Optimasi Metaheuristik Menggunakan Metode Back Propagation Neural Network, Genetic Algorithm Dan Particle Swarm Optimization Proses Hot Forging Pada Produksi Roda Kereta Cepat

Oetomo, Ghassani Darystyo (2021) Optimasi Metaheuristik Menggunakan Metode Back Propagation Neural Network, Genetic Algorithm Dan Particle Swarm Optimization Proses Hot Forging Pada Produksi Roda Kereta Cepat. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
02111640000204-Undergraduate-Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img] Text
02111640000204-Undergraduate-Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Transportasi merupakan kebutuhan dasar masyarakat dalam kehidupan sehari-hari. Perkembangan pembangunan sistem transportasi di Indonesia saat ini sedang digencarkan guna memenuhi kebutuhan masyarakat terutama sektor transportasi darat. Transportasi darat yang disediakan pemerintah dalam memenuhi kebutuhan transportasi masyarakat yang layak, terjangkau,aman dan efisien salah satunya adalah kereta cepat. Sistem roda dari kereta cepat berbeda dengan kereta kecepatan rendah dan kecepatan sedang dan merupakan komponen penting dalam pengoperasian kereta cepat, perbedaan kebutuhan memerlukan optimalisasi dari proses manufaktur dalam pembentukan roda kereta untuk menghindari terjadinya kegagalan dan over-design roda kereta yang dapat membahayakan pengguna dan membebani dari segi ekonomi. Penelitian ini berfokus pada proses hot forging yang digunakan dalam proses pembentukan roda kereta yang dipengaruhi temperatur proses forging, dimensi bilet dan dimensi dies yang digunakan dalam proses hot forging. Penelitian dilakukan dengan metode optimalisasi Backpropagation Neural Network (BPNN) sebagai alat untuk merespon nilai minimal forging force dengan parameter input variasi dimensi billet dan temperatur. Prediksi tersebut kemudian digunakan untuk mendapatkan parameter input dengan hasil respon yang optimal dengan menggunakan metode metaheuristik yaitu Genetic Algorithm (GA) dan Particle Swarm Optimization (PSO) yang akan digunakan untuk mencari nilai global optimum minimal forging force dan akurasi dimensi. Hasil dari penelitian ini didapatkan nilai optimum dari dimensi diameter billet dan tebal billet untuk minimum forging force. Dengan ketebalan billet 73,5 mm dan diameter billet 880 mm. Untuk nilai optimal berdasarkan simulasi ANSYS nilai optimum dari temperatur berada pada temperatur 1000°C dimana minimal forging force berbanding lurus dengan peningkatan temperatur. Hasil dari optimasi yang dilakukan menggunakan metode BPNN-GA untuk mencari nilai global optimum dari parameter forging didapatkan parameter optimum pada temperatur 950°C ketebalan billet 73,5 mm dan diameter billet 880 mm dan untuk BPNN-PSO didapatkan parameter optimum pada temperatur 950°C dan terdapat perbedaan prediksi nilai global minimum pada akurasi dimensi web dimana temperatur optimal sebsar 1000°C hal ini terjadi karena PSO tidak mempunyai kemampuan keluar dari nilai lokal optimum sehingga terdapat perbedaan pada prediksi temperatur optimal. Untuk ketebalan billet 73,5 mm dan diameter billet 880 mm. ================================================================================================ Transportation is a basic need of society in everyday life. The development of the transportation system in Indonesia is currently being intensified to meet the needs of the community, especially the land transportation sector. Land transportation provided by the government in meeting the needs of public transportation that is feasible, affordable, safe and efficient, one of which is the fast train. The wheel system of the high-speed train is different from the low-speed and medium-speed trains and is an important component in the operation of the high-speed train, the different requirements require optimization of the manufacturing process in the formation of the train wheels to avoid failure and over-design of the train wheels which can endanger the user and burden the carriage. economics. This study focuses on the hot forging process used in the train wheel formation process which is influenced by the forging process temperature, billet dimensions and dies dimensions used in the hot forging process. The research was conducted using the Backpropagation Neural Network (BPNN) optimization method as a tool to respond to the minimum value of forging force with the input parameters of billet dimensions and temperature variations. These predictions are then used to obtain input parameters with optimal response results using metaheuristic methods, namely Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) which will be used to find the global optimum value for minimum forging force and dimensional accuracy. The results of this study obtained the optimum value of the dimensions of the billet diameter and billet thickness for minimum forging force. With a billet thickness of 73.5 mm and a billet diameter of 880 mm. For the optimal value based on the ANSYS simulation, the optimum value of temperature is at a temperature of 1000°C where the minimum forging force is directly proportional to the increase in temperature. The results of the optimization carried out using the BPNN-GA method to find the global optimum value of the forging parameter obtained the optimum parameter at a temperature of 950°C, billet thickness of 73.5 mm and billet diameter of 880 mm and for BPNN-PSO the optimum parameter was obtained at a temperature of 950°C. and there is a difference in the prediction of the minimum global value on the accuracy of the web dimensions where the optimal temperature is 1000°C this happens because PSO does not have the ability to get out of the optimum local value so that there is a difference in the optimal temperature prediction. For billet thickness 73.5 mm and billet diameter 880 mm.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Roda Kereta, Hot Forging, Back Propagation Neural Network, Genetic Algorithm, Particle Swarm Optimization, Wheel Train, Hot Forging
Subjects: T Technology > T Technology (General) > T57.62 Simulation
T Technology > TF Railroad engineering and operation
T Technology > TF Railroad engineering and operation > TF1327.O58 High speed ground transportation (train)
T Technology > TS Manufactures
T Technology > TS Manufactures > TS171 Product design
T Technology > TS Manufactures > TS176 Manufacturing engineering. Process engineering (Including manufacturing planning, production planning)
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User: Ghassani Darystyo Oetomo
Date Deposited: 01 Mar 2022 02:09
Last Modified: 01 Mar 2022 02:09
URI: https://repository.its.ac.id/id/eprint/94768

Actions (login required)

View Item View Item