Algorithms for improving models of optimal control for multi-parametric technological processes based on artificial intelligence

Abstract

This article highlights scientific approaches to solving problems that arise in the development of models for optimal control of multi-parameter technological processes. In particular, at the modeling specification stage, the necessity of developing artificial intelligence algorithms aimed at creating derivative parameters and ensuring their effectiveness for the optimal parametric and structural formulation of the problem is revealed. It is justified that the creation of neural rules is a relatively simple process in improving the formal model of complex systems using combinatorial derivatives of the relationships of significant elements over the full range. Usually, in the modeling of sufficiently complex, multi-parameter, uncertain technological systems, it is impossible to fully cover all the elements of the system that can have a strong influence on its reaction. There are several reasons for this. Nevertheless, the main scientific idea of the research is that it is possible to develop mathematical models that preserve the general effect of all elements and allow for its multi-level assessment, which are tasked with making management decisions.