Designing Optimal Systems with Stochastic Programming

A complex and demanding task in control theory is the optimization of the behaviour of systems. But what does optimal mean? Roughly spoken it means that one tries to find the "best" solution of an existing problem, for example to maximize the acceleration of a car or to minimize the energy consumption of a plant. The search for a control that fulfills necessary constraints while maximizing (or minimizing) the desired objective defines - according to the fundamental problem of optimization theory in the field of dynamic systems. If there are uncertainties that act upon the system which has to be controlled, it is necessary to make decisions without knowing what the e.ects might be in detail. The contribution of this book shall be to present algorithms that are able to imprint a desired optimal behavior onto a system - despite of occurring uncertainties - while the algorithms itself shall remain practically implementable. This means that the effort (necessary hardware, time needed to obtain the desired solution) has to stay within reasonable limits - which are given by the respective task at hand.