Perancangan Kendali Manouvre Kapal Trimaran Berbasis Neural Network

Samloy, Queen Fitri Maulidina (2021) Perancangan Kendali Manouvre Kapal Trimaran Berbasis Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 04211740000018_Undergraduate Thesis.pdf]
Preview
Text
04211740000018_Undergraduate Thesis.pdf - Accepted Version

Download (2MB) | Preview

Abstract

Kapal trimaran merupakan pengembangan desain lambung kapal. Desain 3 buah lambung ini menguntungkan dari sisi berkurangnya nilai resistance, meningkatkan straight line stability dan lain-lain. Akan tetapi kemampuan maneuver trimaran lebih rendah jika dibandingkan dengan kapal monohull. Pada penelitian ini, kapal trimaran didesain menggunakan perbedaan putaran motor di sisi outtriggers hull-nya. Perbedaan putaran antara kedua motor ini mengakibatkan salah satu bagian buritan sisi outrigger memiliki trust yang lebih besar dan menimbulkan gerak yawing pada body kapal trimaran. Untuk mengetes kemampuan maneuver trimaran digunakan standard IMO ITTC 2002 mengenai “turning ability”. Jika belum memenuhi standard maneuverability, maka sistem kembali menggunakan rudder sebagai komponen maneuver-nya. Data performa maneuver nantinya digunakan sebagai data input dan ouput target yang dimasukkan ke program neural network. Neural network ini berfungsi untuk membuat jaringan dan menentukan korelasi antara data input dan output. Neural network memiliki sifat adaptif learning, yakni berulang kali mempelajari data yang diberikan hingga menemukan target nilai error seminim mungkin.
=================================================================================================
The trimaran is one of the inventions of the hull design. The multihull
design has benefits to reduce resistance, increases straight line stability and
others. However, the trimaran has lower maneuverability compared to the
monohull ship. In this paper, the trimaran maneuver system is designed using the
different RPM between two Brushless DC motor at outtrigger hulls. The
difference causes one of the stern parts of the outriggers having higher trust than
other and causes yawing motion on the trimaran body. the IMO ITTC 2002
standard was used regarding the "turning ability" to test the maneuverability of
the trimaran. Maneuver performance data such as, drift angle, rudder angle, etc
will be used as an input and target output data. Then, those data are used for
neural network program to create a network and determine the correlation
between input and output data. Neural network has the merits in adaptive
learning, it can repeatedly learn the data provided to find the minimum error
value targeted during creating the network.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: trimaran ship, maneuver and neural network.
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering
Divisions: Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User: Queen Fitri Maulidina Samloy
Date Deposited: 08 Mar 2021 04:06
Last Modified: 25 Jul 2024 03:27
URI: http://repository.its.ac.id/id/eprint/83772

Actions (login required)

View Item View Item