Two New Applications of Artificial Neural Networks: Estimation of Instantaneous Performance Ratio and of the Energy Produced by PV Generators

Abstract

AbstractThe growth of photovoltaic (PV) for electricity generation is one of the highest in the field of the renewable energies and this tendency is expected to continue in the next years. As an obvious consequence, an increasing number of new PV components and devices, mainly arrays and inverters, are coming into the PV market. The need for PV arrays and inverters to be characterized has then become a more and more important aspect. Due to the variable nature of the operating conditions in PV systems, the complete characterization of these elements is quite a difficult issue.One aspect that can help to achieve this goal is to improve methods for estimating the energy produced by photovoltaic generators. Overall, the annual energy provided by a PV generator is directly proportional to the annual radiation incident on the plane of PV generator and the installed nominal power or peak power. However, there are a number of reasons that cause a decrease in the expected energy and include; mismatch losses, dirt and dust, ohmic losses and many more. In this chapter we present two new studies in the PV field. The first one concerns the application of the Artificial Neural Networks (ANN) for estimating the instantaneous Performance Ratio, which is the fundamental parameter in the characterization of PV systems. The second study aims to compare the results of several methods for estimating the annual energy produced by a PV generator, three classical and one based on artificial neural networks, in different types of systems with different settings and types of modules.KeywordsArtificial Neural NetworkMulti Layer PerceptronAnnual EnergyInternational Energy AgencyTime Series PredictionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.