ALGORITHMS FOR MULTI-FACTORY POLYNOMIAL MODELING OF TECHNOLOGICAL PROCESSES

Abstract

The article reveals the meaning and essence of the modeling method in the study of complex technological processes based on the problematic aspects of building multifactorial relationships and the principles for their elimination. In the case of non-linearity of multifactorial connections of a complex system, the expediency of approximating such functions by polynomials is substantiated. The authors' ideas about the increasing importance of interpolation methods in practice against the background of the possibility of simplification based on the use of modern mathematical hardware and software for integration into the process of modeling and optimizing technological processes due to the complexity of the approximation issue. In this case, the process of determining the parameters of the polynomial model is based on the principle of similarity to the methodology for determining the interpolation coefficients. If the requirements for the adequacy of the linear control model are not met, a technology is proposed for using the laws of linear modeling using polynomial nonlinear models. An algorithm for polynomial approximation has been developed using the example of an oil and gas separator for control models of a constructive and non-parametric technological system. When the reaction of the technological system depends on the linear expression of input factors, although this model is initially adequate, it is practically analyzed that the results of factor control of changes in the main indicator lead to significant errors.