ANALISIS PENGARUH KONDISI PEMOTONGAN TERHADAP KEKASARAN PERMUKAAN Ti6Al4V PADA PROSES FRAIS DENGAN METODE ARTIFICIAL NEURAL NETWORK

RIDLO, M. RIV’AT and Mohruni, Amrifan Saladin (2016) ANALISIS PENGARUH KONDISI PEMOTONGAN TERHADAP KEKASARAN PERMUKAAN Ti6Al4V PADA PROSES FRAIS DENGAN METODE ARTIFICIAL NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_21201_03121005036_0011096407_01_front_ref.pdf] Text
RAMA_21201_03121005036_0011096407_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (819kB)
[thumbnail of RAMA_21201_03121005036_0011096407_02.pdf] Text
RAMA_21201_03121005036_0011096407_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (525kB) | Request a copy
[thumbnail of RAMA_21201_03121005036_0011096407_03.pdf] Text
RAMA_21201_03121005036_0011096407_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (584kB) | Request a copy
[thumbnail of RAMA_21201_03121005036_0011096407_04.pdf] Text
RAMA_21201_03121005036_0011096407_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (950kB) | Request a copy
[thumbnail of RAMA_21201_03121005036_0011096407_05.pdf] Text
RAMA_21201_03121005036_0011096407_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (91kB) | Request a copy
[thumbnail of RAMA_21201_03121005036_0011096407_06_ref.pdf] Text
RAMA_21201_03121005036_0011096407_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (161kB) | Request a copy
[thumbnail of RAMA_21201_03121005036_0011096407_07_lamp.pdf] Text
RAMA_21201_03121005036_0011096407_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (2MB) | Request a copy

Abstract

Milling process is the process of cutting the workpiece using a cutting tool with many cutting tooth rotating pivot either clockwise or anti clockwise. Cutting or slicing the workpiece can occur if the movement of the chisel cuts imposed on the workpiece that has gripped there will be friction / collision that will result in cuts in the workpiece, this can happen because the constituent material hardness over the chisel has a hardness of the workpiece. In machining surface roughness quality is very important because surface roughness is defining factor to know good or not product resulting form machining process. In this study used a endmill tool with carbide material and titanium alloy Ti6Al4V as workpieces using CNC milling. Cutting process parameters such as cutting speed (Vc), feed rate (fz), depth of cut radial (ar), and depth of cut axial (aa), to know the optimization of surface roughness values with artificial neural network method. Artificial neural networks (ANN) is a computational method that mimics a biological neural network system in the human brain. Based on the results testing of surface roughness affects the variation of cutting conditions. Ti6Al4V surface roughness maximum value obtained was 1.76 μm in the can on testing to 20. When the value of Vc = 100 m / min, fz = 0.158 mm/tooth, ar = 0.32 mm, and 7.07 mm aa = 10 mm , Meanwhile, the minimum value obtained Ti6Al4V surface roughness was 0.22 μm is on testing 10. With test cutting conditions, Vc = 125 m / min, feed rate (fz) = 0.04 mm / tooth, Depth of cut radial (ar) = 0 , 25 mm, Depth of cut axial (aa) = 10 mm. It can be concluded that the bigest value of the cutting speed generated the bigest value of Ra. Similarly, the feed rate, the bigest of feed rate valued generated the bigest of surface roughness Ra value so the rough surface is generated. Then the test result data were analyzed using ANN method back propagation with using neura network architecture 4-1-1, 4-2-1, 4-3-1, 4-4-1, 4-5-1, 4-6-1, 4-7- 1, and 4-8-1 model. To knowing what neural network architecture have the best prediction by looking at the architecture which has the smallest error rate and the correlation surface roughness values between the surface roughness predictive values based analyzed ANN and the surface roughness values from test. Analysis was done by using MATLAB software, be obtained neural network ANN 4-7-1 is the best neural network with MSE (mean square error) value 0,02. Prediction value Ra analysis ANN optimum is 0,29 μm, on the testing with cutting speed condition (Vc= 125 m/min) and feed rate (fz = 0,04 mm/tooth).

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Surface Roughness, Thin walled machining, Ti6Al4V, Artificial Neural Network, MATLAB
Subjects: T Technology > TJ Mechanical engineering and machinery
Divisions: 03-Faculty of Engineering > 21201-Mechanical Engineering (S1)
Depositing User: Prof Amrifan Saladin Mohruni
Date Deposited: 18 Jul 2019 15:57
Last Modified: 18 Jul 2019 15:57
URI: http://repository.unsri.ac.id/id/eprint/415

Actions (login required)

View Item View Item