Analisis Prediksi Faktor Intensitas Tegangan Pada Sambungan Tubular Jacket Platform Berbasis Surrogate Model

Hardian, Muhammad Akbar (2023) Analisis Prediksi Faktor Intensitas Tegangan Pada Sambungan Tubular Jacket Platform Berbasis Surrogate Model. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 04311940000073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (7MB) | Request a copy

Abstract

Berdasarkan data SKK migas pada tahun 2016, 54,65% anjungan lepas pantai di Indonesia telah berumur lebih dari 20 tahun. Struktur yang telah melebihi umur operasinya perlu ditinjau ulang dari segi kekuatan struktur apakah masih mampu untuk beroperasi. Dalam penelitian ini, penulis akan berfokus pada analisis prediksi faktor intensitas tegangan pada struktur jacket berkaki empat berbasis surrogate model. Faktor intensitas tegangan merupakan faktor yang menentukan kelelahan pada sambungan tubular dengan metode fracture mechanics. Dalam rangkat meningkatkan akurasi dan mengoptimalkan waktu analisis dikembangkan surrogate model dari analisis variasi retak yang didapat dengan metode elemen hingga. Metode analisis yang digunakan dalam penelitian ini meliputi analisis statis inplace untuk analisis tegangan struktur secara global, analisis lokal sambungan tubular kritis, analisis retak pada tubular dengan titik tegangan kritis tertinggi, dan pemodelan faktor intensitas tegangan berbasis surrogate model menggunakan machine learning model SVM berdasarkan variasi kedalaman retak dan panjang retak. Analisis retak menggunakan 30 variasi model retak yang berada pada tegangan maksimum di analisis lokal metode elemen hingga. Analisis lokal struktur berdasarkan hasil analisis statis inplace global struktur dengan sambungan kritisnya adalah sambungan tubular multiplanar DKT. Diperoleh tegangan von-mises tertinggi pada brace 5 dengan tegangan sebesar 327 MPa. Hasil surrogate model variasi model retak dengan algoritma RBF memberikan hasil prediksi dengan validasi R2 dengan variabel rasio (a/2c) sebesar 1 lalu dengan algoritma SVM memberikan hasil prediksi dengan validasi R2 dengan variabel rasio (a/2c) sebesar 0,99.
=================================================================================================================================
Based on SKK Migas data in 2016, 54.65% of offshore platforms in Indonesia were over 20 years old. Structures that have exceeded their operational life need to be reviewed in terms of their structural strength to determine if they are still capable of operating. In this study, the author will focus on predicting the stress intensity factor on a four-legged jacket structure based on a surrogate model. The stress intensity factor is a factor that determines fatigue in tubular connections using fracture mechanics methods. To improve accuracy and optimize analysis time, a surrogate model was developed from the analysis of crack variations obtained using the finite element method. The analysis methods used in this study include static analysis in-place for global structural stress analysis, local critical tubular connection analysis, crack analysis in tubular with the highest critical stress point, and modeling of stress intensity factors based on a surrogate model using a machine learning SVM model based on variations in crack depth and crack length. Crack analysis used 30 crack model variations located at maximum stress in the local analysis of the finite element method. Local structural analysis is based on the results of global in-place static analysis of the structure with its critical connection, which is the DKT multiplanar tubular connection. The highest von-Mises stress was obtained at brace 5 with a stress of 327 MPa. The surrogate model results of crack model variations with the RBF algorithm provided a prediction result with a validation R2 value of 1 for the ratio variable (a/2c), while the SVM algorithm provided a prediction result with a validation R2 value of 0.99 for the ratio variable (a/2c).

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Manifold Wellhead Platform, Surrogate Model, Analisis Lokal, Machine Learning, Faktor Intensitas Tegangan, Local Analysis, Machine Learning, Stress Intensity Factor
Subjects:	T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TC Hydraulic engineering. Ocean engineering T Technology > TC Hydraulic engineering. Ocean engineering > TC1665 Offshore structures--Materials. T Technology > TC Hydraulic engineering. Ocean engineering > TC1680 Offshore structures
Divisions:	Faculty of Marine Technology (MARTECH) > Ocean Engineering > 38201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Akbar Hardian
Date Deposited:	13 Jul 2023 07:43
Last Modified:	13 Jul 2023 07:43
URI:	http://repository.its.ac.id/id/eprint/98455

Actions (login required)

View Item