Probability Analysis of Weather Data for Energy Assessment of Hybrid Solar/Wind Power System G. TINA, S. GAGLIANO D.I.E.E.S University of Catania V. le A. Doria 6, 95125 Catania ITALY Abstract: - In this paper a procedure for the probabilistic treatment of irradiance and wind meteorological data is reported in order to evaluate the energy potential of a given site as well as the generation of electricity from photovoltaic systems (PVSs) and wind systems (WECSs). In particular the aim of the proposed analysis of recorded meteorological data is twofold: first of all to check if the real probability distribution functions (PDF) of both clearness index and wind speed are respectively overlapped with Hollands-Huget distribution and Weibull distribution then to find the parameters of these two distributions. The results of this procedure stands for the input of an algorithm for the optimized design of grid-connected Hybrid Solar Wind Power Systems (HSWPS). The core of this algorithm is a probabilistic model based on the convolution technique, that allows to assess the long-term performance of a hybrid solar–wind power system for both stand-alone and grid-linked applications. In this paper, the applicability of this procedure has been tested for a site, Acireale (Italy), finding the fitting parameters of the probabilistic models. Key-Words: - Solar radiation, wind speed, probability analysis, Hybrid Solar Wind Power System, Solar Energy System, Wind Energy conversion System 1 Introduction Solar and wind energy is non-depletable, site- dependent, non-polluting, and potential sources of alternative energy. Utilization of solar and wind power has become increasingly significant, attractive and cost-effective, since the oil crises of early 1970s [1]. However, common drawback with solar and wind energy is their unpredictable nature. In general, the variations of solar and wind energy do not match with the time distribution of demand. The independent use of both the systems results in considerable over-sizing for system reliability, which in turn makes the design costly [1]. As the advantages of solar and wind energy systems became widely known, system designers have tarted looking for their integration. In this scenario Hybrid Solar Wind Power System (HSWPS) can be considered as a viable option for the energy market. The HSWPS design is mainly dependent on the performance of an individual system. In order to predict performance, individual components should be modeled and then their mix can be evaluated to meet the demand reliably. Various modeling techniques are developed by researchers to model components of HWPS. Performance of individual component is either modeled by deterministic or probabilistic approaches [2]. Many attempts have been tried to explore a relatively simple method for designing hybrid energy systems. An algorithm based on energy concept to optimally size solar photovoltaic (PV) array in a PV/wind hybrid system was reported [3]. Different system developments in hybrid energy system for Thailand were published [4]. A simple numerical algorithm was used for unit sizing and cost analysis of a stand-alone wind, solar PV hybrid system [5]. A linear programming technique was developed for optimal design of a hybrid wind/ solar PV power system for either autonomous or grid-linked applications [6]. In [7] Salerno et al. propose a new method for the optimized design of grid-connected HSWPS. The system’s model of HSWPS is an analytical model using statistical approach and in particular convolving the PDFs of power generated by Solar Energy System (SES) and Wind Energy Conversion System (WECS) [8]. The hypothesis assumed in [8] to calculate the probabilistic model of HSWPS and in particular to calculate the PDFs for HSWPS are the following: 1) the expression used for the PDF of clearness index (k t ) is the one proposed by Hollands and Huget [9], 2) the expression used for the PDF of the wind speed is the Weibull Distribution. 4th IASME/WSEAS International Conference on ENERGY, ENVIRONMENT, ECOSYSTEMS and SUSTAINABLE DEVELOPMENT (EEESD'08) Algarve, Portugal, June 11-13, 2008 ISBN: 978-960-6766-71-8 217 ISSN: 1790-5095