Optimasi Desain Berbasis Keandalan Multi Objektif dengan Surrogate Model pada Sistem Penopang DeepCWind Turbin Angin Lepas Pantai Terapung

Syarifudin, Muhammad Rizky (2025) Optimasi Desain Berbasis Keandalan Multi Objektif dengan Surrogate Model pada Sistem Penopang DeepCWind Turbin Angin Lepas Pantai Terapung. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6020222001-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (4MB) | Request a copy

Abstract

Turbin angin lepas pantai diprediksi mengalami peningkatan seiring dengan target pemanfaatan energi angin sebesar 225 MW pada tahun 2025 di Indonesia. Potensi pemanfaatan angin lepas pantai lebih maksimal pada kedalaman laut lebih dari 100 m. Namun, biaya pembangunan struktur lepas pantai merupakan fungsi linear dari kedalaman air dan jarak dan pantai, sehingga menyebabkan mahalnya biaya proyek yang mencapai 19% dari total biaya proyek pembangunan. Selain itu, keselamatan juga faktor krusial. Probabilitas terbesar kegagalan turbin angin lepas pantai terletak pada komponen struktur penopang, salah satunya kegagalan pada sistem tambat akibat beban siklik. Optimasi desain berbasis keandalan menjadi solusi sebagai konsep proses optimasi di bawah batasan keandalan tiap performansi untuk mendapatkan desain yang optimal dari segi ekonomis dan keselamatan. Surrogate model digunakan sebagai pengganti proses implisit menjadi proses eksplisit untuk mengurangi waktu komputasi selama proses optimasi. Metode Least Square Boosting (LSBoost) terbukti membentuk surrogate yang layak untuk variabel desain multi dimensi dengan mencegah oversmoothing pada data outlier pada penelitian ini. Penelitian ini merupakan optimasi multi objektif dengan meminimalkan biaya struktur dan cumulative damage. Biaya struktur dianggap sebagai fungsi berat struktur penopang. Pareto solution digunakan sebagai solusi pencarian bobot antar objektif dengan hasil trade off menunjukkan cumulative damage memiliki bobot pertimbangan yang lebih tinggi dibandingkan dengan biaya struktur. Optimasi desain berbasis keandalan multi objektif menghasilkan dimensi yang optimal dengan meminimalkan biaya sebesar 35% dan meminimalkan cumulative damage menjadi 0,124 dengan umur lelah struktur 161,3 tahun. Hasil memenuhi batas keandalan yang ditetapkan yakni pengaruh indeks keandalan gaya tarik tali tambat 3,16 di atas indeks keandalan normal 3,09. Penelitian ini menunjukkan bahwa integrasi antara teknik probabilistik dan metode optimasi dapat meningkatkan efisiensi desain, memastikan keamanan struktural, dan mengurangi potensi overdesign. Dengan demikian, pendekatan RBDO direkomendasikan sebagai metode yang layak untuk diterapkan dalam perancangan sistem teknik yang membutuhkan keseimbangan antara kinerja, biaya, dan keandalan.
===================================================================================================================================
Offshore wind turbines are predicted to increase in line with the target of wind energy of 225 MW by 2025 in Indonesia. The potential for offshore wind utilization is maximized at more than 100 m sea depths. However, the cost of building offshore structures is a linear function of water depth, distance, and beach, resulting in high project costs that reach 19% of the total cost of construction projects. In addition, reliability is also a crucial factor. The highest probability of failure of an offshore wind turbine lies in the components of the support structure, one of which is the failure of the mooring system due to cyclic loads. Reliability-based design optimization is a solution as an optimization process concept under the reliability constraints of each performance to obtain an optimal design in terms of economy and reliability. The surrogate model is used instead of an implicit process to an explicit process to reduce the computational time during the optimization process. The Least Square Boosting (LSBoost) method was proven to form a feasible surrogate for multi-dimensional design variables by preventing over-smoothing in the outlier data in this study. This study is a multi-objective optimization that minimizes structural costs and cumulative damage. The structure's cost is considered a weight function of the support structure. The Pareto solution was used to find the weight between objectives, with the trade-off results showing that cumulative damage had a higher weight of consideration than the structure's cost. Multi-objective reliability-based design optimization results in optimal dimensions by minimizing cost by 35% while the cumulative damage drops to 0.124 with a structural fatigue life of 161.3 years. The result satisfies the constraint, namely the influence of the reliability index of the mooring rope tensile force of 3.16 above the limit reliability index of 3.09. This study shows that integrating probabilistic techniques and optimization methods can improve design efficiency, ensure structural safety, and reduce the potential for overdesign. Thus, the RBDO approach is recommended as a feasible engineering design method that requires a balance between performance, cost, and reliability.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Keandalan, Kelelahan, Multi Objektif, Model Pengganti, Optimasi Desain, Optimasi Desain Berbasis Keandalan, Turbin Angin Lepas Pantai Terapung, Design Optimization, Fatigue, Floating Offshore Wind Turbine, Multi Objective, Reliability, Reliability Based Design Optimization, Surrogate Model
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA658 Structural design T Technology > TC Hydraulic engineering. Ocean engineering > TC1680 Offshore structures
Divisions:	Faculty of Marine Technology (MARTECH) > Ocean Engineering > 38101-(S2) Master Thesis
Depositing User:	Muhammad Rizky Syarifudin
Date Deposited:	03 Feb 2025 15:33
Last Modified:	03 Feb 2025 15:33
URI:	http://repository.its.ac.id/id/eprint/117909

Actions (login required)

View Item