抄録
This paper proposes a computational algorithm for the optimal control of coupled nonlinear systems. The dynamical system under consideration is composed of several subsystems with relatively strong interconnections and/or relatively strong nonlinearities. The objective is to minimize a performance index of quadratic type.
The basic idea of the present algorithm is to remove the interconnections and nonlinearities from system variables, and then the overall problem is decomposed into a number of smaller and linear subproblems. At the same time, the performance index is modified by using the interaction variables. Parameters, called weights, are introduced in the modified performance index. The values of these parameters have significant influence on the rate of convergence of the algorithm, and they are major factors determining the total computing time necessary to reach a converged result.
On the other hand, the interaction variables are adjusted directly by a gradient algorithm until the interaction balances are satisfied.
In the paper, a sufficient condition for convergence of the iterative algorithm is discussed in detail, and several features of the present algorithm are illustrated by examining an example.
