Using artificial neural networks in estimating wood resistance

Abstract

The purpose of this research was to evaluate the potential of Artificial Neural Networks in estimating the properties of wood resistance. In order to do so, a hybrid of eucalyptus (Eucalyptus urograndis) planted in the Northern Region of the State of Mato Grosso was selected and ten trees were collected. Then, four samples of each tree were removed, totaling 40 samples, which were later subjected to non-destructive testing of apparent density, ultrasonic wave propagation velocity, dynamic modulus of elasticity obtained by ultrasound, and Janka hardness. These properties were used as estimators of resistance and compressive strength parallel to fibers, and hardness. Multilayer Perceptron networks were also employed, training 100 of them for each of the evaluated parameters. The obtained results indicated that the use of Artificial Neural Networks is an efficient tool for predicting wood resistance.

Artificial intelligence tools and vegetation indices combined to estimate aboveground biomass in tropical forests

“…21,26,27,75,76 In addition, the ANNs and remotely sensed data combined can provide higher modeling precision in forest sites than the regression models, as they are able to assimilate a high complexity and variety of vegetation, environment, and climatic aspects. 21,24 Although developing prediction models using ANNs is a high-complexity task that requires skilled labor and high computational capacity for training them, the ANN has achieved higher estimate precisions when modeling complex relationships among different variables. 22,75,77 More specifically, in this study, it was required approximately 50 working hours of one highly skilled person to train, validate, and apply the ANN model using our datasets and a classical personal computer.…”

“…The adjustment and training of the artificial neural networks (ANNs) to estimate the AGB were carried out on the Statistica software version, 26,43 aiming to achieve the lowest prediction error rate. 24 The ANNs were trained based on supervised approach by defining the input and output variables and using the multilayer perceptron (MLP) architecture type, as applied by Ref. 23.…”

“…19,20 Similarily, the artificial neural networks (ANNs) have been applied to predict forest production. [21][22][23][24] The ANNs are computing systems interconnected with nodes (types of functions) showing learning capacity to perform a task. 25,26 The ANNs and remotely sensed data combined have been successfully applied to estimate dendrometric variables in tropical forests.…”

Combination of non-destructive tests for assessing decay in existing timber elements

“…Moreover, decay assessment using NDTs was researched in Branco et al (2017) and Sousa et al (2017) where, through the use of damage maps, it was possible to pinpoint regions of the structure and even of segments of the elements which had different levels of decay and their probable sources. To analyse the variation of timber properties along one element, due to different levels of conservation, scale and natural variability of wood, some of the more recent studies tried different prediction models, such as and Sousa et al (2018) with hierarchical models and probabilistic models using Bayesian methods, García-Iruela, et al (2016) and Miguel, et al (2018) with artificial neural networks. Furthermore, some research already showed the promising results using combination of different NDTs, such as Cavalli and Togni (2013) found that using pin penetration depth and stress wave velocity, the prediction of bending modulus is high, and Vega, et al (2012) also found the highest predictability of bending modulus by the combination of vibration wave velocity, density and sample length.…”

Ahşap Kompozit Malzemelerin Mekanik ve Fiziksel Özelliklerine göre Tahmininde Radyal Temelli Fonksiyon Sinir Ağının Kullanımı

“…There are a growing number of papers in the field of wood science employing artificial neural network (ANN), such as predicting physical and mechanical properties in wood and wood composites (Fernandez et al, 2008;Fernandez et al, 2012;Melo and Miguel, 2016;Ilkucar et al, 2018;Miguel et al, 2018), calculating wood thermal conductivity (Avradimis and Iliadis, 2005;Xu et al, 2007), classifying wood defects (Marcano-Cedeño et al, 2009;Shahnorbanun et al, 2010;Qayyum et al, 2016), optimizing of bonding strength of the various wood products (Cook and Chiu, 1997;Tiryaki et al, 2014) and analysing of moisture in wood (Zhang et al, 2006;Esteban et al, 2010;Özşahin, 2012).…”