Optimasi Pemilihan Jenis Material dan Ketebalan Loop Spring pada Loopwheel Menggunakan Metode Elemen Hingga dan BPNN-GA

Natasha, Yoana Devita (2023) Optimasi Pemilihan Jenis Material dan Ketebalan Loop Spring pada Loopwheel Menggunakan Metode Elemen Hingga dan BPNN-GA. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 02111940000073-Undergraduate_Thesis.pdf] Text
02111940000073-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (11MB) | Request a copy

Abstract

Suspensi pada sepeda berperan penting bagi sebagian orang yaitu sebagai peredam getaran yang terjadi ketika roda sepeda berbenturan dengan permukaan yang tidak rata. Perkembangan teknologi di masa kini telah mendorong manusia untuk menciptakan inovasi pada roda seperti Loopwheel. Pada temuan ini, roda sepeda tidak lagi menggunakan jari-jari, namun digantikan dengan pegas berbentuk bulatan yang lentur di dalam roda sehingga menekan getaran ketika berkendara. Loopwheel merupakan sebuah inovasi pelek roda bersuspensi yang memberikan gaya pegas bagi pengendara yang terintegrasi dalam roda itu sendiri sehingga tidak memerlukan shock atau fork dengan suspensi. Loopwheel menggunakan tiga pegas komposit karbon yang dilingkarkan di antara velg dengan hub. Dalam penelitian ini, dilakukan simulasi statis dengan Metode Elemen Hingga (FEA) pada struktur roda Loopwheel untuk mendapatkan nilai tegangan dan defleksi menggunakan software ANSYS Workbench 19.2. Tahapan penelitian dimulai dari menentukan data awal seperti dimensi pada Loopwheel dan pembebanan yang terjadi melalui beberapa referensi. Kemudian, model Loopwheel serta jalan didesain menggunakan software Solidworks 2020. Model 3D yang telah didesain akan disimulasikan dengan FEA untuk menentukan nilai tegangan dan defleksi pada 21 variasi desain. Pada simulasi tersebut, digunakan beberapa variasi yakni 3 variasi jenis material serta 7 variasi ketebalan loop spring. Dari hasil yang didapatkan, dilakukan analisa data serta optimasi desain Loopwheel menggunakan metode Backpropagation Neural Network (BPNN) yang dilanjutkan metode Genetic Algorithm (GA). Desain optimum dari loop spring pada roda Loopwheel akan disimulasikan kembali dengan FEA sebagai konfirmasi hasil. Hasil dari penelitian ini diperoleh 21 data yang akan di-training dengan BPNN. Pada BPNN output defleksi, net terbaik diperoleh dengan 6 hidden layer, 3 neuron tiap hidden layer, serta activation function tiap hidden layer adalah tansig. Pada BPNN output tegangan, net terbaik diperoleh dengan 6 hidden layer, 3 neuron tiap hidden layer, serta activation function tiap hidden layer adalah logsig. Dengan menggunakan GA, diperoleh parameter terbaik yaitu dengan jenis material Epoxy E-Glass dan ukuran ketebalan loop spring sebesar 3,7299 mm. Kemudian parameter tersebut disimulasikan kembali pada software ANSYS sebagai konfirmasi hasil. Desain loop spring terbaik tersebut diprediksi oleh BPNN-GA menghasilkan defleksi sebesar 17,85386 mm dan tegangan sebesar 137,0779 MPa. Sedangkan, untuk simulasi menggunakan ANSYS menghasilkan defleksi sebesar 18,269 mm dan tegangan sebesar 135,17 MPa berdasarkan parameter optimasi sebelumnya.
===============================================================================================================================
Suspension on bicycles plays a significant role for many people, serving as a damper and vibration absorber when the bicycle wheel encounters uneven surfaces, especially in off-road areas designed for extreme biking. Advances in science and technology have driven humans to design and create innovations in wheels, such as the Loopwheel. In this discovery, bicycle wheels no longer use spokes but are replaced by flexible circular springs inside the wheel's circumference, which reduces vibrations while riding. Loopwheel is an innovative suspension wheel that provides a spring force to the rider, integrated within the wheel itself, eliminating the need for a separate shock or fork suspension. Loopwheel utilizes three carbon composite springs wound between the Rim and Hub. When the wheel hits bumps, the vibrations generated are dampened by these three springs. In this final project research, static simulations using the finite element method were conducted on the structure of the Loopwheel to obtain stress and deflection values using ANSYS Workbench 19.2 software. The research process started with determining initial data, such as the dimensions of the Loopwheel and the applied loads, based on various references. Subsequently, the Loopwheel model and road surface were designed using Solidworks 2020 software. The designed 3D model was then simulated using the finite element method to determine stress and deflection values for 21 design variations. The simulations involved three variations of materials and seven variations of loop spring thickness. From the obtained results, data analysis and Loopwheel design optimization were performed using the Backpropagation Neural Network (BPNN) method, followed by the Genetic Algorithm (GA) method. The optimized loop spring design was then re-simulated using FEA for validation and confirmation of the results. The research resulted in 21 datasets used for training the BPNN. For the deflection output in BPNN, the best network was obtained with 6 hidden layers, 3 neurons per hidden layer, and the activation function for each hidden layer was set as tansig. For the stress output in BPNN, the best network was obtained with 6 hidden layers, 3 neurons per hidden layer, and the activation function for each hidden layer was set as logsig. Using the Genetic Algorithm, the best parameters were determined as Epoxy E-Glass material and a loop spring thickness of 3.7299 mm. These parameters were then used for further simulation in ANSYS software to confirm the results. The optimized loop spring design predicted by BPNN-PSO resulted in a deflection of 17,85386 mm and stress of 137,0779 MPa. Meanwhile, the simulation using ANSYS yielded a deflection of 18,269 mm and stress of 135,17 MPa, based on the previously optimized parameters.

Item Type: Thesis (Other)
Uncontrolled Keywords: Loopwheel, suspension, deflection, stress, Loopwheel, suspensi, defleksi, tegangan.
Subjects: Q Science > Q Science (General) > Q325.78 Back propagation
Q Science > QA Mathematics > QA402.5 Genetic algorithms. Interior-point methods.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > TJ Mechanical engineering and machinery > TJ230 Machine design
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL240.5 Composite materials
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL257 Springs and suspension
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL410 Bicycles and bicycling--Design and construction
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL671.6. Materials--Fatigue.
T Technology > TS Manufactures > TS320 Steel--Metallurgy.
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User: Yoana Devita Natasha
Date Deposited: 14 Aug 2023 04:21
Last Modified: 14 Aug 2023 04:21
URI: http://repository.its.ac.id/id/eprint/102770

Actions (login required)

View Item View Item