Signal Adaptation in Complex Dynamic Systems

The problem of applying signal adaptation methods in complex systems with high dynamic order is established. It is noted that the main obstacle to constructing such systems is a small margin of their stability, which does not allow implementing the required level of system adaptation to changes in object parameters. To overcome this obstacle, it is proposed to form the general structure of the system based on the evolutionary principle of dividing its functions between different subsystems. It is demonstrated that in practical problems with a high level of generality, it is sufficient to limit ourselves to subsystems of adaptation to non-stationary parameters of the object; ensuring the stability of the signal adaptation loop; ensuring the monotonic nature of the transient process and the specified control time, and the subsystem of ensuring the required control error in the steady state. The content of the synthesis procedures for the listed subsystems is disclosed. Attention is paid to the issues of choosing the type of the reference model, calculating the feedback coefficient and the correcting link in the adaptation loop, as well as implementing astatism in the system. The problem of ambiguity of constructing the correcting link is considered, and a method for resolving this ambiguity using mathematical programming methods is proposed. An example of constructing a system with signal adaptation is presented, illustrating the performance of the proposed approach to the synthesis of such systems.