Abstract

     Modern and future high precision pointing space missions face increasingly high challenges related to the widespread use of large flexible structures. The development of new modeling tools which are able to account for the multidisciplinary nature of this problem becomes extremely relevant in order to meet both structure and control performance criteria. This paper proposes a novel methodology to analytically model large truss structures in a sub-structuring framework. A three dimensional unit cube element has been designed and validated with a Finite Element commercial software. This model is composed by multiple two-dimensional sub-mechanisms assembled using block-diagram models. This constitutes the building block for constructing complex truss structures by repetitions of the element. The accurate vibration description of the system and its minimal representation, as well as the possibility of accounting for parametric uncertainties in its mechanical parameters, make it an appropriate tool to perform robust Structure/Control co-design. In order to demonstrate the strengths of the proposed approach, a structure/control co-design study case is proposed and solved using structured robust -synthesis. The objective is to optimize the pointing performances of an antenna, minimizing the perturbations coming from the Solar Array Driving Mechanisms (SADM) of two solar panels, performing active control by means of multiple Proof Mass Actuators (PMA), and simultaneously reduce the mass of the truss-structure which connects the antenna to the main spacecraft body.

Introduction

     In order to systematically face the challenges associated with the next generation of satellites, the European Space Agency (ESA) and NASA have combined their past experiences to cope with the fine pointing requirements of high accuracy observation and Science missions [1]. This represent a domain which is extremely multi-disciplinary: structural, control and system engineering considerations must coalesce to limit the propagation and amplification of internally generated disturbances through the satellite’s flexible structures. For these reasons, the development of rigorous methodologies and design tools that can handle all these domains is crucial at early stages of design. The works of Preda et al. [2] and Sanfedino et al. [3,4] are example or this approach.

     In the past decades, structural and control co-design has attracted a lot of attention due to its ability of merging these multiple multidisciplinary requirements into a single design flow. Moreover, the increasing use of large structures and appendages for Space applications has rendered flexible modal analysis mandatory for the design of proper spacecraft control laws.

Conclusions

     This paper aimed at introducing new analytical tools to model large complex truss structures for space applications in the TITOP/NINOP framework with the specific objective of developing models for structure/control co-design and robust analysis and control. A series of 2D mechanisms block has been introduced to build a unitary 3D cubic element which serves as a building block for truss structures in a sub-structuring approach. The analysis displayed the potentialities of the approach, as large structures composed by a high number of beam elements can be easily assembled by using blocks of decreasing complexity. In addition, a full validation campaign by comparison with Nastran validated the representativeness of these models.

     A case study was then introduced to represent the strengths of the TITOP/NINOP approach in performing robust structure/control co-design in presence of parametric uncertainties. A complex 3D truss structure was built using the previously introduced cube elements and attached to a spacecraft to act as support for an high precision antenna. The objective of the co-design was to reduce the structural mass of the system while satisfying a fine pointing requirement. This study case highlighted the potential of these analytical blocks in performing complex multi disciplinary optimization problems. The direct co-design using structured ?∞ control synthesis allowed for computational cost reduction and brought to a mass saving of almost 76% of the original structural mass, while coping with stringent pointing performances and a large set of uncertainties in the mechanical design parameters. The main conclusion of this work is that it is possible to develop knowledge-based models for complex truss structures with an analytical dependence on the sizing, variable or uncertain parameters. These models can be directly used for robust design, robustness analysis or parametric optimization, thus simplifying model uncertainty quantification and model reduction in the overall control/structure integrated design process.