OPTIMIZATION OF CNC OPERATING PARAMETERS TO MINIMIZE SURFACE ROUGHNESS OF Pinus sylvestris USING INTEGRATED ARTIFICIAL NEURAL NETWORK AND GENETIC ALGORITHM

GÜRGEN A., Cakmak A., YILDIZ S., Malkocoglu A.

MADERAS-CIENCIA Y TECNOLOGIA, cilt.24, ss.1-12, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 24
  • Basım Tarihi: 2022
  • Doi Numarası: 10.4067/s0718-221x2022000100401
  • Dergi Adı: MADERAS-CIENCIA Y TECNOLOGIA
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Fuente Academica Plus, Aerospace Database, CAB Abstracts, Communication Abstracts, Metadex, Veterinary Science Database, Directory of Open Access Journals, DIALNET, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.1-12
  • Anahtar Kelimeler: Artificial neural network, genetic algorithm, modeling, Pinus sylvestris, optimization, surface roughness, EDGE-GLUED PANELS, MACHINING PARAMETERS, WOOD, PREDICTION
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

The surface roughness of wood is affected by the processing conditions and the material structure. So, optimization of operation parameters is very crucial to have minimum surface roughness. In this study, modeling and optimization of surface roughness (R-a) of Scotch pine (Pinus sylvestris) was investigated. Firstly, the samples were cut under different conditions 8 nun, 9 mm and 11 mm depth of cut and 12 nun, 14 mm and 16 mm axial depth of cut) in computer numerical control (CNC) machine, and then surface roughness (R.) values of samples were calculated. Then a prediction model of surface roughness was developed using artificial neural networks (ANN). Optimization process was carried out to reach minimum surface roughness of wood samples by the genetic algorithm (GA) method. MAPE value of the ANN model was found lower than 4,0 %. The optimum CNC operation parameters were 1874.5 rad/s, 3,0 m/min feed rate. 9,7 mm depth of cut and 12 mm for axial depth of cut for minimum surface roughness. As a result of study, surface roughness of Scotch pine wood can be modeled and optimized using integrated ANN and GA methods by saving time and cost.