Fast and accurate online short-term reliability assessment

This paper proposes a method to obtain a fast and accurate evaluation of the short-term reliability in the online operational timeframe. This method thus provides operators with information on the near-future reliability of the power system and has the ability to warn them when preventive actions might be required. To achieve a sufficient accuracy, this assessment must take into account multi-state components such as generating units with derated states, multiple transmission lines with common-mode outages and the uncertainty on the load forecast. To achieve the speed required for online use, the method relies on the State Space Partitioning (SSP) technique, which divides the system state space into two subspaces. The system states with a large probability of occurrence comprise the first subspace and are identified using the Fast Sorting Technique (FST). System states that have a low probability of occurrence but which could have severe consequences are sampled using Monte Carlo Simulation (MCS) over the residual subspace. This computational acceleration is particularly important when multi-state components are taken into account. Tests carried out using the IEEE-RTS demonstrate the speed and accuracy of this enhanced SSP method.