Pemodelan Estimasi Konsumsi Fuel Oil pada Mesin Diesel Kapal LCT Surya Agung Menggunakan Machine Learning dengan Metode Kernel Ridge Regression.

Zagota, Samuel (2024) Pemodelan Estimasi Konsumsi Fuel Oil pada Mesin Diesel Kapal LCT Surya Agung Menggunakan Machine Learning dengan Metode Kernel Ridge Regression. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5019201166-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (16MB) | Request a copy

Abstract

Penelitian ini mengembangkan model estimasi konsumsi bahan bakar pada mesin diesel kapal LCT Surya Agung menggunakan Kernel Ridge Regression. Data operasional kapal yang digunakan dalam penelitian ini berasal dari Buku Harian untuk mesin, yang meliputi RPM, suhu gas buang, suhu air pendingin, dan tekanan minyak lumas, serta Buku Harian untuk dek, yang meliputi kecepatan kapal, jarak, draft kapal, dan keadaan laut. Data tersebut dikumpulkan dan dibersihkan melalui proses data preprocessing untuk menghilangkan missing data dan outlier. Berdasarkan analisa menggunakan permutation feature importance pada phyton, fitur yang paling berpengaruh terhadap nilai prediksi adalah draft tengah kapal dan kecepatan kapal. Untuk mengoptimalkan kinerja model prediksi, dilakukan tuning hyperparameter melalui metode grid search. Hyperparameter yang dioptimalkan termasuk nilai alpha, jenis kernel, dan degree. Kombinasi hyperparameter terbaik yang ditemukan adalah alpha sebesar 1.0, degree sebesar 2, dan kernel 'poly'. Evaluasi model dilakukan dengan membandingkan hasil prediksi terhadap data aktual menggunakan metrik Root Mean Squared Error (RMSE) dan R-squared (R²). Hasil pemodelan menunjukkan bahwa model memiliki akurasi prediksi yang tinggi dengan nilai RMSE sebesar 45.11735740277411 dan R² sebesar 0.9286707446214904. Hasil penelitian model ini juga dapat diadaptasi dan diterapkan pada jenis kapal lainnya dengan penyesuaian yang sesuai, sehingga memiliki potensi aplikasi yang luas dalam industri maritim.
======================================================================================================
This study develops a fuel consumption estimation model for the diesel engine of the LCT Surya Agung vessel using Kernel Ridge Regression. The operational data of the ship used in this research were sourced from the Engine Logbook, which includes RPM, exhaust gas temperature, cooling water temperature, and lubricating oil pressure, as well as the Deck Logbook, which includes ship speed, distance, ship draft, and sea conditions. The collected data underwent preprocessing to eliminate missing data and outliers. Based on the analysis using permutation feature importance in Python, the most influential features on the prediction values were the ship's midship draft and speed. To optimize the model's predictive performance, hyperparameter tuning was conducted through grid search. The hyperparameters optimized included the alpha value, kernel type, and degree. The best combination of hyperparameters was found to be an alpha of 1.0, a degree of 2, and a 'poly' kernel. The model's performance was evaluated by comparing the predicted results with actual data using the Root Mean Squared Error (RMSE) and R-squared (R²) metrics. The modeling results showed that the model has a high prediction accuracy, with an RMSE of 45.11735740277411 and an R² of 0.9286707446214904. The model also has the potential for adaptation and application to other types of vessels with appropriate adjustments, thus offering broad applicability within the maritime industry.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Konsumsi Bahan Bakar, Mesin Diesel Kapal, Machine Learning, Kernel Ridge Regression, Hyperparameter Tuning. Fuel Consumption, Ship Diesel Engine, Machine Learning, Kernel Ridge Regression, Hyperparameter Tuning.
Subjects:	V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM276.A1 Fuel (Including supplies, costs, etc.)
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Samuel Zagota
Date Deposited:	06 Aug 2024 04:03
Last Modified:	06 Aug 2024 04:03
URI:	http://repository.its.ac.id/id/eprint/112679

Actions (login required)

View Item