摘要

In the design of products, we are often faced with optimization problems, in which design parameters are chosen carefully to optimize the performances. Moreover, there are usually multiple design optimization problems that need to be solved simultaneously because designs are often carried out concurrently in multiple disciplines. These design optimization problem are often coupled with each other. Namely, the decision variables of one design optimization problem might be the parameters of another. This kind of coupling makes the optimization process of concurrent and multidisciplinary design more complex than a single optimization design problem. In this article, we investigate the control of two coupled design optimization processes, for which a proportional-integral (PI) control policy is proposed and the influences of the design variable of one optimization problem on the objective function of the other are taken into consideration. Some numerical experiments in the case of robot designing are done to verify the effectiveness of this control policy. Finally, some issues about the control of multiple coupled optimization processes are put forward for future research.