Robust control design framework for substructure models

Lim, Kyong B.

doi:10.2514/3.21596

Cited by 8 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much of the work done in this field has been focused on the use of sub-block or substructure analysis for control. [21][22][23] This has largely been done to take advantage of the work done by Ikeda et. al.…”

Section: Sub-block Controller Formulationmentioning

confidence: 99%

Discrete Time Finite Element Transfer Matrix Method Development for Modeling and Decentralized Control

Cramer

Swei

Cheung

et al. 2015

Volume 8: 27th Conference on Mechanical Vibration and Noise

View full text Add to dashboard Cite

The current emphasis on increasing aeronautical efficiency is leading the way to a new class of lighter more flexible airplane materials and structures, which unfortunately can result in aeroelastic instabilities.To effectively control the wings deformation and shape, appropriate modeling is necessary. Wings are often modeled as cantilever beams using finite element analysis. The drawback of this approach is that large aeroelastic models cannot be used for embedded controllers. Therefore, to effectively control wings shape, a simple, stable and fast equivalent predictive model that can capture the physical problem and could be used for in-flight control is required.The current paper proposes a Discrete Time Finite Element Transfer Matrix (DT-FETMM) model beam deformation and use it to design a regulator. The advantage of the proposed approach over existing methods is that the proposed controller could be designed to suppress a larger number of vibration modes within the fidelity of the selected time step. We will extend the discrete * Address all correspondence to this author time transfer matrix method to finite element models and present the decentralized models and controllers for structural control. NomenclatureA n = Acceleration integration scaling value for the n th node B n = Acceleration integration constant for the n th node C i j = Galerkin finite element damping matrix sub-block D n = Velocity integration constant for the n th node E n = Velocity integration scaling value for the n th node f n = Control input force at n th node F n = Forward propagation matrix of the right force for the n th node H n = Reverse propagation matrix of the left force for the n th node J n = Reverse propagation matrix for the n th node K i j = Galerkin finite element elastic matrix sub-block M i j = Galerkin finite element mass matrix sub-block P n = Forward propagation matrix for the n th node Q n = Transfer matrix from left boundary condition to n th node T n = Transfer matrix from right boundary condition to n th node v n = Propagation vector 1Copyright c 2015 by ASME x n = n th node position stateṡ x n = n th node velocity states x n = n th node acceleration states τ n = Internal forces at the n th node for either left or right side 7

show abstract

Section: Sub-block Controller Formulationmentioning

confidence: 99%

Discrete Time Finite Element Transfer Matrix Method Development for Modeling and Decentralized Control

Cramer

Swei

Cheung

et al. 2015

Volume 8: 27th Conference on Mechanical Vibration and Noise

View full text Add to dashboard Cite

show abstract

“…As residual mass does not explicitly appear anywhere, it is difficult to introduce these inequality constraints into the LFT framework. The interest of substructure from the uncertainty modeling point of view is adressed in [11,12] for robust control design purposes. In [11], the parametric uncertainty is focused on the interface stiffness.…”

Section: Introductionmentioning

confidence: 99%

“…The interest of substructure from the uncertainty modeling point of view is adressed in [11,12] for robust control design purposes. In [11], the parametric uncertainty is focused on the interface stiffness. The substructure uncertainty is restricted to an unstructured uncertainty to handle high frequency truncated modes.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Modeling and Analysis of Spacecraft With Variable Tilt of Flexible Appendages

Guy¹,

Alazard

Cumer³

et al. 2014

Journal of Dynamic Systems, Measurement, and Control

View full text Add to dashboard Cite

This article describes a general framework to generate linearized models of satellites with large flexible appendages. The obtained model is parameterized according to the tilt of flexible appendages and can be used to validate an attitude control system over a complete revolution of the appendage. Uncertainties on the characteristic parameters of each substructure can be easily considered by the proposed generic and systematic multibody modeling technique, leading to a minimal linear fractional transformation (LFT) model. The uncertainty block has a direct link with the physical parameters avoiding nonphysical parametric configurations. This approach is illustrated to analyze the attitude control system of a spacecraft fitted with a tiltable flexible solar panel. A very simple root locus allows the stability of the closed-loop system to be characterized for a complete revolution of the solar panel.

show abstract

“…As in ref. [9], we are ultimately concerned with the issues of closed loop stability and performance robustness due to inevitable modeling errors, configuration changes, or exogenous disturbances. In particular, we examine the issues of modeling uncertainties for the substructures and their influence on the response of the connected structure.…”

mentioning

confidence: 99%

“…However, when ground testing of the assembled structure is not possible, uncertainty models can only be validated for the components. Consequently, methods are needed to provide for robust control design and analysis of flexible systems with the uncertainty defined at the component levels [9].…”

mentioning

confidence: 99%