Optimisation algorithm for sampled-data control systems with delay by L2 -norm estimations

The paper considers the problem of numerical optimisation regarding the quality index of the transient processes in SISO sampled-data systems. Here the mathematical model of the system allows for computational delay time, which is assumed to be nearly constant. As cost function for the system optimisation, an expression for the L2 norm of the output signal is applied by implementing the delay as a constant quantity. The cost function is constructed by a special approach, which does not need the system description in state space. This approach is based on the application of the parametric transfer function concept and the Laplace transformation in continuous time. Finally an example arising from a maritime control task illustrates the method. © 2014, Springer-Verlag Berlin Heidelberg.