A penalty optimization technique for a class of regulator problems. III (Q1823864)

The frequent encounter is economics, disease epidemics, resource depletion, as well as electro-mechanical systems of phenomena that cannot be readily modelled unless equations involving time delays are admitted has drawn a great deal of research attention to differential-delay systems which have consequently grown at unprecedented rate. In part I [the author and \textit{J. E. Rubio}, J. Optimization Theory Appl. 58, No.1, 39-62 (1988; Zbl 0622.49014)] and part II [the author, \textit{J. E. Rubio} and \textit{I. Orisamolu}, J. Nigerian Math. Assoc. 17, 19-50 (1986)] of this work, we provided and generalized a function space algorithm, knows as the extended conjugate-gradient method (ECGM) algorithm, as a penalty optimization technique for solving a continuous optimal control problem governed by a system of differential-delay equations under the influence of certain inhomogeneous forcing terms; however, numerical results to support our analysis were not presented. To fill the gap, this paper presents the numerical behavior of the ECGM algorithm through the solution of a significant example. There is a noticeable improvement in the accuracy of the solutions obtained via the ECGM algorithm over those obtained via the averaging approximation principles developed by \textit{H. T. Banks} and \textit{J. A. Burns} [SIAM J. Control Optimization 16, 169-208 (1978; Zbl 0379.49025)]. Furthermore, the paper establishes some functional inequalities motivated from practical situations.