Neural network modeling of conditions of destruction of wood plank based on measurements

2016 Joint IMEKO TC1-TC7-TC13 Symposium: Metrology Across the Sciences: Wishful Thinking?

Filkin, V.^1 ; Kaverzneva, T.^1 ; Lazovskaya, T.^1 ; Lukinskiy, E.^1 ; Petrov, A.^1 ; Stolyarov, O.^1 ; Tarkhov, D.^1

Peter the Great St-Petersburg Polytechnic University, 29 Politechnicheskaya Str, Saint-Petersburg

195251, Russia^1

关键词: Anisotropic structure;    Individual characteristics;    Loading force;    Neural network approximation;    Neural network model;    Prediction of mechanical properties;    Safety controls;    Ultimate loads;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/772/1/012041/pdf
DOI  :  10.1088/1742-6596/772/1/012041

来源: IOP

【 摘 要 】

The paper deals with the possibility of predicting the ultimate load breaking timber sample based on the loading force dependence on the deflection before destruction. Prediction of mechanical properties of wood is handicapped by complex anisotropic structures. The anisotropic nature of the material and, in a great measure, the random nature of wood grain local features defining moment of destruction lead to a significant dependence of the required load on the individual characteristics of a particular bar. The ultimate load is sought as a function of the coefficients of the neural network approximation of the loading force dependence on the deflection. For this purpose, a number of experiments on timber sample loading until the destruction is conducted. Modeling of the conditions of material destruction may provide the required safety control in building industry.