Sizing and energy management for photovoltaic pumping

Resumen

Over the last few decades, photovoltaic energy has become an effective source to produce electricity that will be used either in isolated sites or injected into the grid. In isolated areas in particular, photovoltaic installations are already being used for pumping water for agriculture or human purposes, since photovoltaic installations are easy to install and, after installation, the maintenance cost is low. However, the inherent variability of the sources means that the installation has to be carefully sized so as to provide an adequate energy management algorithm. So, this thesis focuses on the sizing and energy management of an autonomous photovoltaic installation used to pump water for irrigation in an isolated site. Typically, this type of installation is widely used in arid and semi-arid regions, such as the Maghreb and the South of Europe, where in addition, there is an important availability of solar radiation. The correct operation of these installations is needed, not only to fulfil the water demand, but also to optimize the use of the photovoltaic energy and to extend the life of the components. These objectives can be ensured by a good sizing of the components and an optimum energy management, which represents the two main contributions of this thesis. In fact, the first part of this thesis deals with the component sizing of the photovoltaic irrigation installation, namely the photovoltaic panels and the battery bank. Hence, an algorithm for the optimum sizing of the installation components has been established, based on the crops water requirements, the site climatic characteristics and the restrictions inherent to the components. For this, some models of the components are selected, which have also been validated experimentally. In addition, some techniques related to the maximum photovoltaic power extraction have been studied. Then, the sizing algorithm has been validated using measured data of the target area (Medjez El Beb, Northern Tunisia). The second part of the thesis deals with the energy management of the photovoltaic irrigation installation. Hence, a fuzzy logic based algorithm has been established, to manage the energy generated by the panels and stored in the battery bank. Fuzzy logic has been used, since it is easy to implement and our study is based on the knowledge of the user. The main idea behind the algorithm is as follows: depending on the photovoltaic power generated, the battery depth of discharge, the water level in the reservoir and the water flux, the connection and disconnection of the components is deduced by using some proposed fuzzy rules. The algorithm efficiency has been firstly evaluated by simulation and validated secondly in a plant installed in the laboratory, with satisfactory results. Hence, this thesis has satisfactorily contributed to the components sizing and energy management of photovoltaic irrigation installations.