Research on prediction algorithm of solar activity base on MgII index for FY-3/OMS

Solar spectral irradiance fluctuates periodically. MgII can describe solar activity and convert solar reference spectrum irradiance according to solar activity. In this paper, we first calculate the MgII, based on multiple solar spectra and construct a time-dependent MgII curve after considering the aggregation level of the curve. Then the error transfer formula of the MgII prediction value participating in the spectral time-varying calculation is derived. The requirement of absolute calibration accuracy is 3%, and the acceptable maximum prediction algorithm error need to below 2.24%. In the prediction algorithm, the Fourier first-class fitting is used for the principal components of the MgII curve, and the Fourier multi-class fitting is used for the high-frequency components. The order is determined by the fitting effect of the Fourier series at the end of the curve. Finally, MgII of the next 200 days and 3 years is predicted at 10 time points, and maximum average error between the true values and the predicted values is 1.863%(200 days) and 1.922%(3 years), which is less than the acceptable error of 2.24%. Predicting solar activity based on MgII is helpful for satellites to obtain accurate observation data.