Article Preview
TopIntroduction
The rapid increase of the population and the consumption of energy in the world as well as the depletion of fossil resources generated a significant rise in the prices of natural resources such as petrol and natural gas. This elevation led to the use of hybrid renewable energy systems (HRES).
In contrast to fossil energy, renewable energy comes from resources which are continually refilled such as sunlight, wind, rain, tides, waves and geothermal heat, and it must be harvested when it is available. Among all the renewable sources of energy, photovoltaic (PV) and wind energies are being widely utilized.
HRESs composed of PV panels, a wind turbine (WT), and one or two energy storage systems (ESS) have a long lifetime and normally low maintenance cost. They also offer many advantages over other generation systems: Low pollution, high efficiency, reusability of exhaust heat, and on-site installation (Atia, Fahmy, Ahmed, & Dorrah, 2012). For these reasons, Advances in WT and PV energy technologies have increased their use in these systems (Mermoud, 2010; El-Shater, Eskander, & El-Hagry 2006). Over the past three decades, several models have been developed in order to generate the solar irradiance data based on stochastic models such as autoregressive (AR), autoregressive moving average (ARMA), autoregressive integrated moving average (ARIMA) and Markov chain (Aguiar & Collares-Pereira, 1992; Mora-Lopez & de Cardona, 1998; Santos, Pinazo, & Canada, 2003; Maafi & Adane, 1989). However, these models, based on the probability estimation, do not always give accurate results as they require the precise definition of problem domains as well as the identification of mathematical functions. This is why most stochastic models were found with relatively big errors and sometimes difficult to be adopted widely. On the other hand, energy output estimation for wind turbines of different power ranges has been the subject of a number of papers (see Jafarian & Ranjbar, 2010). The wind speed pattern of a region was used to predict the energy output of the turbine in most of these studies. However, under many conditions, crisp data are insufficient to represent mathematically real-life decision problems. Indeed, human judgments including preference information are vague or fuzzy in nature and may not be appropriate to represent them by accurate numerical values. Fuzzy sets, linguistic variables and intuitionistic fuzzy (IF) sets are better solutions to model these problems (Shu-Ping & Deng-Feng, 2013). In the same vein, complications inherent in systems that combine cooling, heating, and power make it difficult to attain exact numerical data for quantitative criteria. Fuzzy theory can, therefore, play an essential role in the decision-making process whereby fuzzy evaluation can quantitatively represent the qualitative evaluation from decision makers (Jing, Bai, & Wang, 2012).
This paper presents a study of a hybrid renewable energy system (HRES) combining 16 PV panels and one generic WT and proposes an algorithm based on fuzzy logic and natural parameters to predict the wind and solar powers under any weather conditions without dealing with complex mathematical formulas. The system can provide a maximum of 4.5KW of energy per day. Linguistic control rules derived from expert knowledge were written based on three natural input parameters: The wind speed, the temperature and the solar lighting.
The fuzzy control language was incorporated into the algorithm to implement the fuzzy logic control (FLC) phases. A further method of energy management will be required in future work to control the flow of energy in the HRES between the input sources and the batteries.