RATIONAL CONTROL OF OBJECTS WITH UNCERTAIN DYNAMICS

Abstract

The reasons causing uncertainty in the dynamics of objects are considered on the example of aerospace engineering. Objects of automatic control, consisting of control objects, servo drives, and sensors, are subjects to the influence of indefinite external and internal disturbing influences, which destabilize their performance. Classical methods of automatic control of such objects do not allow obtaining the desired quality indicators. For the rational control of such objects with uncertain dynamics, a new approach has been proposed, which is based on a deep diagnosis of the causes of destabilization and flexible restoration of working capacity. The main provisions of the method of rational control are described. A new class of mathematical models called diagnostic is presented. In-depth diagnostic procedures are formed using formalized and weakly formalized models of the processes of regular and abnormal operation of the control system. For the formation of a priori and a posteriori knowledge of the causes of malfunctioning, structures of developing dichotomous trees are used, representing a kind of product knowledge bases with a design in nodes “if ... then ...”, with logical rules for obtaining a diagnosis of the causes of destabilizing effects. A method for solving diagnostic problems for objects with uncertain dynamics is described. To restore the performance of objects with uncertain dynamics, the use of redundant means has been proposed: signal and parametric adjustment, reconfiguration of algorithms and hardware. The process of parrying destabilization, which was diagnosed, is formed using the second method A. M. Lyapunov. This method ensures the stability of the system and the specified quality control. Rational adaptation to destabilizing effects is based on the use of intelligent procedures for in-depth diagnosis of the causes of a malfunction of the control system and a flexible restoration of working capacity in real time. Such procedures allow for productive training in order to preserve the efficiency of the control system throughout the entire life cycle. The results of applying the method of rational control to a number of model samples of aircraft are given.

Keywords

• rational control
• objects with uncertain dynamics
• the event uncertainty destabilizing impact efficiency
• diagnostic model
• deep diagnostics
• flexible recovery
• dichotomous tree
• lyapunov function