FINITE-TIME SELF-OPTIMIZING SYSTEMS,
Abstract
A correlation method proposed earlier for the optimization of linear systems by altering parameters to reduce a mean-square error criterion is extended to the cases of transient operation, non-quadratic criteria and time varying systems. Stability analysis indicates that the open-loop self-optimizing system is stable regardless of the speed with which the parameters are altered. When parameters of a closed-loop system are permitted to vary rapidly with time, it is found that the problem becomes analogous to that of optimal terminal control, and its solution is a generalization of the methods of optimal programming. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 13, 1963
- Accession Number
- AD0430094
Entities
People
- K. S. Narendra
- L. E. Mcbride Jr.
Organizations
- Harvard University