Optimasi Gaya Potong, Kekasaran Permukaan, Dan Keausan Pahat Pada Proses MQL-Assisted Turning Baja S45C Dengan Metode Back Propagation Neural Network–Genetic Algorithm (BPNN-GA)

Hutajulu, Yabes Bolas (2023) Optimasi Gaya Potong, Kekasaran Permukaan, Dan Keausan Pahat Pada Proses MQL-Assisted Turning Baja S45C Dengan Metode Back Propagation Neural Network–Genetic Algorithm (BPNN-GA). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02111940000095-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (3MB) | Request a copy

Abstract

Pelumasan merupakan salah satu komponen yang sangat penting dalam proses pemesinan. Pelumasan dapat memperpanjang umur pahat, mengurangi gaya potong, dan menurunkan kekasaran permukaan benda kerja. Salah satu teknik pelumasan yang mulai banyak digunakan dalam proses pemesinan adalah Minimum Quantity Lubrication (MQL). MQL merupakan pelumasan yang didasarkan pada konsep pemesinan nyaris kering. MQL banyak diterapkan sebagai teknik pelumasan yang ramah lingkungan dibandingkan dengan pelumasan banjir (flood lubrication). MQL sangat mengurangi jumlah pelumas yang digunakan dalam proses pemesinan dan juga mengatasi masalah lingkungan, ekonomis, dan kinerja proses. Penelitian ini membahas tentang proses turning dengan MQL pada material S45C. Rancangan eksperimen yang dipakai adalah Taguchi L9. Parameter-parameter yang digunakan yaitu kecepatan potong (97, 192, dan 308 m/min), kecepatan makan (0,15; 0,20; dan 0,25 mm/rev), dan kedalaman potong (0,50; 0,75; dan 1,0 mm). Variabel respon yang diteliti adalah gaya potong, kekasaran permukaan, dan keausan pahat. Pengaruh parameter-parameter terhadap variabel respon dianalisis menggunakan Analysis of Variance (ANOVA). Metode optimasi yang digunakan dalam penelitian ini adalah Back Propagation Neural Network - Genetic Algorithm (BPNN-GA). Genetic Algorithm (GA) merupakan metode optimasi berdasarkan prinsip-prinsip genetika dan seleksi alam. Elemen-elemen dasar GA meliputi seleksi, crossover, dan mutasi. Metode seleksi yang digunakan adalah roulette wheel. Kelebihan GA dibandingkan metode optimasi lainnya adalah GA dapat melakukan optimasi masalah dengan masalah yang kompleks dan ruang pencarian yang sangat luas. Hasil penelitian menunjukkan bahwa metode optimasi BPNN-GA dapat memberikan variabel respon yang optimal. Optimasi BPNN-GA mampu memprediksi variabel respon yang dihasilkan dengan rata-rata perbedaan antara hasil prediksi dengan hasil eksperimen sebesar 1,40 %. Pengaturan parameter-parameter proses yang dapat memberikan nilai gaya potong, kekasaran permukaan, dan keausan pahat minimum adalah kecepatan potong 97 m/min, kecepatan makan 0,15 mm/rev, dan kedalaman potong 0,55 mm.
==================================================================================================================================
Lubrication is one of the most important components in the machining process. Lubrication can extend tool life, reduce cutting forces and reduce workpiece surface roughness. One lubrication technique that is starting to be widely used in the machining process is Minimum Quantity Lubrication (MQL). MQL is a lubrication based on the concept of near-dry machining. MQL is widely applied as an environmentally friendly lubrication technique compared to flood lubrication. MQL greatly reduces the amount of lubricant used in the machining process and also addresses environmental, economic, and process performance issues. This study discusses the turning process with MQL on S45C material. The experimental design used is Taguchi L9. The parameters used were cutting speed (97, 192, and 308 m/min), feeding speed (0.15; 0.20; and 0.25 mm/rev), and depth of cut (0.50; 0.75; and 1.0 mm). The response variables studied were cutting force, surface roughness, and tool wear. The effect of the parameters on the response variables was analyzed using Analysis of Variance (ANOVA). The optimization method used in this research is Back Propagation Neural Network - Genetic Algorithm (BPNN-GA). Genetic Algorithm (GA) is an optimization method based on the principles of genetics and natural selection. The basic elements of GA include selection, crossover, and mutation. The selection method used is roulette wheel. The advantage of GA over other optimization methods is that GA can optimize problems with complex problems and a very wide search space. The results showed that the BPNN-GA optimization method can provide optimal response variables. BPNN-GA optimization is able to predict the resulting response variable with an average difference between the predicted results and the experimental results of 1.40%. The process parameter settings that can provide the minimum cutting force, surface roughness, and tool wear values are cutting speed 97 m/min, feeding speed 0.15 mm/rev, and depth of cut 0.55 mm.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Back Propagation Neural Network, Genetic Algorithm, MQL, Turning, S45C
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ1077 Lubrication and lubricants. T Technology > TS Manufactures > TS183 Manufacturing processes. Lean manufacturing.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User:	Yabes Bolas Hutajulu
Date Deposited:	14 Sep 2023 01:29
Last Modified:	14 Sep 2023 01:29
URI:	http://repository.its.ac.id/id/eprint/102842

Actions (login required)

View Item