Optimization of diode laser cutting parameters for wood veneers using artificial neural networks

Küçük Resim Yok

Tarih

2025

Dergi Başlığı

Dergi ISSN

Cilt Başlığı

Yayıncı

Taylor & Francis Ltd

Erişim Hakkı

info:eu-repo/semantics/closedAccess

Özet

In this study, an artificial neural network (ANN) model was developed to estimate the heat-affected zone (HAZ) width in laser cutting of walnut (Juglans nigra L.) and acer (Acer pseudoplatanus L.) veneers using a diode laser. The model used input parameters such as material type, laser power, cutting speed, and angle. It achieved high prediction accuracy, with MAPE values of 4.44% (training) and 7.01% (test), and RMSE values of 0.60% and 0.70%, respectively. A strong correlation (R > 91%) and high R-2 values (> 94%) between predicted and observed HAZ widths confirmed the model's reliability. Visual analyses supported these results. The HAZ width was influenced by cutting speed and angle, especially when the material type was held constant. Generally, increasing the cutting speed reduced the HAZ width. Similarly, raising the laser power (1.67 W -> 6.66 W) also led to narrowing of the HAZ. Walnut veneer, being dark-colored, extractive-rich, and having large tracheal structure, exhibited wider HAZ due to higher heat absorption and pyrolysis. In conclusion, the ANN model effectively reflects parameter effects on the HAZ in laser cutting and serves as a reliable decision-support tool, offering both time and cost efficiency.

Açıklama

Anahtar Kelimeler

Diode Laser, Artificial Neural Network, Laser Cutting Parameters, Wood Veneers Marquetry

Kaynak

Wood Material Science & Engineering

WoS Q Değeri

Q2

Scopus Q Değeri

Q2

Cilt

Sayı

Künye