Evaluation of the mechanism of the destruction of metals based on approaches of artificial intelligence and fractal analysis
Yu. G. Kabaldin, M. S. Anosov, Д. А. Шатагин
Abstract
Abstract The article examines modern informational approaches to assess the degree of damageability of materials based on their fractographic images. The possibility of using the fractal dimension, wavelet transform and convolutional artificial neural networks for tiling and classifying the share of viscous and brittle destructions on fractures is shown. The results of experimental studies of the impact viscosity of materials with different types of crystal lattices in a wide range of temperatures are presented.
Topics & Concepts
Fractal dimensionFractalBrittlenessMechanism (biology)Artificial neural networkConvolutional neural networkArtificial intelligenceWavelet transformFractal analysisMaterials scienceBiological systemWaveletPattern recognition (psychology)Computer scienceMathematicsPhysicsMetallurgyMathematical analysisQuantum mechanicsBiologyMaterial Properties and Failure MechanismsEconomic and Technological Systems AnalysisMining and Gasification Technologies