A set of three methods are proposed for forecasting amounts of global solar radiation. The forecasts are used to predict the output power of photovoltaic systems installed in power systems and control the output of other generators to meet the electricity demand. The methods forecast two-day-ahead, one-day-ahead and three-hour-ahead global solar radiation, respectively, based on weather information using descriptive statistics, factor analysis and binary trees. The main focus is on estimation of the anticipated variations in the forecasts, finding relevant input variables or signal components, and dealing with the nonlinearity of the relationship between the global solar radiation and the input variables. The methods have been tested on the meteorological and global solar radiation data obtained at an observation site, and the results show that they are promising to be used to keep the balance between demand and supply of electric power in the near future power systems with a large number of photovoltaic systems installed.