OPTIMISASI DAYA, KONSUMSI BAHAN BAKAR DAN BIAYA OPERASIONAL MESIN PADA MOBIL SAPUANGIN 4 DENGAN JARINGAN SYARAF TIRUAN DAN ALGORTIMA KILLER WHALE

Auliya, Muhammad Fajar (2017) OPTIMISASI DAYA, KONSUMSI BAHAN BAKAR DAN BIAYA OPERASIONAL MESIN PADA MOBIL SAPUANGIN 4 DENGAN JARINGAN SYARAF TIRUAN DAN ALGORTIMA KILLER WHALE. Undergraduate thesis, Intitut Teknologi Sepuluh Nopember.

Text 2413100112-Undergraduate_Theses.pdf Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Pada lomba Formula Society of Automotive Engineering (SAE) desain motor bensin yang efisien dan bertenaga adalah sebuah keharusan. Penelitian ini membahas tentang pemodelan dan optimisasi mesin bensin dengan algoritma Killer Whale dengan algoritma genetik dan duelist algorithm sebagai pembanding. Model untuk mesin bensin dibuat dengan Jaringan Syaraf Tiruan (JST). Untuk memenuhi data yang dibutuhkan JST, maka mesin bensin disimulasikan menggunakan Ricardo Wave dengan parameter yang didapat dari pembacaan sensor. Algoritma yang digunakan pada JST adalah feed forward dengan metode pelatihan Levenberg Marquardt. Fungsi obyektif pada penelitian ini adalah untuk memaksimalkan daya, meminimalkan Brake Spessific Fuel Consumption (BSFC) dan meminimalkan biaya pemakaian. Variabel yang dioptimasi adalah putaran mesin (RPM), Air Fuel Ratio (AFR), Mass Fuel Flow (MF), Intake Pressure (IP), Intake Air Temperature (IAT) , Combustion Start (CS) dan Throttle Angle (TA). Hasil pemodelan JST memiliki Root Mean Square Error (RMSE) 0,021 kW untuk daya dan 0,00032 kg/kW.hr untuk konsumsi bahan bakar. Hasil optimasi menunjukkan bahwa adanya peningkatan daya sebesar 13%, penurunan konsumsi bahan bakar 11% dan penurunan biaya 23%. Daya maksimum diperoleh dengan nilai desain variabel RPM 4201, AFR 15,76, MFF 5,09 kg/hr, PI 0,98 bar, TI 294,89 K, CS -24,24o, CA_50 12,51o, b_dur 33,01o, TA 61,11o. BSFC minimun diperoleh dengan nilai desain variabel RPM 4442, AFR 13,37, MFF 4,88 kg/hr , PI 0,99 bar, TI 290,32 K, CS -21,3o, CA_50 12,259o, b_dur 32,94o, TA 69,59 o. Operational Cost minimum diperoleh dengan nilai desain variabel RPM 4201, AFR 15.76, MFF 5,87 kg/hr , PI 0,98 bar, TI 294,89 K, CS -24,24o, CA_50 12,51o, b_dur 33,94o, TA 69,59o.
========================================================================= In Formula Society of Automotive Engineering (SAE) competition, the design of efficient and powerful combustion engine is a must. This research discussed about optimization of gasoline engine using Killer Whale algorithm with genetic algorithm and duelits algorithm as comparation. The modelling for gasoline engine was built using Artificial Neural Network (ANN). To acquire data for training and testing the proposed ANN , a gasoline engine was simulated using Ricardo Wave. Parameters used were obtained from sensor logging. Algorithm that is used for ANN was feed forward and Levenberg Marquadt as training method. Objective function used is to maximize the power, minimize the Brake Spessific Fuel Consumption(BSFC) and minimize the operational cost. Optimized variable are engine speed (rpm), Air Fuel Ratio (AFR), Mass Fuel Flow (MFF), Intake Pressure (IP), Intake Air Temperature (IAT), Combustion Start (CS) and Throttle Angle (TA). Root Mean Square Error (RMSE) of ANN modelling is 0,021 kW for power and 0,00032 kg/kWhr for BSFC. The optimization result show that the power increases 13%, BSFC reduces to 11% and the cost operation decreases 23% compare with existing design. The optimum design variables for power are RPM 4201, AFR 15,76, MFF 5,09 kg/hr, PI 0,98 bar, TI 294,89 K, CS -24,24o, CA_50 12,51o, b_dur 33,01o, TA 61,11o. The optimum design variable for BSFC are RPM 4442, AFR 13,37, MFF 4,88 kg/hr , PI 0,99 bar, TI 290,32 K, CS -21,3o, CA_50 12,259o, b_dur 32,94o, TA 69,59 o. The optimum design variable for cost operation are RPM 4201, AFR 15.76, MFF 5,87 kg/hr , PI 0,98 bar, TI 294,89 K, CS -24,24o, CA_50 12,51o, b_dur 33,94o, TA 69,59o.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Kata Kunci: JST, algoritma optimisasi, daya, BSFC, biaya operasi
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions:	Faculty of Industrial Technology > Physics Engineering
Depositing User:	Auliya Muhammad Fajar
Date Deposited:	15 Oct 2025 11:40
Last Modified:	15 Oct 2025 11:45
URI:	http://repository.its.ac.id/id/eprint/48479

Actions (login required)

View Item