Abstract: The prediction methods of photovoltaic power generation are currently divided into:point value prediction and interval prediction,but point value prediction methods are difficult to adapt to the randomness and volatility of photovoltaic power. So this paper proposes a support vector machine（SVM）short-term photovoltaic power interval prediction model based on ensemble empirical mode decomposition（EEMD）and ant lion optimizer based on chaos optimization（ALOCO）. First,similar daily sample sets with different environmental conditions are screened out by grey correlation degree,and the photovoltaic output sequence is decomposed into different eigenmode functions by EEMD. Then,the error penalty factor C and kernel function parameter γ of SVM are optimized by chaotic ant lion algorithm,The and the quantile regression method is used to predict the photovoltaic output power in a short-term interval. Finally,the validity of the established model is verified by example data.