Along with the advancement of current smart grid construction

the characteristics of flexibility and interaction of power system have become more obvious. The uncertainties and interaction caused by the access of renewable energy and active load have brought more challenges to power system security assessment. To solve this problem

a novel method of uncertainty security assessment is proposed. Firstly

composite security indices are developed based on the theory of Hyper-box and Hyper-ellipse space. Secondly

according to the indices

the probabilistic load flow solved by Latin hypercube sampling is combined with the security assessment. By means of Latin hypercube sampling

the composite security indices are calculated in the variable operation states of power systems. Finally

the composite security assessment information can be easily obtained by statistical technology and interval numbers ranking method. Simulations show that the algorithm is able to take into account the probability distribution of power system security indices

reflect the security level of power systems suffering uncertainty of renewable energy

as well as interaction and response of the active load

providing an effective index for selection of active load’s benign interaction. This work is supported by National High Technology Research and Development Program of China (863 Program) (No. 2011AA05A105).