Parametric control to quasi-linear systems based on dynamic compensator and multi-objective optimization

Title:

Parametric control to quasi-linear systems based on dynamic compensator and multi-objective optimization

Creator:

Gu, Da-Ke and Zhang, Da-Wei

Identifier:

https://cdk.lib.cas.cz/client/handle/uuid:a4065867-75f3-4c4f-9ad8-c2703ac7e913
uuid:a4065867-75f3-4c4f-9ad8-c2703ac7e913
doi:10.14736/kyb-2020-3-0516

Subject:

quasi-linear systems, parametric control, dynamic compensator, multi-objective design and optimization, and utilize DOFs in parameter matrices

Type:

model:article and TEXT

Format:

bez média and svazek

Description:

This paper considers a parametric approach for quasi-linear systems by using dynamic compensator and multi-objective optimization. Based on the solutions of generalized Sylvester equations, we establish the more general parametric forms of dynamic compensator and the left and right closed-loop eigenvector matrices, and give two groups of arbitrary parameters. By using the parametric approach, the closed-loop system is converted into a linear constant one with a desired eigenstructure. Meanwhile, it also proposes a novel method to realize multi-objective design and optimization. Multiple performance objectives, containing overall eigenvalue sensitivity, H2 norm, H∞ norm and low compensation gain, are formulated by arbitrary parameters, then robustness and low compensation gain criteria are expressed by a comprehensive objective function which contains each performance index weighted. By utilizing degrees of freedom (DOFs) in arbitrary parameters, we can optimize the comprehensive objective function such that an optimized dynamic compensator is found to satisfy the robustness and low compensation gain criteria. Finally, an example of attitude control of combined spacecrafts is presented which proves the effectiveness and feasibility of the parametric approach.

Language:

English

Rights:

http://creativecommons.org/publicdomain/mark/1.0/
policy:public

Coverage:

516-542

Source:

Kybernetika | 2020 Volume:56 | Number:3

Harvested from:

CDK

Metadata only:

false