Optimizing characteristic properties of visually graded soft- and hardwoods lamellas for the glulam production
A. Kovryga (Technische Universität München)
Peter Stapel (Technische Universität München)
JWG Van De Kuilen (TU Delft - Steel & Composite Structures, Technische Universität München)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
For the production of lamellas for GLT or CLT visually or machine strength graded timber is required. Visual grading allows only for a limited number of classes while reject rates tend to be high. The present paper shows the potential of an evolutionary algorithm NSGA-II optimizing the boundaries of multiple visual grading criteria in a reliable way. The optimization routine is applied to optimize the boundaries of the visual grading criteria given in DIN 4074-1 and DIN 4074-5. Destructive and non-destructive test data of 1515 specimens of Norway spruce (Picea abies) and 704 specimens of European beech (Fagus sylvatia) tested in tension were analysed. The optimization aims at: a) a maximization of the yield; 2) grade timber to the desired strength classes. Using this optimization routine for both beech and spruce higher yield figures compared to the grading according to DIN 4074 can be obtained while the desired characteristic strength are being reached.
help_outline