Research on Clearness Index Support Vector Regression Method for Solar Radiation Forecasting

Aiming at the problem of intermittent and fluctuating solar radiation, which is difficult to forecast accurately, a short-term forecasting method of solar radiation intensity based on support vector regression is proposed. First, the trend curve of the daily clearness index is calculated by using the K-means algorithm to cluster the curves to obtain the cluster center of the curve under the general weather type. Then, the gradient boosting decision tree is used to judge the forecasting weather type, and the corresponding weather type is used. The meteorological information builds a support vector regression model from time to time to obtain the clearness index and calculate the solar radiation intensity, which avoids the problem of low forecasting accuracy caused by using the same fitting model in different weather conditions. Finally, the proposed forecasting method is verified with measured data. Experimental results show that the proposed method has high forecasting accuracy.