For the detailed impact assessment of the total power output fluctuation of high penetration photovoltaic power generation system in terms of the load-frequency control, this study evaluated the relationship between the standard deviation (STD) including only shorter cycles than 32 minute and the maximum fluctuation width (MFW) calculated with various window width by using the two data-sets of multi-points observed insolation data. The main results are as follows. The R² of regression line of STD - MFW correlation diagram is larger than 0.85 for various seasons, while the slope of regression line slightly varies with seasons. The slope of regression line is almost the same for various area sizes during the same season, although the variation ranges of both STD and MFW reduce with larger window width due to a so-called smoothing effect. The results suggest that if the STD of geographical average insolation can be calculated by using stochastic method, the MFW can be calculated with a linear function of STD because of the good correlation between STD and MFW independently of seasons and area sizes.