Zahra, Haya Aqilah (2024) Optimisasi Rate of Penetration (ROP) pada Operasi Drilling menggunakan Predictive Modelling dan Particle Swarm Optimization (PSO) untuk Meminimalkan Waktu dan Biaya (Studi Kasus : Sumur X Lapangan Mudi). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5009201062-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5009201062-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (6MB) | Request a copy |
Abstract
Operasi pemboran diketahui merepresentasikan 30% dari total biaya produksi pada sumur minyak dan gas. Biaya pemboran memiliki hubungan yang erat dengan waktu pemboran, yang mana semakin singkat waktu pemboran, maka biaya pemboran akan semakin murah, dan begitu juga sebaliknya. Parameter utama yang mempengaruhi secara langsung waktu pemboran adalah Rate of Penetration (ROP). Untuk itu, dalam memecahkan masalah biaya dan waktu pemboran, dilakukan penelitian untuk mengoptimasi ROP dengan mempertimbangan tiga parameter utama, yaitu Weight on Bit (WOB), Rotation per Minute(RPM),dan Flowrate. Metode Predictive Modelling dan Particle Swarm Optimization (PSO) diterapkan untuk mengoptimisasi ROP. Metode Predictive Modelling merupakan metode data-driven based yang menggantikan persamaan tradisional. Penelitian ini menggunakan empat model regresi, memungkinkan model untuk dapat mengidentifikasi hubungan kompleks antara parameter pemboran dengan membaca data
historis yang diberikan. Empat algoritma predictive modelling yang disimulasikan lalu dievaluasi menggunakan nilai RMSE,R², MAE, dan MAPE. Random Forest Regressor dipilih sebagai model yang paling akurat dengan nilai R² mencapai 0.92. Optimisasi ROP memberikan hasil yang signifikan, yaitu pengurangan waktu pemboran sampai 6.2 hari dan biaya hingga Rp1,812,531.93 atau 16% lebih murah dari biaya aktual.
=====================================================================================================================================
Drilling operations are known to represent 30% of the total production cost in oil and gas wells. Drilling costs are closely related to drilling time; the shorter the drilling time, the cheaper the drilling costs, and vice versa. The primary parameter that directly influences drilling time is the Rate of Penetration (ROP). Therefore, to address the issues of drilling costs and time, a study was conducted to optimize ROP by considering three main parameters: Weight on Bit (WOB), Revolutions per Minute (RPM), and Flowrate. Predictive Modeling and Particle Swarm Optimization (PSO) methods were applied to optimize ROP. Predictive Modeling is a data-driven method that replaces traditional equations. This study utilized four regression models, allowing the model to identify complex relationships between drilling parameters by reading the provided historical data. Four predictive modeling algorithms were simulated and then evaluated using RMSE, R², MAE, and MAPE values. Random Forest Regressor was selected as the most accurate model with an R² value of 0.92. The ROP optimization yielded significant results, reducing drilling time by up to 6.2 days and costs by Rp1,812,531.93 or 16% cheaper than actual costs.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Machine Learning, Optimisasi, Pemboran, Predicitive Modelling, Optimization, Drilling |
| Subjects: | T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.62 Simulation T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | Haya Aqilah Zahra |
| Date Deposited: | 01 Aug 2024 01:35 |
| Last Modified: | 09 Sep 2024 08:26 |
| URI: | http://repository.its.ac.id/id/eprint/109285 |
Actions (login required)
 |
View Item |