Method for improving resolution of solar radiation quantity forecast data

The invention relates to a method for improving the resolution of solar radiation quantity forecast data, and belongs to the technical field of predicting solar radiation quantity by using a computer algorithm model. According to the method, numerical weather forecast data with the time resolution being T is processed through linear interpolation, a convolutional neural network model and a recurrent neural network model to obtain data with the time resolution being t, and t is smaller than T; the mapping relation between the numerical weather forecast data with the time resolution being T and the site actually-measured irradiation intensity data with the time resolution being t is established, the site actually-measured data are fully utilized, and the change of the site irradiation intensity data along with time can be better captured; and the performance of subsequently predicting the generation power of the station after the model predicts the irradiation intensity of the station is improved. 本发明涉及一种提升太阳辐射