Amperage prediction in mono-wire cutting operation using multiple regression and artificial neural network models

YILMAZKAYA E.

Neural Computing and Applications, cilt.35, sa.18, ss.13343-13358, 2023 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 35 Sayı: 18
  • Basım Tarihi: 2023
  • Doi Numarası: 10.1007/s00521-023-08443-x
  • Dergi Adı: Neural Computing and Applications
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Applied Science & Technology Source, Biotechnology Research Abstracts, Compendex, Computer & Applied Sciences, Index Islamicus, INSPEC, zbMATH
  • Sayfa Sayıları: ss.13343-13358
  • Anahtar Kelimeler: Mono-wire cutting, Amperage, Uniaxial compressive strength, Bohme, Artificial neural network, Multiple linear regression
  • Hacettepe Üniversitesi Adresli: Evet

Özet

Operational parameters such as cutting speed and peripheral speed in diamond wire cutting operation greatly affect the efficiency of the machine. The cutting machine’s amperage draw measures how hard the machine must work to run, and it is an alternative way to understand the cutting performance of rocks. High amperage values in cutting indicate that the machine has difficulties in cutting process. The retreat rates of the quarry type cutting machines and downward rates for stationary diamond wire cutting machines change the cutting rates of natural stone blocks. In addition to operational parameters, rock properties such as strength, abrasivity also affect cutting performance. In this study, variations of amperage values during mono-wire cutting were investigated and the effects of cutting parameters and some rock properties on amperage values were examined. While analyzing the basic relationships between cutting parameters and amperage values, obtained experimental data were grouped depending on rock properties. In the final part of the study, amperage values were predicted using multiple regression and artificial neural network models. Produced models were compared by using R2, RMSE, VAF and MAPE performance indices. This comparison showed that the constructed ANN model is highly acceptable for prediction of amperage.