Analisa Pengaruh Variasi Setting Parameter Terhadap Prediksi Surface Roughness Pada 3D Print Dengan Material PLA+ Menggunakan Machine Learning

Pamungkas, Bagas Adi (2024) Analisa Pengaruh Variasi Setting Parameter Terhadap Prediksi Surface Roughness Pada 3D Print Dengan Material PLA+ Menggunakan Machine Learning. Diploma thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 2038201042-Undergraduate_Thesis.pdf] Text
2038201042-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (4MB) | Request a copy

Abstract

Manufaktur aditif khususnya cetak tiga dimensi (3D printing) dengan teknik Fused Deposition Modeling (FDM), telah menjadi inovasi penting dalam industri manufaktur. Namun, FDM sering menghadapi masalah terhadap hasil cetakan seperti warping, stringing, surface roughness dan lain-lain. Penelitian ini bertujuan untuk mengidentifikasi faktor-faktor yang mempengaruhi kekasaran permukaan dan menentukan parameter cetak optimal. Selain itu tujuan dari penelitian ini yaitu pengembangan system machine learning untuk dilakukan prediksi dengan regresi polinomial derajat dua dengan menambahkan data dari penelitian saat ini dan penelitian sebelumnya untuk melihat berapa akurasi setelah dilakukan penambahan data.
Melalui variasi parameter printing speed, layer height, dan extruder temperature, dengan menggunakan metode Taguchi Smaller is Better ditemukan bahwa kombinasi terbaik untuk kekasaran permukaan halus adalah printing speed 40 mm/s, layer height 0.1 mm, dan extruder temperature 220ºC. Analisis ANOVA menunjukkan bahwa layer height memiliki memiliki pengaruh signifikan terhadap kekasaran permukaan. Dengan parameter pencetakan yang tepat, akurasi prediksi meningkat menjadi 82% pada data pelatihan dan 91% pada data pengujian. Hasil ini menegaskan pentingnya optimasi parameter cetak dalam meningkatkan kualitas produk dan efisiensi proses manufaktur.
================================================================================================
Additive manufacturing, especially 3D printing using the Fused Deposition Modeling (FDM) technique, has become an important innovation in the manufacturing industry. However, FDM often encounters problems with mold results such as warping, stringing, surface roughness and others. This research aims to identify the factors that affect surface roughness and determine the optimal molding parameters. In addition, the purpose of this research is to develop a machine learning system for prediction with second-degree polynomial regression by adding data from current and previous research to see how much accuracy after adding data.
Through the variation of printing speed, layer height, and extruder temperature parameters, using the Taguchi Smaller is Better method, it was found that the best combination for smooth surface roughness was printing speed 40 mm/s, layer height 0.1 mm, and extruder temperature 220ºC. ANOVA analysis showed that layer height has a significant effect on surface roughness. With proper printing parameters, the prediction accuracy increased to 82% in training data and 91% in testing data. These results confirm the importance of printing parameter optimization in improving product quality and manufacturing process efficiency.

Item Type: Thesis (Diploma)
Uncontrolled Keywords: 3D Printing, Additive Manufacturing, Surface Roughness, Machine Learning, Regression, 3D Print, Manufaktur Aditif, Kekasaran Permukaan, Machine Learning, Regresi
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA355 Vibration.
T Technology > TS Manufactures
T Technology > TS Manufactures > TS156 Quality Control. QFD. Taguchi methods (Quality control)
Divisions: Faculty of Vocational > Mechanical Industrial Engineering (D4)
Depositing User: Bagas Adi Pamungkas
Date Deposited: 08 Aug 2024 07:12
Last Modified: 08 Aug 2024 07:12
URI: http://repository.its.ac.id/id/eprint/111130

Actions (login required)

View Item View Item