Prediksi Gerak Berpasangan Rolling, Swaying, Dan Yawing Pada Floating Production Unit (EPU) Dalam Kondisi Beam Seas Dengan Model Arima, Var, Dan Hybrid Var-Ann

Suhermi, Novri (2014) Prediksi Gerak Berpasangan Rolling, Swaying, Dan Yawing Pada Floating Production Unit (EPU) Dalam Kondisi Beam Seas Dengan Model Arima, Var, Dan Hybrid Var-Ann. Other thesis, Insititut Teknologi Sepuluh Nopember.

Text 1310100010_Undergraduate_Thesis.pdf Download (3MB)

Abstract

Floating Production Unit (FPU) mempunyai peran penting dalam proses produksi minyak dan gas mentah yang berada di daerah lepas pantai. Oleh karena itu, stabilitas FPU sangat perlu diperhatikan. Stabilitas suatu kapal sangat ditentukan dari perilaku gerak rolling. Gerak kapal dapat dipelajari baik secara individu (uncoupled) maupun secara berpasangan (coupled). Salah satu sistem gerak coupling yang sering diteliti adalah gerak berpasangan rolling-swaying-yawing. Pada penelitian ini, dilakukan peramalan (forecasting) gerak rolling, swaying, dan yawing dalam kondisi beam seas pada FPU dengan menggunakan model ARIMA, VAR, dan Hybrid VAR- ANN. Data yang digunakan adalah data hasil simulasi eksperimen gerak FPU. Data dibagi menjadi 3000 data in-sample dan 1000 data out-of-sample. Kriteria model terbaik didasarkan pada kriteria out-of-sample dengan menggunakan ukuran Root Mean Square Error (RMSE). Dari ketiga model, didapatkan model terbaik yaitu model Hybrid VAR-ANN dengan nilai RMSE terkecil.
==============================================================================================================================
Floating Production Unit (FPU) has an important role in the production of crude oil and gas located in offshore area. Therefore, the stability of the FPU is very noteworthy. The stability of a ship is determined from the behavior of the rolling motion. Ship motion can be studied either in uncoupled system or in coupled system. One of the coupling motion system that is often observed is rolling-swaying-yawing motions. In this study, the rolling, swaying and yawing motions on a FPU are forecasted in beam seas condition by using ARIMA, VAR, and Hybrid VAR- ANN models. The data used are simulation data taken from the experiment. The data are divided into two parts, in-sample and out-of-sample. In-sample data consist of 3000 observations and out-of-sample data consist of 1000 observations. The best model selection is based on out-of-sample criteria. Out-of-sample criteria used is Root Mean Square Error (RMSE). From the three models, Hybrid VAR-ANN model became the best model with the smallest RMSE.

Item Type:	Thesis (Other)
Additional Information:	RSSt 519.535 Suh p
Uncontrolled Keywords:	ANN, ARIMA, forecasting, FPU, gerak kapal, Hybrid VAR-ANN, RMSE, time series, VAR.
Subjects:	T Technology > T Technology (General) > T174 Technological forecasting
Divisions:	Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	02 Jan 2024 07:03
Last Modified:	02 Jan 2024 07:03
URI:	http://repository.its.ac.id/id/eprint/105340

Actions (login required)

View Item