Prediksi Performa Mesin Diesel Berbahan Bakar Biodiesel Menggunakan Pemodelan Artificial Neural Network

Wardhana, Angelo Decente (2025) Prediksi Performa Mesin Diesel Berbahan Bakar Biodiesel Menggunakan Pemodelan Artificial Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5019211068-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (11MB) | Request a copy

Abstract

Pemodelan performa mesin melibatkan hubungan kompleks dan non-linier antar variabel, sehingga metode machine learning seperti Artificial Neural Network (ANN) digunakan karena mampu meniru cara kerja otak dan belajar dari data. ANN terbukti efektif memprediksi performa mesin secara cepat dan akurat tanpa perhitungan iteratif, sehingga dapat menghemat waktu dan biaya dibandingkan metode eksperimental atau simulasi konvensional. Penelitian ini bertujuan untuk mengembangkan model prediksi performa mesin diesel berbasis Artificial Neural Network (ANN) pada program MATLAB terhadap empat parameter utama, yaitu Fuel Oil Consumption (FOC), Brake Horse Power (BHP)/Power, Specific Fuel Oil Consumption (SFOC), dan Torsi. Dua jenis mesin diesel digunakan dalam studi ini, yaitu mesin MAK 6M43C dan mesin Yanmar TF-85, dengan masing-masing arsitektur jaringan ANN yang disesuaikan untuk mencapai akurasi optimal. Hasil pelatihan dan pengujian menunjukkan performa yang sangat baik, dengan nilai koefisien determinasi (R²) mencapai 0.9997 untuk training dan 0.9997 untuk testing pada mesin MAK 6M43C, serta 0.9958 untuk training dan 0.9954 untuk testing pada mesin Yanmar TF-85. Temuan ini membuktikan bahwa model ANN mampu memprediksi performa mesin diesel dengan akurat dan stabil, termasuk saat menggunakan bahan bakar alternatif seperti biodiesel. Model ini berpotensi diterapkan sebagai sistem prediktif berbasis machine learning di bidang kelautan, khususnya untuk mendukung optimalisasi operasional mesin dan pengembangan transportasi laut yang ramah lingkungan.
========================================================================================================================================
Machine performance modeling involves complex and non-linear relationships between variables, so machine learning methods such as Artificial Neural Networks (ANN) are used because they can mimic the way the brain works and learn from data. ANNs have proven to be effective in predicting machine performance quickly and accurately without iterative calculations, saving time and costs compared to conventional experimental or simulation methods. This study aims to develop a diesel engine performance prediction model based on Artificial Neural Network (ANN) in MATLAB for four main parameters, namely Fuel Oil Consumption (FOC), Brake Horse Power (BHP)/Power, Specific Fuel Oil Consumption (SFOC), and Torque. Two types of diesel engines were used in this study, namely the MAK 6M43C engine and the Yanmar TF-85 engine, each with a customized ANN network architecture to achieve optimal accuracy. The training and testing results showed excellent performance, with a coefficient of determination (R²) value of 0.9997 for training and 0.9997 for testing on the MAK 6M43C engine, and 0.9958 for training and 0.9954 for testing on the Yanmar TF-85 engine. These findings prove that the ANN model can accurately and stably predict diesel engine performance, including when using alternative fuels such as biodiesel. This model has the potential to be applied as an machine learning based predictive system in the maritime field, particularly to support engine operational optimization and the development of environmentally friendly maritime transportation.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Artificial Neural Network, Biodiesel, MATLAB, Prediksi, Prediction
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control T Technology > TJ Mechanical engineering and machinery > TJ785 Internal combustion engines. Spark ignition T Technology > TJ Mechanical engineering and machinery > TJ797 Diesel motor--Fuel systems--Testing.
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Angelo Decente Wardhana
Date Deposited:	05 Aug 2025 04:01
Last Modified:	05 Aug 2025 04:01
URI:	http://repository.its.ac.id/id/eprint/127419

Actions (login required)

View Item