Optimal capacity planning for stand-alone photovoltaic generation in Taiwan

The optimal capacity planning of stand-alone photovoltaic (SPV) system can significantly improve the reliability and economical performance of power supply. This paper proposes a procedure based on genetic algorithms to get global optimum capacity of solar array and battery in a SPV system more efficiently. Different degrees of system reliability are investigated statistically to achieve the long-term optimal capacity allocation for a SPV system. System reliability can be described properly by the corresponding sixth order polynomial equations. Variations resulted from the cost of SPV component and load profiles are conducted to demonstrate the impacts of system uncertainty during planning. The solar radiation and temperature data from the Central Weather Bureau of Taiwan at six different locations during years of 2003 to 2009 were used.