МЕТОД СТРУКТУРНОЇ ІДЕНТИФІКАЦІЇ НЕЛІНІЙНИХ ІНТЕРВАЛЬНИХ МОДЕЛЕЙ СТАТИЧНИХ ОБ’ЄКТІВ

Abstract

The process of building mathematical models of static objects and systems, which includes the solution of two problems: structural and parametric identification, is considered. At the same time, the task of identifying the structure of the model is more difficult and primary. The problem of structural identification of interval models of the characteristic of a static object is the problem of multiple solving of problems of parametric identification of this model, and therefore, from a computational point of view, it is NP complex. The procedure for finding the optimal structure of the model is considered as a directed selection of a defined set of structures in such a way as to minimize the number of iterations of the formation of interval systems of nonlinear algebraic equations. The article formulates the problem of structural identification of interval models of static objects, as a problem of repeatedly searching for solutions of interval systems of nonlinear algebraic equations, in the form of optimization problems with a nonlinear objective function and nonlinear constraints. For the first time, a method of structural identification of interval models of the characteristics of static objects based on the analysis of interval data is proposed and substantiated, which, unlike the existing ones, is based on self-organization and self-adaptation procedures of computing procedures by analogy with artificial bee colony (АBC), which gives the ability to implement model structure identification procedures with lower computational complexity and obtain interval models with simpler structures compared to known methods. The proposed method was tested on the example of building an interval model of the characteristics of a small hydroelectric plant for the purpose of research and ensuring the maximum efficiency of the use of hydropower resources, which demonstrated the effectiveness of using computational procedures based on the artificial bee colony. Accordingly, the proposed method makes it possible to obtain simple, from the point of view of complexity, interval models of complex static objects with a given guaranteed accuracy and with a lower computational complexity of identifying these models. Such features of the method ensure the effective development of the mathematical apparatus, which is used both in decision-making processes and in the processes of preparing decisions in intellectualized data-oriented systems.