Intelligent State Space Pruning with local search for power system reliability evaluation

A methodology called Intelligent State Space Pruning (ISSP) has recently been developed and applied in order to reduce the computational resources necessary to achieve convergence when using non-sequential Monte Carlo Simulation (MCS). The main application of this algorithm has been the probabilistic evaluation of composite power system reliability. ISSP has been shown to perform differently when implemented using different population based metaheuristic algorithms, though computation resources are typically reduced by more than 50%. This reduction in computation resources is particularly important when considering the smart grid - a system whose complexity will be far beyond that of the present power grid. In order to further this line of research, this paper focuses on four contributions: 1) The presentation of a binary version of the deterministic Central Force Optimization (CFO) optimization algorithm, 2) The role of this new algorithm regarding ISSP, 3) The integration of a local search technique with three flavors of the ISSP algorithm in order to improve performance, and 4) A discussion of the role that ISSP may play in the reliability evaluation of the smart grid.