Efficient identification of transient instability states of uncertain power systems

This paper investigates the use of a game theoretic approach, namely the Monte Carlo Tree Search method, to identify critical scenarios considering transient stability of power systems. The method guides dynamic time domain simulations towards the cases that the system exhibits instability in order to explore efficiently the entire domain of possible operating conditions under uncertainty. Since the method focuses the search within the domain on the cases that are more probable to cause instability, information on stability boundaries and values of parameters critical for transient stability are also provided. Critical lines, penetration level of renewable energy sources and system loading can be defined this way.