Input-to-state stability for nonlinear time-varying systems via averaging (Q5945297)
From MaRDI portal
scientific article; zbMATH DE number 1656482
Language | Label | Description | Also known as |
---|---|---|---|
English | Input-to-state stability for nonlinear time-varying systems via averaging |
scientific article; zbMATH DE number 1656482 |
Statements
Input-to-state stability for nonlinear time-varying systems via averaging (English)
0 references
13 November 2002
0 references
Averaging has been a powerful tool to draw information about a time-varying equation from a time-independent one, namely, the averaged equation. Averaging has been also used in a control framework. The present paper initiates the examination of the role of averaging in the framework of a time-varying system with external disturbances, with the goal of drawing stability properties of the time-varying system from the averaged one. The averaging itself is carried out in the standard manner; however, two notions of approximation of the original system are identified, yielding, respectively, the inference of distinct stability properties. The two approximation characteristics are, roughly, as follows: The weak average property assumes that for each fixed disturbance parameter the averaged equation approximates, in the averaging sense, the time-varying system. The strong average property assumes that for each time-varying disturbance input the averaged equation (which is now time-varying) approximates, in the averaging sense, the time-varying one. Clearly, the strong average property implies the weak average one since fixed input disturbances can be considered as degenerate time-varying disturbances. The two approximation notions are natural ones as demonstrated in the paper by enlightening examples and characterizations of the two properties. The stability properties sought after are phrased within the framework of input-to-state stability. Under natural conditions it is verified that (roughly) input-to-state practical stability of the time-varying equation can be inferred from the input-to-state stability of the averaged one if the latter is a strong average. An example demonstrates that a weak average would not suffice to draw such a conclusion. This by itself is an interesting observation since it is the weak averaging which is the straightforward generalization of averaging equations. It is pointed out in the paper, providing both theoretical results and an example, that the weak averaging property suffices for inferring the input-to-state stability of the time-varying system in case of cascaded systems, namely, when the disturbance is input to a time-varying equation whose (possibly coupled) state is the input to the averaged one.
0 references
averaging
0 references
time-varying system
0 references
external disturbances
0 references
weak average property
0 references
strong average property
0 references
input-to-state stability
0 references
cascaded systems
0 references