Optimization of an off-grid hybrid PV-Wind-Diesel system with different battery technologies using genetic algorithm

The power supply of remote sites and applications at minimal cost and with low emissions is an important issue when discussing future energy concepts. This paper presents the modelling and optimization of a stand-alone hybrid energy system. The system consists of photovoltaic (PV) panels and a wind turbine as renewable power sources, a diesel generator for back-up power and batteries to store excess energy and to improve the system reliability. For storage the technologies of lithium-ion, lead-acid, vanadium redox-flow or a combination thereof have been considered. To be able to use different battery technologies at the same time, a battery management system (BMS) is needed. The presented BMS minimizes operation costs while taking into account different battery operating points and ageing mechanisms. The system is modelled and implemented in Matlab/Simulink. As input, the model uses data of the irradiation, wind speed and air temperature measured in 10 min intervals for 10 years in Aachen - Germany. The load is assumed to be that of a rural UMTS/GSM base station for telecommunication. For a time frame of 20 years, the performance is evaluated and the total costs have been determined. Using a genetic algorithm, component sizes and settings have then been varied and the system re-evaluated to minimize the overall costs. The optimization has been also done for a site in Quneitra - Syria which has very good solar radiation that allows for the comparison between two countries, as the weather data in the two countries differ greatly (different weather data). The optimization results show that using batteries in combination with the renewables is economical and ecological. However, the best solution is to combine redox-flow batteries with the renewables. In addition, a power supply system consisting only of batteries, PV and wind generators may be applicable as well to satisfy the power demand. (C) 2013 Elsevier Ltd. All rights reserved.