Identifying the critical nodes of power systems correctly and quickly is significant to prevent the power systems from a big blackout. First

considering the topological and electrical characteristics

we put forward indexes for evaluating the critical nodes of power systems

i.e. the power concentration degree and electrical betweenness of power nodes and the changes of transmission efficiency

agglomeration degree and spinning tree of power networks. Then

we proposed a TOPSIS（technique for order preference by similarity to an ideal solution） based node importance evaluating method and CRITIC（criteria importance through intercriteria correlation） based objective weighting method. Finally

a regional power system of Guangdong was served for demonstrating the effectiveness of the proposed node importance evaluating method. The results show that the proposed node importance evaluating method can identify the critical nodes well

which can avoid subjectively determining the weight of indexes and the difference among various values of an index for different nodes and the conflicts between any two indexes are taken into consideration

which can correspond to the real operation of power systems.