Sequential design of decentralized dynamic compensators using the optimal projection equations (Q3783883)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4047502
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Sequential design of decentralized dynamic compensators using the optimal projection equations |
scientific article; zbMATH DE number 4047502 |
Statements
Sequential design of decentralized dynamic compensators using the optimal projection equations (English)
0 references
1987
0 references
sequential design
0 references
decentralized dynamic compensators
0 references
optimal projection equations
0 references
0.7835050821304321
0 references
0.7526311278343201
0 references
0.7520807981491089
0 references
0.7446476817131042
0 references
This paper considers the sequential design of decentralized dynamic compensators in an iterative way using optimal projection equations. To explain the approach, consider a plant P to be controlled by two controllers \(C_ 1\) and \(C_ 2\). The procedure starts with designing an optimal \(C_ 1\) as a centralized controller for P while ignoring \(C_ 2\), after which \(C_ 2\) is designed as an optimal centralized controller for the augmented system consisting of P and \(C_ 1\). Then, \(C_ 1\) and \(C_ 2\) are iteratively redesigned until the reduction in the overall cost is smaller than a prespecified percentage. However, unless the controllers are static (that is, constant output feedback is used), their dimensions increase at each iteration. To avoid this difficulty and to make use of filtering abilities of dynamic controllers, in the paper, the dimension of the compensators are fixed at the beginning, and the design is performed by solving a set of optimal projection equations which consist of a pair of Riccati and a pair of Liapunov equations. In this way, the i-th decentralized controller is obtained as a projection of a full-order centralized controller for the combination of the plant and other controllers. It is shown that, if the plant is stabilizable by decentralized dynamic controllers of the given order, then the iterative design scheme using optimal projection equations always converges to an optinmal set of controllers. The procedure is applied to decentralized control of an 8-th order model of a pair of simply supported beams connected by a spring.
0 references