CD Skripsi

Prediksi Fluid Rate Untuk Pompa Sucker Rod Pump Di Lapangan X Wk Rokan Menggunakan Algoritma Random Forest

---

The research focuses on the development of a fluid rate prediction method for SRP pumps in Rokan's X WK Field using Random Forest algorithms. The field plays a crucial role in global energy production, with more than 85 active wells and thousands of artificial extraction systems involved. The analysis involved more than 2600 Online Dyno, 1500 ESP Scada monitoring, and more than 150,000 well tests per year. SRP pumps are considered a critical element in oil and gas production operations, and accurate predictions of fluid rates can provide valuable information for production management. This study compared the use of default parameters with the use in the Random Forest algorithm of a tuning hyperparameter. Evaluations showed improvements in performance with the application of a hyperparameters tuning, although the differences were small. R2 scores increased by 0.05% on test data, while MAPE scores decreased by 0.573%. Model development results using training data, correlation analysis, and user experience interviews showed that the model was able to make predictions with a score of 0.9415 and an error rate of 17.36%. Advanced testing was carried out with new data. However, there are differences in accuracy levels in the prediction model, indicating fundamental differences between the two datasets.

Keywords: Sucker Rod Pump Pump, Fluid Rate Prediction, Random Forest Algorithm, Spearman Correlation Analysis

---

Ketersediaan

Tidak ada salinan data

Informasi Detail

Judul Seri

-

No. Panggil

2003111141

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003111141

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar