An improved state selection technique for power system reliability evaluation

Most existing reliability techniques determine credible system outage states by considering system states up to a certain contingency level. The probability of a high level outage may be larger than those of low level outage states since reliabilities of components are different. The significant impact of these high level outage states may be neglected when considering up to certain level of contingencies. An improved fast sorting algorithm (IFSA) for generating system reliability evaluation is proposed in this paper. In this technique, the sequence of units is decided by not only outage probability but also available capacity of a unit in descending order. In order to accelerate the convergence, i.e. minimizing the number of computations and comparisons, a replaceable neighboring states and minimum neighboring states set are proposed to reduce the selected states. The merit of the algorithm is validated by its application to a five units system and East-China power grid. The results show that the efficiency of reliability evaluation has been improved.