Optimasi Desain Jacket Minimalis Berdasarkan Surrogate Model Dengan Metode Machine Learning

Syalsabila, Ferdita (2022) Optimasi Desain Jacket Minimalis Berdasarkan Surrogate Model Dengan Metode Machine Learning. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6020211009-Master Thesis.pdf Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Pemerintah Indonesia menargetken produksi 1 juta barel minyak dan 12 miliar standar kaki kubik gas per hari. Salah satu strategi untuk mencapai ini adalah dengan mempercepat pengembangan bidang marjinal yang sangat dipengaruhi faktor ekonomi. Optimasi pada tahap desain mengarah pada minimalisasi biaya produksi. Pada tesis ini, penulis akan mengoptimasi desain struktur jacket minimalis. Fungsi objektif optimasi adalah meminimumkan biaya struktur dengan membentuk surrogate model sebagai batasan. Variabel desain yang dipertimbangkan adalah ketebalan dan diameter luar bracing, serta ketebalan kaki jacket. Surrogate model yang dibuat dalam penelitian ini untuk mengevaluasi struktur dengan memprediksi hubungan antara variabel desain dengan batas kelelahan dan serviceability limit state. Kelelahan diestimasi dengan membuat persamaan faktor konsesntrasi tegangan atau stress concentration factor (SCF) untuk sambungan tubular tipe multiplanar DKT. Sedangkan untuk kekuatan yang dipertimbangkan seperti rasio tegangan maksimum. Metode untuk membentuk surrogate model menggunakan machine learning berupa radial basis function. Proses optimasi menggunakan genetic algorithm (GA). Fungsi aktivasi radial basis function yang cocok untuk evaluasi kondisi operasi berupa multi- quadric. Sedangkan fungsi aktivasi untuk evaluasi kelelahan sambungan kritis adalah fungsi kubik. Desain optimal menurunkan biaya sebesar 22% dengan rasio kontribusi tertinggi diberikan oleh member brace diagonal. Penelitian ini diharapkan mampu menjadi rekomendasi pendekatan model optimasi struktur minimalis yang dapat digunakan untuk perairan laut Indonesia.
==============================================================================================================================
The Indonesian government is working to achieve its production target of 1 million barrels of oil and 12 billion standard cubic feet of gas per day. One strategy to achieve this mission is to accelerate the development of marginal fields. Economic factors largely determine the marginal field development. In addition, of course, the technical aspects must remain capable and reliable. Op- timization in the design phase leads to production cost minimization. Currently, there are still many results of offshore structures that are overdesigned at a certain level. Therefore, this study analyzes the optimization of jacket structure to get an economical and reliable minimum jacket structure. Dimension optimization will be carried out on the main structure of the jacket and the objective function in the form of fabrication cost based on structural weight and the weld volume. Computational time can be minimized using surrogate models to evaluate the structure in each iteration. The surrogate model used radial basis function to predict the relationship between design variables with serviceability limit strength and the fatigue life. The optimization process used the genetic algorithm method. The radial basis function that suitable for the evaluation of operating conditions is multi-quadric. While the activation function for the evaluation of fatigue life is cubic function. The optimal design lowered costs by 22% with the highest contribution ratio provided by diagonal brace members. This research is expected to become a recommendation for a minimalist structure optimization model approach that can be used for Indonesian marine waters. This research is expected to recommend a minimum jacket structure optimization model approach that can be used especially for Indonesian waters.

Item Type:	Thesis (Masters)
Additional Information:	RTKe 627.98 Sya o-1 2022
Uncontrolled Keywords:	Genetic algorithm, Jacket minimalis, Optimasi, Radial basis function, Surrogate model. Genetic algorithm, Offshore jacket structure, Optimization, Radial basis function, Surrogate model.
Subjects:	T Technology > TC Hydraulic engineering. Ocean engineering
Divisions:	Faculty of Marine Technology (MARTECH) > Ocean Engineering > 38101-(S2) Master Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	06 Jul 2026 03:14
Last Modified:	06 Jul 2026 03:14
URI:	http://repository.its.ac.id/id/eprint/134330

Actions (login required)

View Item