Solar Cells, 25 (1988) 127 - 142 127 A NOVEL METHOD FOR DETERMINING THE OPTIMUM SIZE OF STAND-ALONE PHOTOVOLTAIC SYSTEMS C. SORAS and V. MAKIOS Laboratory of Electromagnetics, Department of ElectricalEngineering, University of Patras, Patras (Greece) (Received July 8, 1988; accepted August 8, 1988) Summary A method is presented to select the optimum tilt angle, photovoltaic array area and battery storage capacity of stand-alone photovoltaic systems. This method uses monthly average meteorological data and easily acquirable system parameters in order to determine possible photovoltaic system sizes, capable of supplying any given monthly average hourly load profile. The optimum system selected is that with the minimum life-cycle cost while ensuring a desired reliability level. In the life-cycle cost computations a battery-life model has been used to determine the number of battery bank replacements. The reliability criterion used is the loss-of-energy probability. The method can be implemented on a personal computer and is applied to an illustrative example, where the optimum system size proposed by this methodology is compared with that of a newly installed system on a Greek island. 1. Introduction The existence of analytical methods for selecting the optimum size of stand-alone photovoltaic (SAPV) systems, is the basic requirement for their proliferation. Depending on the approach used to estimate the system performance, which is the prerequisite in any system sizing process, two solutions related to the system sizing problem can be identified, namely detailed simulations and simplified design methods. The models which perform detailed photovoltaic system simulation, e.g. ref. 1, calculate deterministically the energy flow into the system for each hour during the whole period of the analysis, using historical data of solar radiation and ambient temperature. Although these models provide an adequate solution to the system sizing problem, they feature some significant disadvantages, i.e. large computing time, not easily acquirable system parameter values and the need of long-term hourly meteorological data, which are not available for most locations worldwide. 0379-6787/88/$3.50 © Elsevier Sequoia/Printed in The Netherlands