Simulasi Dinamis Dan Optimasi Dimensi Ban Airless Menggunakan ANSYS Explicit Dynamics, Backpropagation Neural Network-Genetic Algorithm

Saputro, Yosef Ardiansyah (2023) Simulasi Dinamis Dan Optimasi Dimensi Ban Airless Menggunakan ANSYS Explicit Dynamics, Backpropagation Neural Network-Genetic Algorithm. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02111940000104-Undergrauate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (5MB) | Request a copy

Abstract

Penggunaan ban pneumatic yang lazim saat ini sangat berpengaruh pada kondisi jalan yang dilalui karena sangat rawan rusak saat terkena benda tajam dan over-pressure yang berpotensi menyebabkan berkurangnya kenyamanan bahkan bisa mengakibatkan kecelakaan, sehingga berpotensi mengakibatkan kecelakaan. Saat ini hadir ban dengan teknologi tanpa udara atau yang dikenal sebagai ban airless, sebagai pendatang baru tentu ban airless menawarkan kelebuhan dibandingkan ban pneumatik di mana ban airless menawarkan pengendalian serta kenyamanan berkendara yang lebih baik, adanya rongga-rongga akan membuat getaran yang dialami ban dapat diserap. Tetapi masih diperlukan penyempurnaan terutama dalam segi kekuatan material dibandingkan ban pneumatic. Pada penelitian ini akan dilakukan analisis untuk mengetahui besaran defleksi dan maximum stress yang terjadi pada ban airless dibandingkan dengan ban pneumatic saat dilakukan uji impak secara dinamis menggunakan software ANSYS Workbench. Pada awal penelitian mula-mula dilakukan penentuan batasan masalah pada ban airless, setelah itu akan dilakukan pemodelan tiga dimensi terhadap ban airless yang sudah ada di pasaran menggunakan aplikasi Solidworks 2020. Dari model tersebut dilakukan simulasi dinamis menggunakan aplikasi ANSYS Workbench 19.2 dengan menjalankan ban airless pada polisi tidur. Backpropagataion Neural Network (BPNN) menggunakan MATLAB 2022b selanjutnya digunakan untuk menentukan network yang berfungsi menghubungan parameter input (lebar ban, tebal ban dan circular spokes) dengan parameter output (defleksi dan stress) pada ban airless. Network ini di-generate dengan cara mevariasikan jumlah hidden layer, jumlah node, fungsi aktifasi dll sampai didapatkan Mean Squared Error (MSE) terkecil. Network ini selanjutnya digunakan Genetic Algorithm (GA) sebagai objective function untuk mendapatkan minimum stress dan maximum defleksi dengan massa ban minimum. Dari proses optimasi menggunakan metode Backpropagation Neural Network-Genetic Algorithm dihasilkan ban airless dengan parameter ukuran tebal 39,54 mm, lebar 48,38 mm dan diameter lubang 9,33 mm di mana pada ukuran terebut dihasilkan tegangan sebesar 0,99 MPa, defleksi vertikal 18,90 mm dan massa 2,99 kg dengan pembebanan 72 kg pada ban belakang
===============================================================================================================================
The use of pneumatic tires that are common today is very influential on the condition of the road being traversed because they are very prone to damage when hit by sharp objects and over-pressure which has the potential to cause reduced comfort and can even lead to accidents, thus potentially causing accidents. Currently there are tires with airless technology or airless tires, as newcomers of course airless tires offer advantages compared to pneumatic tires where airless tires offer better driving control and comfort, the presence of cavities will allow the vibrations experienced by tires to be absorbed. However, improvements are still needed, especially in terms of material strength compared to pneumatic tires. In this study an analysis will be carried out to determine the amount of deflection and maximum stress that occurs on airless tires compared to pneumatic tires when dynamic impact tests are carried out using ANSYS Workbench software. At the beginning of the research, we first determined the problem boundaries for airless tires, after that three-dimensional modeling of airless tires that are already on the market using the Solidworks 2020 application will be carried out. From this model a dynamic simulation is carried out using the ANSYS Workbench 2021 application by running airless tires on speed bump. Backpropagataion Neural Network (BPNN) using MATLAB 2022b is then used to determine a network that functions to relate input parameters (tire width, tire thickness and circular spokes) with output parameters (deflection and stress) on airless tires. This network is generated by varying the number of hidden layers, number of nodes, activation functions etc. until the smallest Mean Squared Error (MSE) is obtained. This network is then used Genetic Algorithm (GA) as an objective function to obtain minimum stress and maximum deflection with minimum tire mass. From the optimization process using the Backpropagation Neural Network-Genetic Algorithm method, an airless tire is produced with a parameter thickness of 39.54 mm, width of 48.38 mm and a hole diameter of 9.33 mm where the size produces a stress of 0.99 MPa, vertical deflection 18.90 mm and a mass of 2.99 kg with a loading of 72 kg on the rear tires

Item Type:	Thesis (Other)
Uncontrolled Keywords:	ANSYS Workbench, Backpropagataion Neural Network (BPNN) ,Ban airless, Defleksi, Genetic Algorithm, Airless Tire, ANSYS Workbench, Backpropagataion Neural Network (BPNN), Reflection, Genetic Algorithm
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA347 Finite Element Method
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User:	Yosef Ardiansyah Saputro
Date Deposited:	14 Aug 2023 07:19
Last Modified:	14 Aug 2023 07:19
URI:	http://repository.its.ac.id/id/eprint/102935

Actions (login required)

View Item