ABOUT MULTISCALE MODELING WITH THE USE OF HAAR DISCRETE WAVELETS FOR TARGETED DATA EXTRACTION IN PROBLEMS OF STRUCTURAL MECHANICS

The distinctive paper is devoted to multiscale modelling of functions with the use of Haar discrete wavelets. Numerical implementation is realized with the use of “Fortran” programming language (“Intel Visual Fortran” software). Particularly several sample functions are under consideration. Three basic parts of corresponding algorithm include full wave reconstruction (implementing the entire Hilbert space of the wave); extraction of wave components constructed solely by wavelet functions; isolation and visualization of the final wavelet layer. Modeling wave behavior from multiple perspectives enables the extraction of highly valuable information, which can reveal defects in mechanical or structural systems. The Haar discrete wavelet, due to its basis and step-like wavelet functions, is particularly suitable for numerical implementation (programming, computation), featuring straightforward algorithms.