Implementación de un algoritmo para el seguimiento del punto de potencia máxima en un panel solar mediante un control predictivo por modelo.

Abstract

The objective was to implement a model of the maximum power point follower algorithm for a photovoltaic panel, seeking to improve the efficiency in electricity generation, through the use of a predictive control by model, it seeks to generate a referential framework towards the reason of the power generation systems to alternative the energy production and the importance of control systems as an alternative to the freedom of resources, trying to explain with data the benefits based on the growing energy need. A control model has been developed that allows to improve the monitoring of the maximum power point of a solar panel and establish this point as the starting line in understanding the strategies for the optimal and efficient production of electrical energy, for this process, Matlab is used as the simulation engine; To validate the improvement, a comparison is made with a control system widely used commercially (perturb and observe),due to its low computational resource, is popular among photovoltaic system controllers. The variant clouds behavior, the working condition in moments of shade and partial shade, makes it necessary to have a control system that can anticipate events in order to take the necessary actions and adjust the strategy that guarantees the highest point power in all conditions, predictive control, which is the alternative addressed in this document; It is presented with examples and comparisons, within this way you have the ability to choose the most optimal type of control based on its functionality and performance.The results achieved allow us to understand in comparison with one of the commercial control systems, applying a predictive control by model improves the power stability generated by the photovoltaic panel, however the heavy computational load makes it clear that from the commercial point of view this alternative It is not yet entirely feasible, the efficiency of the model is improved by carefully developing mathematical work, based on the present work it is recommend, use a functional model.