Claim Missing Document
Check
Articles

Found 5 Documents
Search

STUDY OF PERMEABILITY PREDICTION USING HYDRAULIC FLOW UNIT (HFU) AND MACHINE LEARNING METHOD IN “BSH” FIELD Babas Samudera Hafwandi; Dedy Irawan; Amega Yasutra
PETRO: Jurnal Ilmiah Teknik Perminyakan Vol. 12 No. 2 (2023): JUNI
Publisher : Jurusan Teknik Perminyakan Fakultas Teknologi Kebumian dan Energi Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/petro.v12i2.15763

Abstract

In this study, Decision Tree, Gradient Boosting, AdaBoost, Random Forest, Support Vector Machines, and K-Nearest Neighbor Machine Learning model are presented that use log and core data available as the basis for permeability prediction. The results were then compared to previously available method, namely Hydraulic Flow Unit (HFU) based on MAE and RMSE Value. The approach was taken by considering correlation relationships between existing log data in predicting permeability values. Three correlations, namely Spearman, Pearson, and Kendall, will be used to determine the relationship between existing log data and permeability. The machine learning model is then compared with the Hydraulic Flow Unit (HFU) Method in predicting the permeability value. The Novelty of this Machine Learning Model is to be able to predict permeability value, to solve the problem of accuracy using the existing method, and to save reasonable time to obtain permeability value by coring in the laboratory by utilizing standard computer available.
ANALISIS KEBERHASILAN KERJA ULANG PINDAH LAPISAN PADA SUMUR SKW-33 LAPANGAN SUKOWATI Desyana Nindya Prastiwi; Abror, Hadziqul; Hafwandi, Babas samudera
Journal of Sustainable Energy Development Vol. 2 No. 2 (2024): Journal of Sustainable Energy Development (JSED)
Publisher : Petroleum Engineering, Faculty of Engineering, University of Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19184/jsed.v2i2.1603

Abstract

This study explores the effectiveness of re-perforation operations at Well SKW-33 in the Sukowati Field to enhance oil production. The background highlights the decline in production rates despite the well's initial success with natural flow since June 2015. The primary objective is to evaluate the impact of re-perforation on oil production and water cut. The research employs methods such as squeeze cementing, Cement Bond Log (CBL) analysis, well logging and subsequent re-perforation. The results demonstrate a significant increase in oil production from 100-150 barrels per day (bopd) to a peak of 305 bopd post-re-perforation. Additionally, the water cut reduced from 95-96% to 80%, indicating a successful reduction in water production. The conclusion asserts that re-perforation significantly boosts short-term oil production and reduces water production, though ongoing management is essential for maintaining long-term efficiency.
Studi Prediksi Porositas Dengan Menggunakan Metode Deterministik dan Machine Learning Pada Lapangan “X” Hafwandi, Babas Samudera
Journal of Sustainable Energy Development Vol. 3 No. 1 (2025): Journal of Sustainable Energy Development (JSED)
Publisher : Petroleum Engineering, Faculty of Engineering, University of Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Porosity is one of the most critical parameters in reservoir characterization, as it directly influences hydrocarbon storage capacity. Accurate porosity prediction becomes even more essential in fields with limited core data, such as Field “X”, located in the South Sumatra Basin. This study compares two different porosity prediction approaches: a deterministic method based on well log interpretation using NPHI and RHOB logs, and various Machine Learning (ML) algorithms, including Random Forest (RF), K-Nearest Neighbor (KNN), Gradient Boosting (GBR), AdaBoost (ADA), Support Vector Machine (SVM), and Decision Tree (DT). Data preprocessing involved feature selection using Pearson, Spearman, and Kendall correlation coefficients to identify the most influential log parameters. The dataset was then divided into training (70%) and testing (30%) subsets. Model performance was evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The deterministic method yielded an MAE of 0.0658 and RMSE of 0.0906, while the best ML model, Random Forest, achieved an MAE of 0.0329 and RMSE of 0.0434 on the testing dataset. In conclusion, Machine Learning, especially the Random Forest model, proves to be a more reliable and accurate tool for porosity prediction in geologically complex fields, offering significant potential for enhancing reservoir modeling and field development planning.
STUDI PREDIKSI TIPE REKAHAN PADA RESERVOIR VULKANIK JATIBARANG LAPANGAN “TGB” DENGAN METODE MACHINE LEARNING Hafwandi, Babas Samudera
Petro : Jurnal Ilmiah Teknik Perminyakan Vol. 14 No. 4 (2025): Desember 2025
Publisher : Jurusan Teknik Perminyakan Fakultas Teknologi Kebumian dan Energi Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/petro.v14i4.24179

Abstract

Reservoir vulkanik dikenal memiliki heterogenitas yang tinggi akibat aktivitas geologi kompleks seperti aliran lava, pengendapan tuf, serta proses alterasi hidrotermal. Salah satu karakteristik penting dalam reservoir ini adalah keberadaan rekahan alam, yang secara langsung memengaruhi permeabilitas dan produktivitas fluida. Rekahan dapat terbentuk secara alami maupun akibat proses tektonik dan diklasifikasikan menjadi beberapa tipe, seperti rekahan konduktif (conductive), rekahan tertutup (sealed), dan rekahan campuran (mixed). Penentuan tipe rekahan secara konvensional memerlukan interpretasi dari log citra borehole dan analisis deskripsi core, yang bersifat subjektif, memakan waktu, dan mahal. Penelitian ini bertujuan untuk mengembangkan pendekatan berbasis machine learning untuk memprediksi tipe rekahan pada Reservoir Vulkanik Jatibarang menggunakan data log dan data Well Image Log. Lima algoritma digunakan: Random Forest (RF), Gradient Boosting Machine (GBM), CatBoost, Support Vector Machine (SVM), dan K-Nearest Neighbor (KNN). Validasi dilakukan menggunakan akurasi, precision, recall dan F1-score. Hasil menunjukkan bahwa algoritma Random Forest menghasilkan akurasi tertinggi  dengan nilai Accuracy sebesar 0.8678, Precision sebesar 0.8678, Recall sebesar 0.8870 dan F1 sebesar 0.8756 (89,2%), pada data Blind Testing. dengan fitur paling penting adalah  DEPTH, GR, ILD, NPHI, dan DT . Studi ini membuktikan bahwa machine learning dapat digunakan sebagai metode alternatif yang cepat dan akurat dalam klasifikasi rekahan, membantu proses evaluasi reservoir dan pengambilan keputusan pengembangan lapangan.
PENENTUAN CADANGAN AWAL ISI MINYAK DENGAN MENGGUNAKAN SOFTWARE  MATERIAL BALANCE PADA RESERVOIR TGB Muhammad Dimas Adiguna; Babas Samudera Hafwandi; Lindia Heviyanti; Ardipratama Ilham Vembriatmaja
Petro : Jurnal Ilmiah Teknik Perminyakan Vol. 15 No. 1 (2026): Maret 2026
Publisher : Jurusan Teknik Perminyakan Fakultas Teknologi Kebumian dan Energi Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/petro.v15i1.25526

Abstract

Reservoir TGB is a sandstone reservoir that began production in December 2019 and, as of November 2024, has been developed by two producing wells with a cumulative oil production of approximately 374 MSTB. Along with the increasing production activities and the need for field development planning, an evaluation of hydrocarbon reserves is required to assess the reservoir’s future production potential. This study aims to determine the Original Oil in Place (OOIP), identify the reservoir drive mechanism, and estimate the Estimated Ultimate Recovery (EUR) and remaining reserves of Reservoir TGB using the material balance method. The material balance analysis was conducted using oil, gas, and water production data, reservoir pressure data, and fluid property (PVT) data. History matching was performed using MBAL software to achieve consistency between the reservoir model and actual production behavior. Furthermore, the reservoir drive mechanism was evaluated through energy plot analysis and reservoir performance assessment. Decline curve analysis was subsequently carried out using OFM software to estimate the productive life and ultimate recovery potential of the reservoir. The results indicate that the OOIP of Reservoir TGB is approximately 3,000 MSTB. The evaluation of the drive mechanism shows that the reservoir is dominated by a water drive system. Based on the decline curve analysis, the EUR is estimated at 613.7 MSTB, with remaining reserves of 239.08 MSTB and a recovery factor of approximately 20.46%. These results indicate that a significant portion of hydrocarbons remains unrecovered. Therefore, further field development through the addition of production wells is recommended to improve oil recovery and optimize the utilization of the remaining reserves in Reservoir TGB.