Prastya, Donny Endra (2020) Evaluasi Performa Main Engine Dengan Pendekatan Forecasting Assessment Berdasarkan Data Condition Monitoring Pada Temperatur Gas Buang. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
04211750010008-Master_Thesis.pdf Download (5MB) | Preview |
Abstract
Perawatan dan pemeliharaan adalah merupakan unsur utama dalam setiap kegiatan operasional kapal. Perawatan mesin kapal yang tidak memadai dapat meningkatkan kegagalan peralatan yang mengancam lingkungan, mempengaruhi kinerja, memiliki dampak besar dalam hal kerugian bisnis dengan mengurangi ketersediaan kapal, meningkatkan waktu henti, dan juga meningkatkan potensi kecelakaan besar yang terjadi dan membahayakan nyawa di atas kapal. Mesin utama di kapal tidak hanya merupakan bagian terpenting namun juga komponen yang paling dominan mengalami kendala. Dalam kondisi aktualnya, kerusakan yang terjadi pada sebuah mesin sangatlah kompleks dan akan berakibat fatal apabila tidak dilakukan kegiatan monitoring dan perawatan yang baik pada saat kapal dalam kondisi operasi. Sebelum dilakukan perbaikan pada kerusakan yang dialami mesin diesel kapal, perlu dilakukan deteksi terhadap kerusakan yang ada pada mesin. setelah diketahui jenis kerusakan dan penyebabnya, hal yang perlu dilakukan adalah mengetahui penyebab utama. Dalam penelitian ini penulis mengembangkan sebuah metode kombinasi untuk mendeteksi atau mendiagnosis sebuah mesin berdasarkan data condition monitoring pada temperature gas buang. Metode yang diusulkan merupakan kombinasi atau gabungan dua metode, yaitu Failure Mode and Effect Criticality Analysis (FMECA) dan metode Forecasting Assessment (peramalan) menggunakan Artificial Neural Network (ANN). FMECA dimaksudkan untuk bottom-up kejadian penyebab kegagalan dan mengukur level kekritisan pada engine. Forecasting Assessment (peramalan) digunakan untuk memprediksi nilai yang akan datang dari semua silinder mesin utama suhu gas buang. Dengan kombinasi metode diatas, maka diharapkan dapat memprediksi waktu dan tindakan dalam proses perawatan yang dilakukan pada mesin kapal. Adapun hasil dari penelitian ini terdapat 23 equipment dan terdapat 99 failure mode pada lubricating oil system. Dari keseluruhan failure mode sebaran level kekritisan memiliki level low risk (24%), medium risk (64%), dan high risk (12%). Berdasarkan data condition monitoring temperatur gas buang selajutnya dilakukan forecasting assessment (peramalan) didapatkan nilai error maksimal sebesar 4,31% divalidasi melalui perbandingan pengamatan aktual yang dicatat langsung diatas kapal untuk peramalan 12 bulan yang akan datang.
==================================================================================================================================
Operation and maintenance are the main elements in every operational activity of the ship. Inadequacy in terms of ship engine maintenance could increase the rates of failures in the equipment that could lead to environmental problems, altering the performance, and would possibly create a large impact of business loss over the decline of availability on the vessels, increasing the downtime, and also potentially broaden the potential for major accidents that occur and endangering lives on board. The main engine on the ship is not only the most important part but also the most dominant component that often undergo problems. In actual conditions, the damage that occurs to an engine is very complex and would lead to fatality if there is a lack of proper monitoring and maintenance activities to be carried out when the ship is in operational condition. Before repairs are to be made onto the damage suffered by the ship's diesel engine, it is necessary to detect any damage to the engine in prior. Subsequentially to the type of damage and its cause, the thing to do is to identify the major cause. In this study, the authors developed a combination method to detect or diagnose an engine based on its condition monitoring data on the exhaust gas temperature. The proposed method is a combination or a fuse of two methods, on Failure Mode and Effect Criticality Analysis (FMECA) and the Forecasting Assessment method using Artificial Neural Network (ANN). FMECA is intended to bottom-up the events that lead to failure and measure the level of criticality on the engine. Forecasting Assessment is used to predict future values from all major engine cylinders of the exhaust gas temperature. With the combination of the methods above, it is expected that the prediction of time and action in the maintenance process are adept to be carried out on the ship's engine. The results of this study were consisting of 23 equipment and 99 failure modes detected in the lubricating oil system. From the overall failure mode, the critical level distribution had a low-risk level (24%), medium risk (64%), and high risk (12%). Based on the condition monitoring data on the exhaust gas temperature, the forecasting assessment was obtained and the maximum error value of 4.31% is validated through comparison of actual observations recorded directly on the ship for forecasting towards 12 months forth.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSP 623.872 36 Pra e-1 |
| Uncontrolled Keywords: | Condition Monitoring, FMECA, Forecasting Assessment, Main Engine, Temperatur Gas Buang |
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TJ Mechanical engineering and machinery > TJ785 Internal combustion engines. Spark ignition V Naval Science > VC Naval Maintenance V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM731 Marine Engines |
| Divisions: | Faculty of Marine Technology (MARTECH) > Marine Engineering > 36101-(S2) Master Theses |
| Depositing User: | DONNY ENDRA PRASTYA |
| Date Deposited: | 20 Mar 2025 06:40 |
| Last Modified: | 20 Mar 2025 06:40 |
| URI: | http://repository.its.ac.id/id/eprint/74735 |
Actions (login required)
 |
View Item |