Probabilistic security analysis using SOM and Monte Carlo simulation

This paper proposes a new probabilistic method involving transient stability and voltage stability for power system security assessment by combining Monte Carlo simulation and self organizing map (SOM). This overcomes the problem of large amount of computation time required for Monte Carlo simulation. SOM learns to recognize groups of similar input vectors in such a way that neurons physically near each other in the neuron layer respond to similar input vectors. Data classification by SOM can reduce sampling data, which reduces computation time for the reliability security index when using classified data. A case study of the IEEE RTS is given to demonstrate the efficiency of this approach.