In view of the great importance of automatic risk analysis in smart substations for the safety of increasingly complicated power systems

a risk associated model of smart substation based on Bayesian network (BN) is presented. Firstly

after the main risk factors and their relations are identified through experts’ knowledge

a decision table of risk analysis is built and reduced through the reduction approach of the rough set theory. Then

the risk associated model is built based on BN as well as the reduced decision table

the solution of generating and updating conditional probability distribution (CPD) of model is brought forward in conjunction with experts knowledge and gamma function

while updating with monitored data considered. By taking a 220 kV substation as an example

the probabilities of risk considered with or without monitored data respectively

are predicted using reverse reasoning of BN. As the analysis results under different conditions have pre-warning function

the effectiveness and applicability of the proposed model is verified. This work is supported by National Natural Science Foundation of China (No. 51277023) and Jilin Scientific and Technological Development Program (No. 20130206085SF

No. 20140204071GX).