Abstract: To make precise and reliable estimation of solar radiation,this paper proposes a short-term solar radiation synergy estimation method,which implements adjacent station data to estimate solar radiation at target station. First,the maximum information coefficient is used to perform feature selection on relevant data from all stations. Then the data after feature selection are utilized as input for estimation using different machine learning models. The errors of the synergistic estimation are finally compared to the error taking only the target station data on real data. Experimental results indicate that the performance of the synergy estimation has a more precision and low error for all target stations,compared without synergy estimation.