Odunun Yüzey Pürüzlülüğünün ve Adezyon Direncinin Yapay Sinir Ağları ile Tahmini

Abstract

ÖZMobilya ve dekorasyon elemanlarının üretiminde kullanılan ağaç malzemelerin yüzey pürüzlülüğünün ve adezyon direncinin belirlenmesi, nihai ürünün kalitesinin değerlendirilmesi bakımından çok önemlidir. Bu makalede ilk olarak, odun türü, kesme yönü ve zımpara kağıdı türünün yüzey pürüzlülüğü üzerine etkilerini incelemek için yapay sinir ağı (YSA) ile yüzey pürüzlülüğü tahmin modeli geliştirilmiştir. Daha sonra, vernik türü, odun türü, kesme yönü ve yüzey pürüzlülüğünün adezyon direnci üzerine etkileri YSA ile geliştirilen adezyon direnci tahmin modeliyle araştırılmıştır. En iyi performansa sahip tahmin modelleri istatistiksel ve grafiksel karşılaştırmalar yoluyla belirlenmiştir. YSA modellerinin kabul edilebilir sapmalarla oldukça tatmin edici neticeler elde ettiği görülmüştür. Sonuç olarak bu çalışmanın bulguları, deneysel araştırmalar için zaman, enerji ve maliyeti azaltmak amacıyla mobilya ve dekorasyon endüstrisinde etkili bir şekilde uygulanabilir. ABSTRACTDetermining the surface roughness and adhesion strength of wood materials used in the manufacturing of furniture and decoration elements is very crucial in terms of evaluating the quality of the final product. In this article, firstly, the surface roughness prediction model was developed with the artificial neural network (ANN) to examine the effects of wood species, cutting direction and sandpaper type on surface roughness. Then, the effects of varnish type, wood species, cutting direction and surface roughness on adhesion strength were investigated with the adhesion strength prediction model developed with ANN. The prediction models with the best performance were determined by statistical and graphical comparisons. It has been observed that ANN models yielded very satisfactory results with acceptable deviations. As a result, the findings of this study could be employed effectively into the furniture and decoration industry to reduce time, energy and cost for empirical investigations.

Prediction of Life Quality Index Value Rankings of Countries After the COVID-19 Pandemic by Artificial Neural Networks

“…There are as many neurons in the input layer as there are data coming from the outside. Neurons operating in the input layer pass the information they collect directly to the hidden layer without modifying it [24]. The input units X1, X2, and X3 shown in Figure 1 represent information used to train, test, and validate the network.…”

The use of an artificial neural network for predicting the gloss of thermally densified wood veneers

“…The ANN approach has been widely employed in wood science to model input-output relationships. ANN applications to wood science include analyzing moisture in wood (Avramidis and Wu 2007), predicting fracture toughness (Samarasinghe et al 2007), classifying wood veneer defects (Castellani and Rowlands 2008), wood recognition (Khalid et al 2008), optimization of process parameters in oriented strand board manufacturing (Özşahin 2012(Özşahin , Ozsahin 2013, predicting the bonding strength of wood joints (Bardak et al 2016), determination of optimum power consumption in wood machining (Tiryaki et al 2016), prediction of formaldehyde emission (Akyüz et al 2017), and prediction of surface roughness and adhesion strength of wood (Özşahin and Singer 2019). These studies have shown that the ANN approach produces highly successful results.…”