The state-of-the-art development of spatial correlation based wind speed prediction is reviewed. And the concepts of conditional correlation and its corresponding confidence correlation are introduced to improve traditional spatial correlation. Based on big-data thinking

a framework of integrating data-driven with causality-driven wind speed prediction is proposed. In the framework

correlation is mined from historical data for wind speed prediction. Spatial correlation is employed to import data sources for wind speed prediction to overcome the shortage of historical data in part. Furthermore

spatial correlation with long time lag can be used to predict drastic and sudden change in downstream wind speed. Finally

suggestions for future research under the proposed framework can be made with confidence. This work is supported by the State Key Program of National Natural Science Foundation of China(No. 61533010)

NSFC-NRCT(Sino Thai)Cooperation Research Project(No. 51561145011)and State Grid Corporation of China.