A New Approach of Optimizing the Security Posture Assessment Index System for Electric Power Information Networks

To address the problem of low accuracy and poor real-time performance due to the subjective and absolute nature of the selected indicators, a comprehensive evaluation indicator optimization model is proposed by considering the use of factor analysis and principal component analysis to process the indicators. Firstly, the original data indicators are used to establish the corresponding sets of factor weights, and through factor rotation and estimation, the sets of weights are analyzed and solved in depth to obtain common factors that can describe the interrelationship between multiple indicators; secondly, the principal component analysis method is used to extract the principal components from the information carried by the original indicators, and ensure that there is no correlation between these principal components as far as possible; finally, the original index system can be represented by the comprehensive index parameters which are more independent. The validity of the proposed method is demonstrated by the analysis of experimental cases.