ABSTRACT
System identification problems occur in many diverse fields. Although system parameters are usually identified by applying known experimental conditions and observing the system response, in most situations there are a number of variables which can be adjusted, subject to certain constraints, so that the information provided by the experiment is maximized. As indicated by Walter and Pronzato [348], experimental design includes two steps. The first is qualitative and consists in selecting a suitable configuration of the input/output ports so as to make, if possible, all the parameters of interest identifiable. The second step is qualitative and based on the optimization of a suitable criterion with respect to free variables such as input profiles, sampling schedules, etc. Since the estimation accuracy can be considerably enhanced by the use of the settings so found, a comprehensive optimum experimental design methodology has been developed and applied in areas such as statistics, engineering design and quality control, social science, agriculture, process control, and medical clinical trials and pharmaceuticals [274].
