Catálogo de publicaciones - libros

In this paper we describe two modified versions of the Euler-Picard and Euler-Taylor-Picard discretization methods for nonlinear dynamical control systems. We use an upper bound for the absolute difference of each pair of consecutive Picard iterations, in the first case to control the number of Picard iterations, and in the second case to control the sampling frequency, while keeping the maximum allowed number of Picard iterations fixed. These non-standard discretization methods are used to support the construction of computer animated mimics of nonlinear dynamical control systems.

This article presents a hierarchical control structure aimed at optimizing the electricity production process in solar power plants with distributed collectors. In these systems, a fluid is heated using the energy provided by the solar irradiation until a desired outlet temperature range is achieved, despite of the effect of disturbances (mainly radiation and inlet temperature),using as manipulated variable the fluid flow. The heated fluid is then used for feeding a heat exchanger where steam is produced for electricity generation using a turbine. Nonlinear models are used in the design of the different layers of the control architecture.

This work is devoted to application of explanatory data analysis in the field of solar plant control and monitoring. Data analysis tasks are discussed that allow discovering useful plant knowledge based on historical data obtained from plant control and monitoring systems. Approaches discussed include both OLAP analysis of plant monitoring data (i.e. multidimensional analysis based on a data warehouse), as well as data mining on plant monitoring data. Some of the data analysis tasks discussed have been realized based on monitoring data from distributed solar collector fields at Plataforma Solar de Almeria in Spain.