PREDIKSI NILAI SURFACE ROUGHNESS DAN HARDNESS PADA MULTI 3D PRINTER DENGAN MATERIAL PLA+ MENGGUNAKAN MACHINE LEARNING

Prakusya, Mohamad Rizky Eka (2024) PREDIKSI NILAI SURFACE ROUGHNESS DAN HARDNESS PADA MULTI 3D PRINTER DENGAN MATERIAL PLA+ MENGGUNAKAN MACHINE LEARNING. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2038201077-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Manufaktur Aditif (AM), yang dikenal sebagai 3D Printing adalah “Proses penggabungan bahan untuk membuat objek dari data model 3D melalui tahap lapis demi lapis”. Salah satu prinsip yang digunakan ialah prinsip Fused Deposition Modelling (FDM). Penggunaan material PLA+ semakin umum karena sifatnya yang ramah lingkungan dan mudah dicetak. Namun, kontrol kualitas dalam pencetakan 3D masih menjadi tantangan, terutama dalam mengontrol nilai surface roughness (kekasaran permukaan) dan hardness (kekerasan) pada benda cetakan.
Penelitian ini bertujuan untuk membuat modelling prediktif menggunakan algoritma machine learning untuk memprediksi nilai surface roughness dan hardness. Data yang diperoleh dari pengujian eksperimental berbagai parameter pencetakan akan digunakan sebagai dataset untuk melatih dan menguji model prediktif.
Hasil dari penelitian ini menunjukkan bahwa terdapat perbedaan nilai surface roughness dan hardness dari variasi 3D Printer. Untuk Mesin 3D Printer Ender 3 V3 KE memiliki rata – rata nilai surface roughness dan hardness 9,83 \mu m dan 72,41 HD. Sedangkan untuk mesin 3D Printer Flashforge Creator Pro KE memiliki rata – rata nilai surface roughness dan hardness 7,02 83 \mu m dan 73,2 HD. Untuk memprediksikan nilai surface roughnes yang memiliki keakuratan paling tinggi yakni dengan persentase error terendah menggunakan metode XGBoost dengan nilai persentase error berdasarkan MAPE 15,16% sedangkan untuk nilai Hardness menggunakan metode Random Forest memiliki nilai persentase error terendah berdasarkan MAPE 3,34% .Implementasi model prediktif yang dikembangkan diterapkan dalam sistem UI sehingga dalam mengoptimalkan parameter pencetakan untuk mencapai kualitas yang diinginkan dalam hal kekasaran permukaan dan kekerasan.
================================================================
Additive Manufacturing (AM), known as 3D Printing, is the "process of joining materials to make objects from 3D model data, usually layer upon layer." One of the principles used is Fused Deposition Modelling (FDM). The use of PLA+ material is becoming increasingly common due to its environmentally friendly properties and ease of printing. However, quality control in 3D printing remains a challenge, particularly in controlling surface roughness and hardness values in printed objects.
This research aims to create predictive modeling using several machine learning algorithms to predict surface roughness and hardness values. Data obtained from experimental testing of various printing parameters will be used as a dataset to train and test the predictive models.
The results of this study show that there are differences in surface roughness and hardness values across different 3D printers. For the Ender 3 V3 KE 3D Printer, the average values for surface roughness and hardness are 9.83 and 72.41, respectively. Meanwhile, for the Flashforge Creator Pro KE 3D Printer, the average values for surface roughness and hardness are 7.02 and 73.2. To predict the surface roughness values with the highest accuracy, using the method with the lowest error percentage, the XGBoost method was used, achieving a Mean Absolute Percentage Error (MAPE) of 15.16%. For predicting hardness values, the Random Forest method had the lowest error percentage, with a MAPE of 3.34%. The implementation of the developed predictive model is applied in a user interface system to optimize printing parameters to achieve the desired quality in terms of surface roughness and hardness.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Addictive Manufacturing (AM), Surface Roughness, Hardness, Machine Learning, Multi FDM 3D Printer Addictive Manufacturing (AM), Surface Roughness, Hardness, Machine Learning, Multi FDM 3D Printer
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA355 Vibration. T Technology > TA Engineering (General). Civil engineering (General) > TA418.42 Hardness properties and tests. Hardness--Testing. T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control
Divisions:	Faculty of Vocational > Mechanical Industrial Engineering (D4)
Depositing User:	MOHAMAD RIZKY EKA PRAKUSYA
Date Deposited:	05 Aug 2024 01:13
Last Modified:	05 Aug 2024 01:13
URI:	http://repository.its.ac.id/id/eprint/111562

Actions (login required)

View Item