Rational selection of a blackstart scheme is of great significance to rapid restoration of power systems. Previous studies on optimal selection of blackstart schemes mostly focus on ideal condition of complete evaluation information. This paper proposes a comprehensive evaluation model based on difference weight and collaborative filtering

which can address the issue of assessing blackstart schemes under the condition of incomplete evaluation information. First

differences between indices are computed based on the raw evaluation data to obtain difference weights. Then the missing data is inputed by the collaborative filtering approach for a complete evaluation matrix. Finally

the score of each scheme is calculated to sort the blackstart schemes. Experiments are carried out based on the data of blackstart path of Tianjin power grid. The results show the effectiveness of the proposed approach. This work is supported by National Natural Science Foundation of China (No. 71201145)

Innovation Program of Shanghai Municipal Education Commission (No. 15ZS064)

and Research Fund of Shanghai University of Electric Power (No. K2014037).