In this talk, we propose functional additive regression (FAR) method to predict the solar flare index. The predictors are 25 variables obtained from the Space-weather HMI Active Region Patches (SHARP) produced by the SDO/HMI team, which can be found as the hmi.sharp data series at the Joint Science Operations Center (JSOC). Each SHARP parameter is a time series observed from May 2010 to December 2017 in one-hour cadence. The FAR method establishes a functional relationship between the solar flare index and the SHARP parameters, so that when a new SHARP parameter enters the model, we are able to predict the future solar flare index. Numerous simulation studies and real-data analysis are provided to demonstrate the effectiveness of the FAR method.