Analisa Komparatif Pada Prediksi Konsumsi Bahan Bakar Dengan Metode Machine Learning

Antoridi, Hizbi Muhamadsyah (2023) Analisa Komparatif Pada Prediksi Konsumsi Bahan Bakar Dengan Metode Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 04211941000023-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (6MB) | Request a copy

Abstract

There were issues with fuel on the older ships, such as a lack of fuel when sailing due to the manual nature of the estimation calculations. One of the causes of problems is also the lack of computers. Consequently, this study was conducted to develop a solution to this problem; fuel consumption estimation will be performed using machine learning. Machine learning is a subset of Artificial Intelligence that uses regression to make predictions. Using twelve variables, including ship speed (knot), engine RPM, engine torque (kNm), engine power (kW), charge air pressure (bar), charge air temperature (°C), calorific heat value (kj/kg), SFOC (g/kWh), period of seawave (s), ambient air pressure (Pa), temperature (°C), humidity (%) and the fuel oil consumption (l/h), it should be possible to accurately estimate fuel consumption. This study's research yielded classification and estimation that are rated as excellent. Polynomial Regression is the most accurate regression method, with a 99.99% accuracy rate and RMSE value of 0.13867 and MAE value of 0.108, when compared to other regression methods.
=================================================================================================================================
Ada beberapa masalah terkait bahan bakar pada kapal yang lebih tua, seperti kekurangan bahan bakar saat berlayar karena perhitungan estimasi yang masih manual. Salah satu penyebab masalah juga kurangnya komputer. Oleh karena itu, penelitian ini dilakukan untuk mengembangkan solusi atas masalah tersebut; Estimasi konsumsi bahan bakar akan dilakukan dengan menggunakan machine learning. Pembelajaran mesin adalah bagian dari Kecerdasan Buatan yang menggunakan regresi untuk membuat prediksi. Dengan menggunakan dua belas variabel, kecepatan kapal (knot), RPM mesin, torsi mesin (kNm), daya mesin (kW), tekanan udara muatan (bar), suhu udara muatan (°C), calorific heat value (kj/kg), SFO (g/kWh), periode gelombang laut (s), tekanan udara ambien (Pa), suhu (°C), kelembaban (%), seharusnya dapat memperkirakan konsumsi bahan bakar secara akurat. Penelitian studi ini menghasilkan klasifikasi dan estimasi yang dinilai sangat baik. Regresi Polinomial merupakan metode regresi yang paling akurat, dengan tingkat akurasi 99,99% dan nilai RMSE sebesar 0,13867 serta nilai MAE sebesar 0,108, jika dibandingkan dengan metode regresi lainnya.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Comparative Analysis, Fuel Consumption Prediction, Machine Learning
Subjects:	T Technology > T Technology (General) > T57.74 Linear programming T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM731 Marine Engines
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Hizbi Muhammadsyah Antoridi
Date Deposited:	09 Nov 2023 02:27
Last Modified:	09 Nov 2023 02:27
URI:	http://repository.its.ac.id/id/eprint/102800

Actions (login required)

View Item