RT info:eu-repo/semantics/doctoralThesis
T1 Effective Neuro-Evolutionary Schemes for Solar Radiation Estimation Problems
A1 Aybar Ruiz, Adrián
K1 Inteligencia Artificial
K1 Heurística
K1 Recursos Renovables
K1 Energía solar
K1 Telecomunicaciones
K1 Telecommunication
AB This Ph.D. thesis’ goal is focused on the optimization of renewable energy resources development,specifically solar PV energy, using different hybrid computational Machine Learningtechniques.Energy is the engine of our society, allowing us performing almost every action taken byhuman beings in our daily routine, and providing a constant evolution and development in allour fields. Currently, fossil fuels entail the higher percentage of energy sources in our planet.They have several advantages, such as easy and constant production, but, at the same time, theypresent substantial disadvantages, like the extreme pollution associated with these resources,and their contribution to global warming and climate change. This is the reason why thelargest and most powerful economies are working for a energy change towards renewable sourcesfor a sustainable development. In the introduction of this thesis, a large number of studiesare presented, which foresee a penetration by over a 50% of this kind of energies in the nextdecades. Strong investments are being made in this field, looking for technology developmentand, besides, introducing these energies into society as a matter to be taken into account, for it isrelated to economic and social status. However, the development of energy systems mainly basedon renewable energy will surely be slow, since these energies depend on variables which are outof our control, mainly atmospheric and climatic variables, which are intrinsically intermittent.This matter must be taken into account, due to the amount of energy demanded by society atthe present, as well as the tremendous increase that is predicted for this demand in the future,due to new technologies and new ways of daily routines, like electric vehicles or IoT, etc.To obtain a solution for this problem, on one hand, it is fundamental to achieve the capabilityof predicting the quantity of energy obtained in each moment, avoiding increases or decreases ofenergy, being this matter the core of this Ph.D. Thesis, where the optimal feature selection forpredicting the quantity of global solar radiation in a given point is studied. On the other hand,all the information to do the prediction process will be obtained from a numerical weather mesoscalemodel called WRF (Weather Research and Forecasting), a static model based on differentphysic equations which involve different variables like humidity, pressure or percentage of cloudfraction in any point and different heights in the planet. Additionally, dynamic information,like global solar radiation can be obtained from a radiometric measuring point in Toledo, Spain,allowing us to get a database of the solar global radiation in the past few years. The result of9mixing both of these data will be added as inputs in our hybrid systems.In this work, a deep analysis in the state of art for machine learning models is performed, soas to solve the problems previously considered. Different contributions have been proposed:1. One of the pillars of this work is focused on the optimal feature selection in the exploitationof solar PV radiation in a given point. For this purpose, Extreme-Learning Machine (ELM)will be used as regressor element in the system, where the output of the ELM will becalculated from the WRF outcome features added as inputs in the system.2. The second contribution of this thesis is related to parameters selection problems. Morespecifically, the use of EAs such as Grouping Genetic Algorithm (GGA) or Coral ReefOptimization (CRO) hybridized with others ML are used as classifiers and regressors.Regarding this, the GGA or CRO look for several subsets of basic parameters to solve theproblem, and the regressor employed provides the prediction in terms of the selected by theGenetic Algorithm (GA), reducing the computational cost maintaining a good accuracy.Finally, the several of the mentioned algorithms are applied in the same problem alreadydefined, in order to get the global solar radiation prediction in different points, dealing to improveprevious results in other works and obtaining new applications and techniques, as new paths ofresearch in the future.
YR 2021
FD 2021
LK http://hdl.handle.net/10017/51087
UL http://hdl.handle.net/10017/51087
LA eng
DS MINDS@UW
RD 26-abr-2024