Análisis de la influencia de las energías renovables sobre el precio de la electricidad en España mediante técnicas de Inteligencia Artificial

Autor: Azofra Rojo, Diego; 

Tipo de documento: Tesis

Director/es: Jiménez Macías, EmilioMartínez Cámara, Eduardo

Universidad: Universidad de La Rioja

Año: 2013 

Resumen: Spain is at the forefront of renewable energies worldwide, with a great integration level of them in its generation mix. This has put Spain in the situation which the other countries supporting renewable energies will be in a short-medium term. However active policies to support renewable energy have been losing strength in recent times, seeking to reduce the deficit. This fact, coupled with the significant rising experienced by the electricity bill lately, have made ordinary people to belief that renewables technologies are expensive for the system and constitute the main cause for that rising. It is true that renewable generation technologies have not been, up to date, competitive enough to challenge the traditional thermoelectric power plants, which has made them earn bonuses from generation. But renewables have in turn benefits for the system as their ranking preference in the energy auction gets the most expensive technologies out of the Pool, lowering thus the electricity prices. That works in practice in a very uneven way depending on the type of renewable energy. This happens because, on the one hand, the bonus for each technology is different and, on the other hand, their integration level in the mix, and thus its influence on the Pool, is disparate. According to that, this Thesis aims to determine, using an ex-post analysis, the influence, in economic terms, of each renewable technology (except for large hydraulic power which receives no bonus) on the Pool and the electricity bill for the Spanish case in 2012. For doing so, artificial intelligence techniques (algorithm M5P) are applied in order to model the energy Pool cassation process. Once, the model has been created, it is time to simulate the new electricity price in five fictional scenarios that suppress wind, photovoltaic, biomass, small hydraulic, and solar thermal, respectively, out of the generation mix. Taking into account the price reduction produced for renewables as well as the bonus received, it is determined, for each renewable technology, whether it is beneficial or in deficit for the system. To go a step further and extrapolate the influence of renewable energies over the Pool to the tariff or charges to be paid by the final consumer in the year 2012, the low voltage most common tariff, known as TUR, is recalculated for the above-mentioned scenarios. It is concluded that wind, biomass and small hydraulic power are beneficial while the photovoltaic and solar thermal technologies are economically deficient for the system, as well as for the electricity tariff. However, the bonus for the new photovoltaic plants has experienced significant declines over the past five years thanks to the price reduction occurred in photovoltaic panels. This means that the economic influence of the photovoltaic as a whole on the system, is highly penalized by the legislative framework during 2007 and 2008, producing the well-known photovoltaic boom. Taking into account the information shown above, it is necessary to undertake a sensitivity analysis that, by eliminating the influence of the bonus of the previous years, determine whether the bonus received by the generation of 1 MWh with a new photovoltaic plant is lower than the savings generated for itself in the Pool. The results are very encouraging for the future of photovoltaic energy, always branded as ¿expensive¿ for the grid; and it can be concluded that photovoltaic facilities type II (ground facilities), under the Royal Decree 1578/2008, since the second call of 2011, have been profitable for the system in the last quarter of 2012, being, also, quite close to profitable in the year 2012 as a whole. Regarding wind power, the incipient energy storage technologies at a large scale are not sufficiently developed; therefore, wind power generation depends entirely on wind speed at any given time. That makes wind power quite unpredictable due to wind variability, and it is really important to the grid due to its significant penetration level thereof. Based on these facts, this study has attempted to make an analysis of the influence of wind power on the power grid based on its level of penetration in it. For this purpose, 111 scenarios have been analyzed, ranging from zero penetration in the grid to a 110% of the real wind power generation of the year 2012. It is concluded that the generation by wind power would have been beneficial to the system whenever the wind power generation level had been equal or higher than 83% of the real case, being 55% the worst scenario, and 110% the optimal.