APPLICATION OF LEVENBERG-MARQUARDT’s ALGORITHM IN TRIGONOMETRIC IDENTIFICATION PROBLEMS OF MULTI-DIMENSIONAL NONLINEAR OBJECTS
Abstract
The article deals with the usage of one of the most popular algorithms of training the artificial neural networks of direct distribution - Levenberg-Markhvarda’s algorithm to optimize the multidimensional Fourier’s series coefficients according to the trigonometric system of paired functions in order to improve the trigonometric identification accuracy of the multidimensional nonlinear objects in control systems.