Admiral, Fulgentius Agiel (2022) Pemodelan Persamaan Faktor Konsentrasi Tegangan Pada Sambungan Tubular Multi-Planar Double Kt Menggunakan Metode Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 04311740000039-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
04311740000039-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (5MB) |
Abstract
Jacket platform adalah struktur bangunan lepas pantai yang banyak digunakan untuk industri migas dan energi terbarukan karena efisiensi dan keandalannya. Bagian penting yang perlu dipertimbangkan dalam mendesain jacket tersebut adalah sambungan tubular, terutama sambungan multi-planar yang banyak ditemukan pada struktur jacket. Pada sambungan, konsentrasi tegangan rentan terjadi, sehingga dapat menyebabkan kelelahan akibat punching shear, local buckling, dan propagasi retak. Dalam mengkalkulasi umur kelelahan, banyak peneliti menggunakan analisis metode elemen hingga untuk memprakirakan faktor konsentrasi tegangan karena tingkat akurasi yang andal. Namun, membutuhkan banyak pemodelan dan memakan biaya komputasi. Sehingga, studi ini bertujuan untuk mengembangkan persamaan SCF untuk sambungan tubular double KT berdasarkan pemodelan numerik metode elemen hingga menggunakan metode machine learning untuk meningkatkan akurasi dan mengurangi waktu komputasi. Persamaan yang diformulasikan menggunakan analisis regresi sebagai salah satu bagian dari metode machine learning. SCF di sekitar las sambungan tubular DKT akibat pembebanan kombinasi dari beban aksial, in-plane bending, dan out-of-plane bending moments dipertimbangkan. Keandalan dari persamaan SCF diperiksa menggunakan acceptance criteria berdasarkan rekomendasi dari UK Department of Energy. Persamaan SCF kemudian dibandingkan dengan hasil dari persamaan SCF konvensional.
===================================================================================================================================
Most of the world’s infrastructure is a jacket stucture applied in the oil and gas also renewable energy industries because of its efficiency and reliability. An important part that needs to be considered in designing a jacket is the tubular joint, especially the multi-planar joint that can cause failures due to punching shear, local buckling, and especially crack propagation leading to fatigue. In predicting fatigue life, many researchers use the finite element (FE) analysis method for estimating the stress concentration factor (SCF) because of its practicability with considerable accuracy. However, in the practical design routine, FE analysis needs high computational time and effort. Therefore, this study will develop an alternative SCF equation for a Double KT (DKT) tubular joint based on numerical results using the finite element method and machine learning techniques to increase accuracy while reducing computational time. Equation formulation uses regression analysis method as one of the machine learning methods. The SCF distribution along welded lines of the DKT joint will also be presented with a loading combination of axial load, in-plane bending, and out-of-plane bending moments. The reliability of the SCF equation is checked using acceptance criteria based on the recommendation of the UK Department of Energy. The proposed SCF equation will be compared to the result of the existing conventional SCF equation.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSKe 627.98 Adm p-1 2022 |
| Uncontrolled Keywords: | Faktor konsentrasi tegangan, Machine learning, Metode elemen hingga, Sambungan tubular DKT. DKT tubular joint, Finite element method, Machine learning, Stress concentration factor. |
| Subjects: | T Technology > TC Hydraulic engineering. Ocean engineering > TC1665 Offshore structures--Materials. |
| Divisions: | Faculty of Marine Technology (MARTECH) > Ocean Engineering > 38201-(S1) Undergraduate Thesis |
| Depositing User: | Mr. Marsudiyana - |
| Date Deposited: | 22 May 2026 04:29 |
| Last Modified: | 22 May 2026 04:29 |
| URI: | http://repository.its.ac.id/id/eprint/133340 |
Actions (login required)
 |
View Item |