Models and Algorithms for Decision Making in Digital Control Systems
Аннотация
Currently, part of the main tasks of digital control information technologies is to recreate a system that thinks logically and intelligently using a computer, and to create devices that make control decision-making like a human under positions of unpredictability and speculative. In this article, the processes of control decision-making in digital control information systems were systematically analyzed and studied. During the analysis of processes control decision-making in digital control information systems, its functions and shortcomings were determined using systematic approach methods. It was the initial stage of a detailed assessment of task positions and the results of statistical models of benchmark for control decision-making were considered. A control decisionmaking rule (master plan) has been developed that allows identifying and choosing the best solutions according to the benchmark and preferences of the control decision-maker. Based on this master plan, an array the most optimal solutions are formed, ranked among the available solutions for the current position. Also, efficiency measures and benchmark are formed, which are used to evaluate the results according to the selected efficiency benchmark and determine the priority level of option solutions corresponding to these results. A model for choosing the optimal solution according to effectiveness measures and benchmark has been developed. In this article, part of the main tasks of digital control information technologies is the creation and calculation of a computer-based information system that thinks logically and intelligently, as well as devices that make human-like control decision-making under positions of unpredictability and speculative. Computational experiments were conducted to solve fuzzy model tasks of control decisionmaking in digital control information systems. Based on the results of computational experiments on task solving, it can be seen that the quality, efficiency and reliability of control decision-making made in digital control information systems are improved.