Method for searching the extremum of multidimensional functions in solving engineering problems of machines for forestry works

This article is devoted to the analysis of the most common optimization methods used in practical engineering problems of finding the extremum of multidimensional functions and the formation on the basis of the identified properties of recommendations for choosing the best on different data sets. In the process of analysis, various implementations of gradient descent methods, pulse methods, adaptive methods and quasi-Newtonian methods were considered, and the advantages and problems of each of the methods in their use were summarized. Developed computer program that implements the use of all considered methods. The computational experiment performed for the three functions showed that the zero -Rosenbrock and zero - Powell methods proved to be the most effective.