workflow and data acquisition process. The process of adjusting the optimal Gas Lift Injection Rate (GLIR) based on the Gas Lift Performance Curve (GLPC) from oil wells is still done manually and not periodically – even though the dynamics of the wells are constantly changing. In this final project, adaptive control is designed and simulated to increase oil production from a network of oil wells with the help of gas lift technology. The dynamics of the ‘OA – 11’ and ‘OA – 12’ wells are modeled using the Neural Network method from PT. Pertamina Hulu Mahakam. Determination of the optimal GLIR is obtained from the estimation of the GLPC curve using a 2nd order polynomial regression algorithm and Sequential Least Square Quadratic Programming (SLSQP) optimization. The increased oil production of the two wells is simulated based on Hardware-in-the-loop (HITL) simulation using virtual instrumentation connected by the Modbus protocol and presented in a Graphical User Interface (GUI). This study succeeded in estimating an average increase in oil production in the range of 13-20% with an accuracy level using the R2 value of the oil well model using LSTM neural network between 88-92%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Data-Driven Gas Lift Well Optimization, Hardware-in-the-loop simulation, Virtual Instrumentation, Neural Network System Identification
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Electrical Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Ananda Cahyo Wibowo
Date Deposited:	26 Jul 2023 06:30
Last Modified:	26 Jul 2023 06:30
URI:	http://repository.its.ac.id/id/eprint/99351

Actions (login required)

View Item