A procedure for evaluating the applicability of a control proxy function to optimal co-design

ABSTRACT

Optimisation is a useful tool in the design of many systems. When designing a system which will be subject to control, the full optimisation of the system must take into account both the physical components of the system and the controller. Many different methods have been developed for the optimisation of such systems. One approach is to define one single optimisation objective, which may consist of a combination of system characteristics; another approach iterates between the physical system, or plant, and the controller. Other approaches are sequential, in which the plant is optimised first, followed by the controller. The optimisation of the plant may, in some cases, include some characteristics relevant to the controller. One such sequential method, optimisation using a control proxy function (CPF), is highlighted in this paper. This method allows a designer to perform sequential optimisation of a coupled co-design problem, while producing optimal or near-optimal results. However, it is necessary to first evaluate whether the CPF method is appropriate for a problem, and then to identify an appropriate CPF. This paper presents a decision process which a designer can use to determine whether the CPF approach is appropriate, and if so, criteria to determine the CPF.

KEYWORDS:

• Co-design
• optimisation
• control proxy function

Acknowledgments

Thanks also to Dr Panos Papalambros and Dr A. Galip Ulsoy of the University of Michigan for their advice in the course of this research, and to Professor Rebecca Reck of Kettering University for her comments on the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Diane L. Peters http://orcid.org/0000-0001-8177-8145

Additional information

Funding

This work was supported in part by this grant 0625060.