Abstract: In view of the problems such as the single assessment dimension and the strong subjectivity in determining weights in the security assessment for integrated range of electric power information systems，a multi-dimensional security assessment method integrating particle swarm optimization algorithm，Bayesian network reasoning and Apriori association rule mining is proposed.Firstly，a five-dimensional security assessment index is constructed，and the particle swarm algorithm is used to optimize the weights to reduce subjective bias.Secondly，the risk probability is calculated through Bayesian network reasoning，and the association rules are mined using the Apriori algorithm.Finally，the effectiveness of this assessment method is verified through simulation experiments.The results show that this assessment method converges quickly，the confidence of low application security and low data security leading to high risk reaches 0.938，and the objectivity，interpretability and predictability of the assessment are superior to the expert scoring method and the random forest method.